Mirror/frr
0
0
Fork 1
mirror of https://github.com/FRRouting/frr.git synced 2025-04-30 21:47:15 +02:00
Code Issues Actions 4 Packages Projects Releases Wiki Activity
frr/zebra/rt_netlink.c

3605 lines
93 KiB
C
Raw Normal View History

Initial revision
2002-12-13 21:15:29 +01:00
/* Kernel routing table updates using netlink over GNU/Linux system.
* Copyright (C) 1997, 98, 99 Kunihiro Ishiguro
*
* This file is part of GNU Zebra.
*
* GNU Zebra is free software; you can redistribute it and/or modify it
* under the terms of the GNU General Public License as published by the
* Free Software Foundation; either version 2, or (at your option) any
* later version.
*
* GNU Zebra is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* General Public License for more details.
*
*: make consistent & update GPLv2 file headers The FSF's address changed, and we had a mixture of comment styles for the GPL file header. (The style with * at the beginning won out with 580 to 141 in existing files.) Note: I've intentionally left intact other "variations" of the copyright header, e.g. whether it says "Zebra", "Quagga", "FRR", or nothing. Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-05-13 10:25:29 +02:00
* You should have received a copy of the GNU General Public License along
* with this program; see the file COPYING; if not, write to the Free Software
* Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
Initial revision
2002-12-13 21:15:29 +01:00
*/
#include <zebra.h>
build: zebra: remove *_method Makefile hacks replace with preprocessor checks in source files. Much simpler... Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-26 19:49:15 +02:00
#ifdef HAVE_NETLINK
lib, zebra: unify link layer type and hardware address handling This removes the BSD specific usage of struct sockaddr_dl hardware address. This unifies to use explict hw_addr member for the address, and zebra specific enumeration for the link layer type. Additionally the zapi is updated to never send platform specific structures over the wire, but the ll_type along with hw_addr_len and hw_addr are now sent for all platforms. Based on initial work by Paul Jakma. Signed-off-by: Timo Teräs <timo.teras@iki.fi> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com> # Please enter the commit message for your changes. Lines starting # with '#' will be kept; you may remove them yourself if you want to. # An empty message aborts the commit. # # Author: Timo Teräs <timo.teras@iki.fi> # # rebase in progress; onto 9c2f85d # You are currently editing a commit while rebasing branch 'renato' on '9c2f85d'. # # Changes to be committed: # modified: isisd/isis_circuit.c # modified: lib/if.c # modified: lib/if.h # modified: lib/zclient.c # modified: zebra/interface.c # modified: zebra/interface.h # modified: zebra/kernel_socket.c # modified: zebra/rt_netlink.c # modified: zebra/rtadv.c # modified: zebra/zserv.c # # Untracked files: # "\033\033OA\033OB\033" # 0001-bgpd-fix-build-on-Solaris.patch # ldpd/ # redhat/ldpd.init # redhat/ldpd.service # tags #
2016-01-15 16:36:33 +01:00
#include <net/if_arp.h>
build: include our own copies of some linux kernel headers This is the definitive solution to avoid build issues on old Linux systems, where the system kernel headers might not contain some constants or macros used by FRR (e.g. MPLS_IPTUNNEL_DST, introduced on 2015). This is the same strategy adopted by other projects, like iproute2, libnl, lldpd, strongswan, etc. These header files don't need to be in sync with upstream, they only need to be updated when necessary (e.g. if we want to use a new feature introduced by a recent kernel). Fixes #962 using the solution suggested by David Lamparter. Signed-off-by: Renato Westphal <renato@opensourcerouting.org>
2017-09-21 03:12:56 +02:00
#include <linux/lwtunnel.h>
#include <linux/mpls_iptunnel.h>
#include <linux/neighbour.h>
#include <linux/rtnetlink.h>
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
#include <linux/nexthop.h>
Initial revision
2002-12-13 21:15:29 +01:00
/* Hack for GNU libc version 2. */
#ifndef MSG_TRUNC
#define MSG_TRUNC 0x20
#endif /* MSG_TRUNC */
#include "linklist.h"
#include "if.h"
#include "log.h"
#include "prefix.h"
#include "connected.h"
#include "table.h"
zebra: read multipath routes and hw addr from netlink * zebra/rt_netlink.c: - Pick up the hardware address of an interface when we receive a netlink link change message. Extract code for parsing the link-layer hardware address into a new function so we can reuse it. - netlink_routing_table(): Update to handle multipath routes. - netlink_route_change(): Update to handle multipath routes. Fix problem where the metric was not being read out. * zebra/zebra_rib.[ch]: Extern nexthop_ipv4_ifindex_add() -- it is now called from the netlink code. From: Josh Bailey <joshb@google.com> Signed-off-by: Avneesh Sachdev <avneesh@opensourcerouting.org> Signed-off-by: David Lamparter <equinox@diac24.net>
2012-03-22 09:09:21 +01:00
#include "memory.h"
*: split & distribute memtypes and stop (re|ab)using lib/ MTYPEs This is a rather large mechanical commit that splits up the memory types defined in lib/memtypes.c and distributes them into *_memory.[ch] files in the individual daemons. The zebra change is slightly annoying because there is no nice place to put the #include "zebra_memory.h" statement. bgpd, ospf6d, isisd and some tests were reusing MTYPEs defined in the library for its own use. This is bad practice and would break when the memtype are made static. Acked-by: Vincent JARDIN <vincent.jardin@6wind.com> Acked-by: Donald Sharp <sharpd@cumulusnetworks.com> [CF: rebased for cmaster-next] Signed-off-by: David Lamparter <equinox@opensourcerouting.org> Signed-off-by: Christian Franke <chris@opensourcerouting.org>
2015-05-29 05:48:31 +02:00
#include "zebra_memory.h"
Initial revision
2002-12-13 21:15:29 +01:00
#include "rib.h"
From havanna_moon@gmx.net Fri Jan 17 23:37:49 2003 Date: Sat, 11 Jan 2003 23:26:28 +0100 (CET) From: Yon Uriarte <havanna_moon@gmx.net> To: "the list(tm) Zebra" <zebra@zebra.org> Subject: [zebra 17217] [PATCH] show thread CPU Hi, a little patch from the 'stupid preprocessor tricks' collection to record thread statistics. Usage: "show thread cpu [r][w][t][e][x]" Output Fields: self explaining I hope. Type is one of RWTEX for: Read, Write (fd threads), Timer, Event, Execute. Overhead vs. vanilla zebra: almost nothing. Vanilla CVS zebra already collects thread run times. Caveats: Under linux getrusage has a granularity of 10ms, which is almost useless in this case. Run ./configure, edit config.h and comment out "#define HAVE_RUSAGE", this way it will use getimeofday which has a much better granularity. IMHO this is better, as cooperative threads are effectively running during all that wall time (dont care if CPU utilization was 3% or 99% during the time the thread was running (an effective rusage combined with getimeofday could give that info)). Maybe someone can give tips for other platforms on API granularity. TODO: change some of the calls to thread_add_$KIND to funcname_thread_add_$KIND with a meaningfull funcname, so users will get a better idea of what's going on. F.ex. (AFAIK): ospf_spf_calculate_timer -> "Routes Step 1, areas SPF" ospf_ase_calculate_timer -> "Routes Step 2, externals" Could this be added to the unofficial patch collection? Could someone with BGP keepalive problems run their bgpd with this patch and post the results? TIA, HTH, HAND, regards yon Example output: -------------------------------- ospfd# show thread cpu Runtime(ms) Invoked Avg uSecs Max uSecs Type Thread 14.829 31 478 585 T ospf_ase_calculate_timer 82.132 9838 8 291 EX ospf_nsm_event 0.029 1 29 29 E ospf_default_originate_timer 0.254 9 28 34 T ospf_db_desc_timer 0.026 7 3 11 T ospf_wait_timer 669.015 523 1279 490696 R vty_read 4.415 45 98 173 TE ospf_network_lsa_refresh_timer 15.026 31 484 588 T ospf_spf_calculate_timer 29.478 1593 18 122 E ospf_ls_upd_send_queue_event 0.173 1 173 173 T vty_timeout 4.173 242 17 58 E ospf_ls_ack_send_event 637.767 121223 5 55 T ospf_ls_ack_timer 39.373 244 161 2691 R zclient_read 12.169 98 124 726 EX ospf_ism_event 0.226 2 113 125 R vty_accept 537.776 14256 37 3813 W ospf_write 4.967 41 121 250 T ospf_router_lsa_timer 0.672 1 672 672 E zclient_connect 7.901 1658 4 26 T ospf_ls_req_timer 0.459 2 229 266 E ospf_external_lsa_originate_timer 3.203 60 53 305 T ospf_maxage_lsa_remover 108.341 9772 11 65 T ospf_ls_upd_timer 33.302 525 63 8628 W vty_flush 0.101 1 101 101 T ospf_router_lsa_update_timer 0.016 1 16 16 T ospf_router_id_update_timer 26.970 407 66 176 T ospf_lsa_maxage_walker 381.949 12244 31 69 T ospf_hello_timer 0.114 22 5 14 T ospf_inactivity_timer 34.290 1223 28 310 T ospf_lsa_refresh_walker 470.645 6592 71 665 R ospf_read 3119.791 180693 17 490696 RWTEX TOTAL ospfd# bgpd# sh t c TeX Runtime(ms) Invoked Avg uSecs Max uSecs Type Thread 21.504 476 45 71 T bgp_keepalive_timer 17.784 1157 15 131 T bgp_reuse_timer 29.080 193 150 249 T bgp_scan 23.606 995 23 420 E bgp_event 317.734 28572 11 69 T bgp_routeadv_timer 0.084 1 84 84 E zlookup_connect 0.526 1 526 526 E zclient_connect 1.348 13 103 147 T bgp_start_timer 19.443 142 136 420 T bgp_connect_timer 16.032 772 20 27 T bgp_import 447.141 32322 13 526 TEX TOTAL bgpd# bgpd# show thread cpu rw Runtime(ms) Invoked Avg uSecs Max uSecs Type Thread 155.043 7 22149 150659 R bgp_accept 129.638 180 720 53844 R vty_read 1.734 56 30 129 R zclient_read 0.255 2 127 148 R vty_accept 58.483 983 59 340 R bgp_read 171.495 29190 5 245 W bgp_write 13.884 181 76 2542 W vty_flush 530.532 30599 17 150659 RW TOTAL bgpd# --------------------------------
2003-01-18 00:47:00 +01:00
#include "thread.h"
2003-06-04 Paul Jakma <paul@dishone.st> * Merge of zebra privileges
2003-06-04 15:59:38 +02:00
#include "privs.h"
nexthop-tracking.patch quagga: nexthop-tracking.patch Add next hop tracking support to Quagga. Complete documentation in doc/next-hop-tracking.txt. Signed-off-by: Pradosh Mohapatra <pmohapat@cumulusnetworks.com> Signed-off-by: Daniel Walton <dwalton@cumulusnetworks.com> Signed-off-by: Dinesh Dutt <ddutt@cumulusnetworks.com>
2015-05-20 02:40:34 +02:00
#include "nexthop.h"
zebra: let FIB stand for its respective VRF A new member "vrf_id" is added to "struct rib", reflecting the VRF which it belongs to. A new parameter "vrf_id" is added to the relative functions where need, except those: - which already have the parameter "vrf_id"; or - which have a parameter in type of "struct rib"; or - which have a parameter in type of "struct interface". All incoming routes are set to default VRF. In fact, all routes in FIB are kept in default VRF. And the logic is not changed. Signed-off-by: Feng Lu <lu.feng@6wind.com> Reviewed-by: Alain Ritoux <alain.ritoux@6wind.com> Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Acked-by: Vincent JARDIN <vincent.jardin@6wind.com> [DL: conflicts fixed + compile warning fix] Signed-off-by: David Lamparter <equinox@opensourcerouting.org> Conflicts: zebra/connected.c zebra/kernel_socket.c zebra/rib.h zebra/rt_netlink.c zebra/zebra_rib.c zebra/zserv.c Conflicts: zebra/connected.c zebra/interface.c zebra/kernel_socket.c zebra/rib.h zebra/rt_netlink.c zebra/rtread_getmsg.c zebra/zebra_rib.c zebra/zebra_vty.c zebra/zserv.c
2015-05-22 11:40:02 +02:00
#include "vrf.h"
all: removed all DEFUN command stomps (fix build errors) Signed-off-by: Daniel Walton <dwalton@cumulusnetworks.com>
2016-10-06 22:23:13 +02:00
#include "vty.h"
Quagga: Install label forwarding entries for statically configured LSPs Install the statically configured LSPs into the FIB (kernel). This is done using the new attributes and definitions for MPLS in the kernel - RTA_VIA, RTA_NEWDST and AF_MPLS. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com> Ticket: CM-4804 Reviewed By: CCR-3088 Testing Done: Manual in SE-1
2016-04-16 04:19:37 +02:00
#include "mpls.h"
zebra: VNI and VTEP handling Implement fundamental handling for VNIs and VTEPs: - Handle EVPN enable/disable by client (advertise-all-vni) - Create/update/delete VNIs based on VxLAN interface events and inform client - Handle VTEP add/delete from client and install into kernel - New debug command for VxLAN/EVPN - kernel interface (Linux/netlink only) Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:38:26 +02:00
#include "vxlan.h"
zebra: Handle nhg_hash_entry encaps/more debugging Add code for handling nexthop group hash entry encaps and sending them to the kernel. Add some more debugging information for the encaps and groups in general. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-12 21:46:22 +02:00
#include "printfrr.h"
Initial revision
2002-12-13 21:15:29 +01:00
zebra: clean up zapi organization zserv.c has become something of a dumping ground for everything vaguely related to ZAPI and really needs some love. This change splits out the code fo building and consuming ZAPI messages into a separate source file, leaving the actual session and client lifecycle code in zserv.c. Unfortunately since the #include situation in Zebra has not been paid much attention I was forced to fix the headers in a lot of other source files. This is a net improvement overall though. Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-04-22 22:01:20 +02:00
#include "zebra/zapi_msg.h"
lib, zebra: Rework zebra_ns to be a bit more modular The struct zebra_ns was littered throughout the code base in a half-hazard fashion. Gather up the references and isolate the code a bit better. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com> Reviewed-by: Don Slice <dslice@cumulusnetworks.com> Reviewed-by: Vivek Venkatraman <vivek@cumulusnetworks.com>
2016-04-14 04:40:18 +02:00
#include "zebra/zebra_ns.h"
zebra: Refactor zebra_vrf Move zebra_vrf_XXX functionality into it's own file so that we can isolate a bit the api edges Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com> Reviewed-by: Don Slice <dslice@cumulusnetworks.com> Reviewed-by: Vivek Venkatraman <vivek@cumulusnetworks.com>
2016-04-14 15:20:47 +02:00
#include "zebra/zebra_vrf.h"
[zebra] fix some small compile errors, mark several functions static 2005-11-23 Paul Jakma <paul.jakma@sun.com> * (general) fix some small compile errors, and mark several functions as static. * kernel_socket.c: (ifan_read) should be static. fix missing brackets. (ifm_read,ifam_read,rtm_read_mesg,kernel_read) Make static (ifam_read_mesg) make static. fix incorrect variable name. (rtm_read) make static. Fix call to rib_delete_ipv4 which should be rib_delete_ipv6. (routing_socket,kernel_init) should be static. Void argument should be specified as such, not left incomplete. * rt_netlink.c: rt.h should be included, contains prototypes of exported functions. (kernel_delete_ipv6_old) fix sign of index argument. * rt_socket.c: Exact same as previous. Also, make various functions static. * rtread_getmsg.c: Include zserv.h, which prototypes route_read. Make static. * rtread_sysctl.c: zserv.h and rt.h should be included. fix definition of route_read.
2005-11-23 14:02:08 +01:00
#include "zebra/rt.h"
Initial revision
2002-12-13 21:15:29 +01:00
#include "zebra/redistribute.h"
#include "zebra/interface.h"
#include "zebra/debug.h"
lib, zebra: The Bulk of the conversion over to NS and VRF Convert the rest of zebra over to use a Namespae and VRF. Signed-off-by: Vipin Kumar <vipin@cumulusnetworks.com> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2016-02-01 19:55:42 +01:00
#include "zebra/rtadv.h"
zebra - BFD client de-registration support CM-10680 Issue: When BGP daemon is stopped, all the BGP BFD sessions are not getting deleted from PTM. Root cause: BGP daemon stop causes BFD de-register message to be sent for every peer on which BFD is enabled. But, all the de-register messages from bgpd to zebra are not processed before the socket close. This results in some stale BGP BFD sessions. Fix: Support for client de-register message has been added in PTM/BFD. Changes in Quagga to support BFD client de-registrations: − The BFD clients de-registration is sent directly from zebra daemon when zebra client (bgpd, ospfd and ospf6d) socket close is detected. − Introduced a BFD flag for the zebra clients to prevent BFD de-registration messages from being sent to zebra daemon when the client is shutting down. This reduces the BFD messaging. CM-10540 Issue: Invalid ptm status “fail” instead of “n/a” being displayed for VRF interfaces. Root cause: ptm status is not being initialized to “unknown” status when VRF interface is added or changed. The uninitialized value is ‘0’ which is the value for “fail” Fix: Initialized the ptm status to the correct value. Signed-off-by: Radhika Mahankali <radhika@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com> Reviewed-by: Kanna Rajagopal <kanna@cumulusnetworks.com> Ticket: CM-10680, CM-10540 Reviewed By: CCR-4653 Testing Done: PTM smoke, BGP smoke and ptmd_test.py:TestMultipleAddrsIntfOspfBgp
2016-05-09 05:11:18 +02:00
#include "zebra/zebra_ptm.h"
Quagga: Install label forwarding entries for statically configured LSPs Install the statically configured LSPs into the FIB (kernel). This is done using the new attributes and definitions for MPLS in the kernel - RTA_VIA, RTA_NEWDST and AF_MPLS. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com> Ticket: CM-4804 Reviewed By: CCR-3088 Testing Done: Manual in SE-1
2016-04-16 04:19:37 +02:00
#include "zebra/zebra_mpls.h"
zebra: Refactor netlink interactions Separate core netlink functions and library functions from route-related interactions and interface-related interactions. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com> Reviewed-by: Don Slice <dslice@cumulusnetworks.com> Ticket: CM-13199 Reviewed By: CCR-5254 Testing Done: bgp-min, ospf-min
2016-10-17 21:39:55 +02:00
#include "zebra/kernel_netlink.h"
#include "zebra/rt_netlink.h"
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
#include "zebra/zebra_nhg.h"
lib, pimd, zebra: Allow pimd to ask the kernel about mroute info When we need to lookup the mroute info for a route. Allow pimd to ask the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2016-08-11 02:04:20 +02:00
#include "zebra/zebra_mroute.h"
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
#include "zebra/zebra_vxlan.h"
zebra: fix includes Add and remove error related includes as necessary. Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-08-24 19:14:09 +02:00
#include "zebra/zebra_errors.h"
lib, pimd, zebra: Allow pimd to ask the kernel about mroute info When we need to lookup the mroute info for a route. Allow pimd to ask the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2016-08-11 02:04:20 +02:00
Quagga: Install label forwarding entries for statically configured LSPs Install the statically configured LSPs into the FIB (kernel). This is done using the new attributes and definitions for MPLS in the kernel - RTA_VIA, RTA_NEWDST and AF_MPLS. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com> Ticket: CM-4804 Reviewed By: CCR-3088 Testing Done: Manual in SE-1
2016-04-16 04:19:37 +02:00
#ifndef AF_MPLS
#define AF_MPLS 28
#endif
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
static vlanid_t filter_vlan = 0;
zebra: Zebra nexthop group re-work checkpatch fixes Checkpatch fixes for the zebra nexthop group re-work. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-09-03 22:12:06 +02:00
static bool supports_nh;
zebra: Move the supports_nh bool to a better place Move the supports_nh bool indicating whether the kernel we are using supports nexthop objects into the netlink kernel interface itself. Since only linux and netlink support nexthop object APIs for now this is fine. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-01 20:53:06 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
struct gw_family_t {
*: use C99 standard fixed-width integer types The following types are nonstandard: - u_char - u_short - u_int - u_long - u_int8_t - u_int16_t - u_int32_t Replace them with the C99 standard types: - uint8_t - unsigned short - unsigned int - unsigned long - uint8_t - uint16_t - uint32_t Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-03-27 21:13:34 +02:00
uint16_t filler;
uint16_t family;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
union g_addr gate;
Quagga: Install label forwarding entries for statically configured LSPs Install the statically configured LSPs into the FIB (kernel). This is done using the new attributes and definitions for MPLS in the kernel - RTA_VIA, RTA_NEWDST and AF_MPLS. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com> Ticket: CM-4804 Reviewed By: CCR-3088 Testing Done: Manual in SE-1
2016-04-16 04:19:37 +02:00
};
zebra: Store commonly used values instead of regenerating the ipv4_ll address used for 5549 routes does not need to be figured out every single time that we attempt to install/remove a route of that type. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-26 03:11:24 +02:00
char ipv4_ll_buf[16] = "169.254.0.1";
struct in_addr ipv4_ll;
/*
* The ipv4_ll data structure is used for all 5549
* additions to the kernel. Let's figure out the
* correct value one time instead for every
* install/remove of a 5549 type route
*/
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
void rt_netlink_init(void)
zebra: Store commonly used values instead of regenerating the ipv4_ll address used for 5549 routes does not need to be figured out every single time that we attempt to install/remove a route of that type. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-26 03:11:24 +02:00
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
inet_pton(AF_INET, ipv4_ll_buf, &ipv4_ll);
zebra: Store commonly used values instead of regenerating the ipv4_ll address used for 5549 routes does not need to be figured out every single time that we attempt to install/remove a route of that type. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-26 03:11:24 +02:00
}
zebra: Use dataplane for evpn neighbor changes Move neighbor programming to the dataplane; remove old apis; remove some ifdef'd use of direct netlink code points, using neutral values outside of the netlink- specific files. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-08-23 19:59:10 +02:00
/*
* Mapping from dataplane neighbor flags to netlink flags
*/
static uint8_t neigh_flags_to_netlink(uint8_t dplane_flags)
{
uint8_t flags = 0;
if (dplane_flags & DPLANE_NTF_EXT_LEARNED)
flags |= NTF_EXT_LEARNED;
if (dplane_flags & DPLANE_NTF_ROUTER)
flags |= NTF_ROUTER;
return flags;
}
/*
* Mapping from dataplane neighbor state to netlink state
*/
static uint16_t neigh_state_to_netlink(uint16_t dplane_state)
{
uint16_t state = 0;
if (dplane_state & DPLANE_NUD_REACHABLE)
state |= NUD_REACHABLE;
if (dplane_state & DPLANE_NUD_STALE)
state |= NUD_STALE;
if (dplane_state & DPLANE_NUD_NOARP)
state |= NUD_NOARP;
if (dplane_state & DPLANE_NUD_PROBE)
state |= NUD_PROBE;
return state;
}
Add source of route as protocol string in ip route pushed into kernel Ticket: CM-14313 Reviewed By: Testing Done: bgpmin, ospfmin, bgp_kitchen_sink_test 'ip route show' displays all routes as belonging to protocol zebra. The user has to run an additional command (in vtysh) to get the actual source of a route (bgp/ospf/static etc.). This patch addresses that by pushing the appropriate protocol string into the protocol field of the netlink route update message. Now you can see routes with the correct origin as well as filter on them (ip route show proto ospf). 'ospf' is used for both IPv4 and IPv6 routes, even though the OSPF version is different in both cases. Sample output (old): 9.9.12.13 via 69.254.2.38 dev swp3.2 proto zebra metric 20 9.9.13.3 proto zebra metric 20 nexthop via 69.254.2.30 dev swp1.2 weight 1 nexthop via 69.254.2.34 dev swp2.2 weight 1 nexthop via 69.254.2.38 dev swp3.2 weight 1 Sample output (new): 9.9.12.13 via 69.254.2.38 dev swp3.2 proto bgp metric 20 9.9.13.3 proto bgp metric 20 nexthop via 69.254.2.30 dev swp1.2 weight 1 nexthop via 69.254.2.34 dev swp2.2 weight 1 nexthop via 69.254.2.38 dev swp3.2 weight 1
2017-01-11 23:33:39 +01:00
static inline int is_selfroute(int proto)
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if ((proto == RTPROT_BGP) || (proto == RTPROT_OSPF)
tools, zebra: Use different protocol value for our statics The re-use of RTPROT_STATIC has caused too many collisions where other legitimate route sources are causing us to believe we are the originator of the route. Modify the code so that if another protocol inserts RTPROT_STATIC we will assume it's a Kernel Route. Fixes: #2293 Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-05-25 20:36:12 +02:00
|| (proto == RTPROT_ZSTATIC) || (proto == RTPROT_ZEBRA)
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
|| (proto == RTPROT_ISIS) || (proto == RTPROT_RIPNG)
|| (proto == RTPROT_NHRP) || (proto == RTPROT_EIGRP)
zebra: Allow zebra to delete self originated routes With the change to make zebra pass routes to the kernel with the 'correct' proto name, it caused zebra to not properly recognize them on startup again the next time such that the route would not be deleted. Modify rt_netlink.c to notice that we have a self originated route and to properly mark the type of route it was. Modify rib_table_sweep to mark the nexthops as active so that when we go to delete the self originated routes it would properly delete from the kernel. Fixes: #1061 Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-09-01 16:28:19 +02:00
|| (proto == RTPROT_LDP) || (proto == RTPROT_BABEL)
zebra: Add PBR and SHARP handling We are missing some handling of PBR and SHARP protocols for netlink operations w/ the linux kernel. Additionally add a bread crumb for new developers( or existing ) to know to fixup the rt_netlink.c when we start handling new route types to hand to the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-04-27 20:53:46 +02:00
|| (proto == RTPROT_RIP) || (proto == RTPROT_SHARP)
zebra: add RTPROT_FABRICD route type 197 Add an iproute2 route type for fabricd Signed-off-by: Christian Franke <chris@opensourcerouting.org>
2018-06-11 14:09:23 +02:00
|| (proto == RTPROT_PBR) || (proto == RTPROT_OPENFABRIC)) {
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return 1;
}
return 0;
Add source of route as protocol string in ip route pushed into kernel Ticket: CM-14313 Reviewed By: Testing Done: bgpmin, ospfmin, bgp_kitchen_sink_test 'ip route show' displays all routes as belonging to protocol zebra. The user has to run an additional command (in vtysh) to get the actual source of a route (bgp/ospf/static etc.). This patch addresses that by pushing the appropriate protocol string into the protocol field of the netlink route update message. Now you can see routes with the correct origin as well as filter on them (ip route show proto ospf). 'ospf' is used for both IPv4 and IPv6 routes, even though the OSPF version is different in both cases. Sample output (old): 9.9.12.13 via 69.254.2.38 dev swp3.2 proto zebra metric 20 9.9.13.3 proto zebra metric 20 nexthop via 69.254.2.30 dev swp1.2 weight 1 nexthop via 69.254.2.34 dev swp2.2 weight 1 nexthop via 69.254.2.38 dev swp3.2 weight 1 Sample output (new): 9.9.12.13 via 69.254.2.38 dev swp3.2 proto bgp metric 20 9.9.13.3 proto bgp metric 20 nexthop via 69.254.2.30 dev swp1.2 weight 1 nexthop via 69.254.2.34 dev swp2.2 weight 1 nexthop via 69.254.2.38 dev swp3.2 weight 1
2017-01-11 23:33:39 +01:00
}
zebra: Allow zebra to delete self originated routes With the change to make zebra pass routes to the kernel with the 'correct' proto name, it caused zebra to not properly recognize them on startup again the next time such that the route would not be deleted. Modify rt_netlink.c to notice that we have a self originated route and to properly mark the type of route it was. Modify rib_table_sweep to mark the nexthops as active so that when we go to delete the self originated routes it would properly delete from the kernel. Fixes: #1061 Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-09-01 16:28:19 +02:00
static inline int zebra2proto(int proto)
Add source of route as protocol string in ip route pushed into kernel Ticket: CM-14313 Reviewed By: Testing Done: bgpmin, ospfmin, bgp_kitchen_sink_test 'ip route show' displays all routes as belonging to protocol zebra. The user has to run an additional command (in vtysh) to get the actual source of a route (bgp/ospf/static etc.). This patch addresses that by pushing the appropriate protocol string into the protocol field of the netlink route update message. Now you can see routes with the correct origin as well as filter on them (ip route show proto ospf). 'ospf' is used for both IPv4 and IPv6 routes, even though the OSPF version is different in both cases. Sample output (old): 9.9.12.13 via 69.254.2.38 dev swp3.2 proto zebra metric 20 9.9.13.3 proto zebra metric 20 nexthop via 69.254.2.30 dev swp1.2 weight 1 nexthop via 69.254.2.34 dev swp2.2 weight 1 nexthop via 69.254.2.38 dev swp3.2 weight 1 Sample output (new): 9.9.12.13 via 69.254.2.38 dev swp3.2 proto bgp metric 20 9.9.13.3 proto bgp metric 20 nexthop via 69.254.2.30 dev swp1.2 weight 1 nexthop via 69.254.2.34 dev swp2.2 weight 1 nexthop via 69.254.2.38 dev swp3.2 weight 1
2017-01-11 23:33:39 +01:00
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
switch (proto) {
case ZEBRA_ROUTE_BABEL:
proto = RTPROT_BABEL;
break;
case ZEBRA_ROUTE_BGP:
proto = RTPROT_BGP;
break;
case ZEBRA_ROUTE_OSPF:
case ZEBRA_ROUTE_OSPF6:
proto = RTPROT_OSPF;
break;
case ZEBRA_ROUTE_STATIC:
tools, zebra: Use different protocol value for our statics The re-use of RTPROT_STATIC has caused too many collisions where other legitimate route sources are causing us to believe we are the originator of the route. Modify the code so that if another protocol inserts RTPROT_STATIC we will assume it's a Kernel Route. Fixes: #2293 Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-05-25 20:36:12 +02:00
proto = RTPROT_ZSTATIC;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
break;
case ZEBRA_ROUTE_ISIS:
proto = RTPROT_ISIS;
break;
case ZEBRA_ROUTE_RIP:
proto = RTPROT_RIP;
break;
case ZEBRA_ROUTE_RIPNG:
proto = RTPROT_RIPNG;
break;
case ZEBRA_ROUTE_NHRP:
proto = RTPROT_NHRP;
break;
case ZEBRA_ROUTE_EIGRP:
proto = RTPROT_EIGRP;
break;
case ZEBRA_ROUTE_LDP:
proto = RTPROT_LDP;
break;
sharpd: Add Super Happy Advanced Routing Protocol Add a daemon that will allow us to test the zapi as well as test route install/removal times from the kernel. The current commands are: install route <starting ip address> nexthop <nexthop> (1-1000000) This command starts installing at <starting ip address>/32 (1-100000) routes that it auto-increments by 1 Installation start time is noted in the log and finish time is noted as well. remove routes <starting ip address> (1-1000000) This command removes routes at <starting ip address>/32 and removes (1-100000) routes created by the install route command. This code can be considered experimental and *is not* something that should be run in a production environment. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-11-10 18:55:16 +01:00
case ZEBRA_ROUTE_SHARP:
proto = RTPROT_SHARP;
break;
zebra: Add PBR and SHARP handling We are missing some handling of PBR and SHARP protocols for netlink operations w/ the linux kernel. Additionally add a bread crumb for new developers( or existing ) to know to fixup the rt_netlink.c when we start handling new route types to hand to the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-04-27 20:53:46 +02:00
case ZEBRA_ROUTE_PBR:
proto = RTPROT_PBR;
break;
zebra: add RTPROT_FABRICD route type 197 Add an iproute2 route type for fabricd Signed-off-by: Christian Franke <chris@opensourcerouting.org>
2018-06-11 14:09:23 +02:00
case ZEBRA_ROUTE_OPENFABRIC:
proto = RTPROT_OPENFABRIC;
break;
zebra: When installing route use a good proto When installing a table route into the kernel choose RTPROT_ZEBRA as the installing/controlling protocol. This way we can know we installed it as well as stop the warnings about this special case of `ip import-table XX` Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-06-25 23:47:51 +02:00
case ZEBRA_ROUTE_TABLE:
zebra: Sweep our nexthop objects out on restart On restart, if we failed to remove any nexthop objects due to a kill -9 or such event, sweep them if we aren't using them. Add a proto field to handle this and remove the is_kernel bool. Add a dupicate flag that indicates this nexthop group is only present in our ID hashtable. It is a dupicate nexthop we received from the kernel, therefore we cannot hash on it. Make the idcounter globally accessible so that kernel updates increment it as soon as we receive them, not when we handle them. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-01 20:07:04 +02:00
case ZEBRA_ROUTE_NHG:
zebra: When installing route use a good proto When installing a table route into the kernel choose RTPROT_ZEBRA as the installing/controlling protocol. This way we can know we installed it as well as stop the warnings about this special case of `ip import-table XX` Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-06-25 23:47:51 +02:00
proto = RTPROT_ZEBRA;
break;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
default:
zebra: Add PBR and SHARP handling We are missing some handling of PBR and SHARP protocols for netlink operations w/ the linux kernel. Additionally add a bread crumb for new developers( or existing ) to know to fixup the rt_netlink.c when we start handling new route types to hand to the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-04-27 20:53:46 +02:00
/*
* When a user adds a new protocol this will show up
* to let them know to do something about it. This
* is intentionally a warn because we should see
* this as part of development of a new protocol
*/
zebra: flog_warn conversion Convert Zebra to user error subsystem. Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-08-16 22:10:32 +02:00
zlog_debug(
"%s: Please add this protocol(%d) to proper rt_netlink.c handling",
__PRETTY_FUNCTION__, proto);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
proto = RTPROT_ZEBRA;
break;
}
return proto;
Add source of route as protocol string in ip route pushed into kernel Ticket: CM-14313 Reviewed By: Testing Done: bgpmin, ospfmin, bgp_kitchen_sink_test 'ip route show' displays all routes as belonging to protocol zebra. The user has to run an additional command (in vtysh) to get the actual source of a route (bgp/ospf/static etc.). This patch addresses that by pushing the appropriate protocol string into the protocol field of the netlink route update message. Now you can see routes with the correct origin as well as filter on them (ip route show proto ospf). 'ospf' is used for both IPv4 and IPv6 routes, even though the OSPF version is different in both cases. Sample output (old): 9.9.12.13 via 69.254.2.38 dev swp3.2 proto zebra metric 20 9.9.13.3 proto zebra metric 20 nexthop via 69.254.2.30 dev swp1.2 weight 1 nexthop via 69.254.2.34 dev swp2.2 weight 1 nexthop via 69.254.2.38 dev swp3.2 weight 1 Sample output (new): 9.9.12.13 via 69.254.2.38 dev swp3.2 proto bgp metric 20 9.9.13.3 proto bgp metric 20 nexthop via 69.254.2.30 dev swp1.2 weight 1 nexthop via 69.254.2.34 dev swp2.2 weight 1 nexthop via 69.254.2.38 dev swp3.2 weight 1
2017-01-11 23:33:39 +01:00
}
zebra: Sweep our nexthop objects out on restart On restart, if we failed to remove any nexthop objects due to a kill -9 or such event, sweep them if we aren't using them. Add a proto field to handle this and remove the is_kernel bool. Add a dupicate flag that indicates this nexthop group is only present in our ID hashtable. It is a dupicate nexthop we received from the kernel, therefore we cannot hash on it. Make the idcounter globally accessible so that kernel updates increment it as soon as we receive them, not when we handle them. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-01 20:07:04 +02:00
static inline int proto2zebra(int proto, int family, bool is_nexthop)
zebra: Allow zebra to delete self originated routes With the change to make zebra pass routes to the kernel with the 'correct' proto name, it caused zebra to not properly recognize them on startup again the next time such that the route would not be deleted. Modify rt_netlink.c to notice that we have a self originated route and to properly mark the type of route it was. Modify rib_table_sweep to mark the nexthops as active so that when we go to delete the self originated routes it would properly delete from the kernel. Fixes: #1061 Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-09-01 16:28:19 +02:00
{
switch (proto) {
case RTPROT_BABEL:
proto = ZEBRA_ROUTE_BABEL;
break;
case RTPROT_BGP:
proto = ZEBRA_ROUTE_BGP;
break;
case RTPROT_OSPF:
*: conform with COMMUNITY.md formatting rules, via 'make indent' Signed-off-by: Lou Berger <lberger@labn.net>
2018-03-06 20:02:52 +01:00
proto = (family == AFI_IP) ? ZEBRA_ROUTE_OSPF
: ZEBRA_ROUTE_OSPF6;
zebra: Allow zebra to delete self originated routes With the change to make zebra pass routes to the kernel with the 'correct' proto name, it caused zebra to not properly recognize them on startup again the next time such that the route would not be deleted. Modify rt_netlink.c to notice that we have a self originated route and to properly mark the type of route it was. Modify rib_table_sweep to mark the nexthops as active so that when we go to delete the self originated routes it would properly delete from the kernel. Fixes: #1061 Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-09-01 16:28:19 +02:00
break;
case RTPROT_ISIS:
proto = ZEBRA_ROUTE_ISIS;
break;
case RTPROT_RIP:
proto = ZEBRA_ROUTE_RIP;
break;
case RTPROT_RIPNG:
proto = ZEBRA_ROUTE_RIPNG;
break;
case RTPROT_NHRP:
proto = ZEBRA_ROUTE_NHRP;
break;
case RTPROT_EIGRP:
proto = ZEBRA_ROUTE_EIGRP;
break;
case RTPROT_LDP:
proto = ZEBRA_ROUTE_LDP;
break;
case RTPROT_STATIC:
tools, zebra: Use different protocol value for our statics The re-use of RTPROT_STATIC has caused too many collisions where other legitimate route sources are causing us to believe we are the originator of the route. Modify the code so that if another protocol inserts RTPROT_STATIC we will assume it's a Kernel Route. Fixes: #2293 Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-05-25 20:36:12 +02:00
case RTPROT_ZSTATIC:
zebra: Allow zebra to delete self originated routes With the change to make zebra pass routes to the kernel with the 'correct' proto name, it caused zebra to not properly recognize them on startup again the next time such that the route would not be deleted. Modify rt_netlink.c to notice that we have a self originated route and to properly mark the type of route it was. Modify rib_table_sweep to mark the nexthops as active so that when we go to delete the self originated routes it would properly delete from the kernel. Fixes: #1061 Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-09-01 16:28:19 +02:00
proto = ZEBRA_ROUTE_STATIC;
break;
zebra: Add PBR and SHARP handling We are missing some handling of PBR and SHARP protocols for netlink operations w/ the linux kernel. Additionally add a bread crumb for new developers( or existing ) to know to fixup the rt_netlink.c when we start handling new route types to hand to the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-04-27 20:53:46 +02:00
case RTPROT_SHARP:
proto = ZEBRA_ROUTE_SHARP;
break;
case RTPROT_PBR:
proto = ZEBRA_ROUTE_PBR;
break;
zebra: add RTPROT_FABRICD route type 197 Add an iproute2 route type for fabricd Signed-off-by: Christian Franke <chris@opensourcerouting.org>
2018-06-11 14:09:23 +02:00
case RTPROT_OPENFABRIC:
proto = ZEBRA_ROUTE_OPENFABRIC;
break;
zebra: Sweep our nexthop objects out on restart On restart, if we failed to remove any nexthop objects due to a kill -9 or such event, sweep them if we aren't using them. Add a proto field to handle this and remove the is_kernel bool. Add a dupicate flag that indicates this nexthop group is only present in our ID hashtable. It is a dupicate nexthop we received from the kernel, therefore we cannot hash on it. Make the idcounter globally accessible so that kernel updates increment it as soon as we receive them, not when we handle them. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-01 20:07:04 +02:00
case RTPROT_ZEBRA:
if (is_nexthop) {
proto = ZEBRA_ROUTE_NHG;
break;
}
/* Intentional fall thru */
zebra: Allow zebra to delete self originated routes With the change to make zebra pass routes to the kernel with the 'correct' proto name, it caused zebra to not properly recognize them on startup again the next time such that the route would not be deleted. Modify rt_netlink.c to notice that we have a self originated route and to properly mark the type of route it was. Modify rib_table_sweep to mark the nexthops as active so that when we go to delete the self originated routes it would properly delete from the kernel. Fixes: #1061 Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-09-01 16:28:19 +02:00
default:
zebra: Add PBR and SHARP handling We are missing some handling of PBR and SHARP protocols for netlink operations w/ the linux kernel. Additionally add a bread crumb for new developers( or existing ) to know to fixup the rt_netlink.c when we start handling new route types to hand to the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-04-27 20:53:46 +02:00
/*
* When a user adds a new protocol this will show up
* to let them know to do something about it. This
* is intentionally a warn because we should see
* this as part of development of a new protocol
*/
zebra: flog_warn conversion Convert Zebra to user error subsystem. Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-08-16 22:10:32 +02:00
zlog_debug(
"%s: Please add this protocol(%d) to proper rt_netlink.c handling",
__PRETTY_FUNCTION__, proto);
zebra: Allow zebra to delete self originated routes With the change to make zebra pass routes to the kernel with the 'correct' proto name, it caused zebra to not properly recognize them on startup again the next time such that the route would not be deleted. Modify rt_netlink.c to notice that we have a self originated route and to properly mark the type of route it was. Modify rib_table_sweep to mark the nexthops as active so that when we go to delete the self originated routes it would properly delete from the kernel. Fixes: #1061 Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-09-01 16:28:19 +02:00
proto = ZEBRA_ROUTE_KERNEL;
break;
}
return proto;
}
lib, zebra: The Bulk of the conversion over to NS and VRF Convert the rest of zebra over to use a Namespae and VRF. Signed-off-by: Vipin Kumar <vipin@cumulusnetworks.com> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2016-02-01 19:55:42 +01:00
/*
Pending: create an efficient table_id (in a tree/hash) based lookup)
*/
*: use C99 standard fixed-width integer types The following types are nonstandard: - u_char - u_short - u_int - u_long - u_int8_t - u_int16_t - u_int32_t Replace them with the C99 standard types: - uint8_t - unsigned short - unsigned int - unsigned long - uint8_t - uint16_t - uint32_t Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-03-27 21:13:34 +02:00
static vrf_id_t vrf_lookup_by_table(uint32_t table_id, ns_id_t ns_id)
lib, zebra: The Bulk of the conversion over to NS and VRF Convert the rest of zebra over to use a Namespae and VRF. Signed-off-by: Vipin Kumar <vipin@cumulusnetworks.com> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2016-02-01 19:55:42 +01:00
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
struct vrf *vrf;
struct zebra_vrf *zvrf;
lib, zebra: The Bulk of the conversion over to NS and VRF Convert the rest of zebra over to use a Namespae and VRF. Signed-off-by: Vipin Kumar <vipin@cumulusnetworks.com> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2016-02-01 19:55:42 +01:00
*: use clang's 'ForEachMacros' format style option This fixes the broken indentation of several foreach loops throughout the code. From clang's documentation[1]: ForEachMacros: A vector of macros that should be interpreted as foreach loops instead of as function calls. [1] http://clang.llvm.org/docs/ClangFormatStyleOptions.html Signed-off-by: Renato Westphal <renato@opensourcerouting.org>
2017-09-15 17:47:35 +02:00
RB_FOREACH (vrf, vrf_id_head, &vrfs_by_id) {
zebra: add a runtime flag to enable vrf with netns The netns backend is chosen by VRF if a runtime flag named vrfwnetns is selected when running zebra. In the case the NETNS backend is chosen, in some case the VRFID value is being assigned the value of the NSID. Within the perimeter of that work, this is why the vrf_lookup_by_table function is extended with a new parameter. Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
2018-01-22 09:42:53 +01:00
zvrf = vrf->info;
if (zvrf == NULL)
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
continue;
zebra: add a runtime flag to enable vrf with netns The netns backend is chosen by VRF if a runtime flag named vrfwnetns is selected when running zebra. In the case the NETNS backend is chosen, in some case the VRFID value is being assigned the value of the NSID. Within the perimeter of that work, this is why the vrf_lookup_by_table function is extended with a new parameter. Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
2018-01-22 09:42:53 +01:00
/* case vrf with netns : match the netnsid */
if (vrf_is_backend_netns()) {
if (ns_id == zvrf_id(zvrf))
return zvrf_id(zvrf);
} else {
/* VRF is VRF_BACKEND_VRF_LITE */
if (zvrf->table_id != table_id)
continue;
return zvrf_id(zvrf);
}
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
}
lib, zebra: The Bulk of the conversion over to NS and VRF Convert the rest of zebra over to use a Namespae and VRF. Signed-off-by: Vipin Kumar <vipin@cumulusnetworks.com> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2016-02-01 19:55:42 +01:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return VRF_DEFAULT;
lib, zebra: The Bulk of the conversion over to NS and VRF Convert the rest of zebra over to use a Namespae and VRF. Signed-off-by: Vipin Kumar <vipin@cumulusnetworks.com> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2016-02-01 19:55:42 +01:00
}
zebra: Add support for static encap mpls labels We were ignoring mpls labels encapped with static routes. Added support for single and multipath labels. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-08-12 00:36:22 +02:00
/**
* @parse_encap_mpls() - Parses encapsulated mpls attributes
* @tb: Pointer to rtattr to look for nested items in.
* @labels: Pointer to store labels in.
*
* Return: Number of mpls labels found.
*/
static int parse_encap_mpls(struct rtattr *tb, mpls_label_t *labels)
{
struct rtattr *tb_encap[MPLS_IPTUNNEL_MAX + 1] = {0};
mpls_lse_t *lses = NULL;
int num_labels = 0;
uint32_t ttl = 0;
uint32_t bos = 0;
uint32_t exp = 0;
mpls_label_t label = 0;
netlink_parse_rtattr_nested(tb_encap, MPLS_IPTUNNEL_MAX, tb);
lses = (mpls_lse_t *)RTA_DATA(tb_encap[MPLS_IPTUNNEL_DST]);
while (!bos && num_labels < MPLS_MAX_LABELS) {
mpls_lse_decode(lses[num_labels], &label, &ttl, &exp, &bos);
labels[num_labels++] = label;
}
return num_labels;
}
zebra: Put unicast nexthop parsing into its own function Move the nexthop unicast parsing into its own function to improve code readability. It was getting a bit too indented. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-11 15:52:33 +01:00
static struct nexthop
parse_nexthop_unicast(ns_id_t ns_id, struct rtmsg *rtm, struct rtattr **tb,
enum blackhole_type bh_type, int index, void *prefsrc,
zebra: Add equivalence function for nhg_depends Add a helper function to allow us to check if two nhg_hash_entry's dependency lists are equal. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-29 15:56:52 +01:00
void *gate, afi_t afi, vrf_id_t vrf_id)
zebra: Put unicast nexthop parsing into its own function Move the nexthop unicast parsing into its own function to improve code readability. It was getting a bit too indented. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-11 15:52:33 +01:00
{
struct interface *ifp = NULL;
struct nexthop nh = {0};
mpls_label_t labels[MPLS_MAX_LABELS] = {0};
int num_labels = 0;
zebra: Add equivalence function for nhg_depends Add a helper function to allow us to check if two nhg_hash_entry's dependency lists are equal. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-29 15:56:52 +01:00
vrf_id_t nh_vrf_id = vrf_id;
zebra: Put unicast nexthop parsing into its own function Move the nexthop unicast parsing into its own function to improve code readability. It was getting a bit too indented. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-11 15:52:33 +01:00
size_t sz = (afi == AFI_IP) ? 4 : 16;
if (bh_type == BLACKHOLE_UNSPEC) {
if (index && !gate)
nh.type = NEXTHOP_TYPE_IFINDEX;
else if (index && gate)
nh.type = (afi == AFI_IP) ? NEXTHOP_TYPE_IPV4_IFINDEX
: NEXTHOP_TYPE_IPV6_IFINDEX;
else if (!index && gate)
nh.type = (afi == AFI_IP) ? NEXTHOP_TYPE_IPV4
: NEXTHOP_TYPE_IPV6;
else {
nh.type = NEXTHOP_TYPE_BLACKHOLE;
nh.bh_type = bh_type;
}
} else {
nh.type = NEXTHOP_TYPE_BLACKHOLE;
nh.bh_type = bh_type;
}
nh.ifindex = index;
if (prefsrc)
memcpy(&nh.src, prefsrc, sz);
if (gate)
memcpy(&nh.gate, gate, sz);
if (index) {
ifp = if_lookup_by_index_per_ns(zebra_ns_lookup(ns_id), index);
if (ifp)
nh_vrf_id = ifp->vrf_id;
}
nh.vrf_id = nh_vrf_id;
if (tb[RTA_ENCAP] && tb[RTA_ENCAP_TYPE]
&& *(uint16_t *)RTA_DATA(tb[RTA_ENCAP_TYPE])
== LWTUNNEL_ENCAP_MPLS) {
num_labels = parse_encap_mpls(tb[RTA_ENCAP], labels);
}
if (rtm->rtm_flags & RTNH_F_ONLINK)
SET_FLAG(nh.flags, NEXTHOP_FLAG_ONLINK);
if (num_labels)
nexthop_add_labels(&nh, ZEBRA_LSP_STATIC, num_labels, labels);
return nh;
}
zebra: Add equivalence function for nhg_depends Add a helper function to allow us to check if two nhg_hash_entry's dependency lists are equal. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-29 15:56:52 +01:00
static uint8_t parse_multipath_nexthops_unicast(ns_id_t ns_id,
struct route_entry *re,
struct rtmsg *rtm,
struct rtnexthop *rtnh,
struct rtattr **tb,
void *prefsrc, vrf_id_t vrf_id)
{
void *gate = NULL;
struct interface *ifp = NULL;
int index = 0;
/* MPLS labels */
mpls_label_t labels[MPLS_MAX_LABELS] = {0};
int num_labels = 0;
struct rtattr *rtnh_tb[RTA_MAX + 1] = {};
int len = RTA_PAYLOAD(tb[RTA_MULTIPATH]);
vrf_id_t nh_vrf_id = vrf_id;
re->ng = nexthop_group_new();
for (;;) {
struct nexthop *nh = NULL;
if (len < (int)sizeof(*rtnh) || rtnh->rtnh_len > len)
break;
index = rtnh->rtnh_ifindex;
if (index) {
/*
* Yes we are looking this up
* for every nexthop and just
* using the last one looked
* up right now
*/
ifp = if_lookup_by_index_per_ns(zebra_ns_lookup(ns_id),
index);
if (ifp)
nh_vrf_id = ifp->vrf_id;
else {
flog_warn(
EC_ZEBRA_UNKNOWN_INTERFACE,
"%s: Unknown interface %u specified, defaulting to VRF_DEFAULT",
__PRETTY_FUNCTION__, index);
nh_vrf_id = VRF_DEFAULT;
}
} else
nh_vrf_id = vrf_id;
if (rtnh->rtnh_len > sizeof(*rtnh)) {
memset(rtnh_tb, 0, sizeof(rtnh_tb));
netlink_parse_rtattr(rtnh_tb, RTA_MAX, RTNH_DATA(rtnh),
rtnh->rtnh_len - sizeof(*rtnh));
if (rtnh_tb[RTA_GATEWAY])
gate = RTA_DATA(rtnh_tb[RTA_GATEWAY]);
if (rtnh_tb[RTA_ENCAP] && rtnh_tb[RTA_ENCAP_TYPE]
&& *(uint16_t *)RTA_DATA(rtnh_tb[RTA_ENCAP_TYPE])
== LWTUNNEL_ENCAP_MPLS) {
num_labels = parse_encap_mpls(
rtnh_tb[RTA_ENCAP], labels);
}
}
zebra: rt_netlink nexthop handling checkpatch Checkpatch was complaining because this code was extending beyond 80 characters on a couple lines. Adjusted a conditional tree to fix that. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-10-25 17:11:06 +02:00
if (gate && rtm->rtm_family == AF_INET) {
if (index)
nh = route_entry_nexthop_ipv4_ifindex_add(
re, gate, prefsrc, index, nh_vrf_id);
else
nh = route_entry_nexthop_ipv4_add(
re, gate, prefsrc, nh_vrf_id);
} else if (gate && rtm->rtm_family == AF_INET6) {
if (index)
nh = route_entry_nexthop_ipv6_ifindex_add(
re, gate, index, nh_vrf_id);
else
nh = route_entry_nexthop_ipv6_add(re, gate,
nh_vrf_id);
zebra: Add equivalence function for nhg_depends Add a helper function to allow us to check if two nhg_hash_entry's dependency lists are equal. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-29 15:56:52 +01:00
} else
nh = route_entry_nexthop_ifindex_add(re, index,
nh_vrf_id);
if (nh) {
if (num_labels)
nexthop_add_labels(nh, ZEBRA_LSP_STATIC,
num_labels, labels);
if (rtnh->rtnh_flags & RTNH_F_ONLINK)
SET_FLAG(nh->flags, NEXTHOP_FLAG_ONLINK);
}
if (rtnh->rtnh_len == 0)
break;
len -= NLMSG_ALIGN(rtnh->rtnh_len);
rtnh = RTNH_NEXT(rtnh);
}
uint8_t nhop_num = nexthop_group_nexthop_num(re->ng);
if (!nhop_num)
nexthop_group_delete(&re->ng);
return nhop_num;
}
Initial revision
2002-12-13 21:15:29 +01:00
/* Looking up routing table by netlink interface. */
zebra: Remove unnecessary function parameter The snl variable is no longer needed to be passed around, so remove it from the calling path. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-05-23 15:37:06 +02:00
static int netlink_route_change_read_unicast(struct nlmsghdr *h, ns_id_t ns_id,
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
int startup)
Initial revision
2002-12-13 21:15:29 +01:00
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
int len;
struct rtmsg *rtm;
struct rtattr *tb[RTA_MAX + 1];
*: use C99 standard fixed-width integer types The following types are nonstandard: - u_char - u_short - u_int - u_long - u_int8_t - u_int16_t - u_int32_t Replace them with the C99 standard types: - uint8_t - unsigned short - unsigned int - unsigned long - uint8_t - uint16_t - uint32_t Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-03-27 21:13:34 +02:00
uint8_t flags = 0;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
struct prefix p;
zebra: fix clang warning Set to 0 to avoid "Branch condition evaluates to a garbage value". Signed-off-by: Vincent Jardin <vincent.jardin@6wind.com>
2017-10-08 23:47:43 +02:00
struct prefix_ipv6 src_p = {};
zebra: add a runtime flag to enable vrf with netns The netns backend is chosen by VRF if a runtime flag named vrfwnetns is selected when running zebra. In the case the NETNS backend is chosen, in some case the VRFID value is being assigned the value of the NSID. Within the perimeter of that work, this is why the vrf_lookup_by_table function is extended with a new parameter. Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
2018-01-22 09:42:53 +01:00
vrf_id_t vrf_id;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
char anyaddr[16] = {0};
zebra: Allow zebra to delete self originated routes With the change to make zebra pass routes to the kernel with the 'correct' proto name, it caused zebra to not properly recognize them on startup again the next time such that the route would not be deleted. Modify rt_netlink.c to notice that we have a self originated route and to properly mark the type of route it was. Modify rib_table_sweep to mark the nexthops as active so that when we go to delete the self originated routes it would properly delete from the kernel. Fixes: #1061 Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-09-01 16:28:19 +02:00
int proto = ZEBRA_ROUTE_KERNEL;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
int index = 0;
int table;
int metric = 0;
*: use C99 standard fixed-width integer types The following types are nonstandard: - u_char - u_short - u_int - u_long - u_int8_t - u_int16_t - u_int32_t Replace them with the C99 standard types: - uint8_t - unsigned short - unsigned int - unsigned long - uint8_t - uint16_t - uint32_t Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-03-27 21:13:34 +02:00
uint32_t mtu = 0;
zebra: Modify metric read to be admin distance and metric For ZEBRA_ROUTE_KERNEL types: The metric/priority of the route received from the kernel is a 32 bit number. We are going to interpret the high order byte as the Admin Distance and the low order 3 bytes as the metric. This will allow us to do two things: 1) Allow the creation of kernel routes that can be overridden by zebra. 2) Allow the old behavior for 'most' kernel route types if a user enters 'ip route ...' v4 routes get a metric of 0 and v6 routes get a metric of 1024. Both of these values will end up with a admin distance of 0, which will cause them to win for the purposes of zebra. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-09-25 14:37:04 +02:00
uint8_t distance = 0;
zebra: Add ability to support tags -> realms in linux Linux has the ability to support a concept of 'realms'. This concept allows you to mark routes with a realm id value of 1-255. If you have marked the realm of a route then you can use the tc program to apply policy to the routes. This commit adds the ability of FRR to interpret a tag from (1-255) as a realm when installing into the kernel. Please note that at this point in time there is no way to set policy from within FRR. This must be done outside of it. The normal methodology for setting tags is valid here via a route-map. Finally this is only applied if the --enable-realms configure option is applied. Signed-off-by: Kaloyan Kovachev <kkovachev@varna.net>
2017-12-08 20:44:15 +01:00
route_tag_t tag = 0;
zebra: Parse in nexthop ID information from new routes Add parsing code for nexthop object ID's when we get a route. When we get a new route with the new kernel, it will come with a nexthop ID and the nexthop full info. We should just reference by ID if it exists and point to the nexthop hash entry that matches it. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-11 15:46:25 +01:00
uint32_t nhe_id = 0;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
void *dest = NULL;
void *gate = NULL;
void *prefsrc = NULL; /* IPv4 preferred source host address */
void *src = NULL; /* IPv6 srcdest source prefix */
zebra: netlink: get blackhole routes from kernel support processing of RTN_BLACKHOLE et al. from kernel and dump them into appropriate blackhole rib entries. Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2010-02-05 04:58:46 +01:00
enum blackhole_type bh_type = BLACKHOLE_UNSPEC;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
rtm = NLMSG_DATA(h);
if (startup && h->nlmsg_type != RTM_NEWROUTE)
return 0;
zebra: netlink: get blackhole routes from kernel support processing of RTN_BLACKHOLE et al. from kernel and dump them into appropriate blackhole rib entries. Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2010-02-05 04:58:46 +01:00
switch (rtm->rtm_type) {
case RTN_UNICAST:
break;
case RTN_BLACKHOLE:
bh_type = BLACKHOLE_NULL;
break;
case RTN_UNREACHABLE:
bh_type = BLACKHOLE_REJECT;
break;
case RTN_PROHIBIT:
bh_type = BLACKHOLE_ADMINPROHIB;
break;
default:
zebra: Increase debugs to understand why we rejected a kernel route Add a bit of extra code to indicate to the operator why we intentionally rejected a kernel route from being used. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-06-18 14:47:28 +02:00
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug("Route rtm_type: %s(%d) intentionally ignoring",
nl_rttype_to_str(rtm->rtm_type),
rtm->rtm_type);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return 0;
zebra: netlink: get blackhole routes from kernel support processing of RTN_BLACKHOLE et al. from kernel and dump them into appropriate blackhole rib entries. Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2010-02-05 04:58:46 +01:00
}
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
len = h->nlmsg_len - NLMSG_LENGTH(sizeof(struct rtmsg));
zebra: Add error messages in places where we have a decode problem When we receive a netlink message from the kernel we have handler functions for when we send a netlink command, if these return a failure ( < 0 ) then we output that we had a parse issue. But if all we get is: 2018-06-21T23:47:45.298156+00:00 qct-ix1-08 zebra[1484]: netlink-cmd (NS 0) filter function error Then it is not very useful to figure out *where* the error happened. Add more error code when in a decode path to hopefully allow us to figure out where this message is coming from. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-06-22 20:22:02 +02:00
if (len < 0) {
zlog_err("%s: Message received from netlink is of a broken size %d %zu",
__PRETTY_FUNCTION__, h->nlmsg_len,
(size_t)NLMSG_LENGTH(sizeof(struct rtmsg)));
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return -1;
zebra: Add error messages in places where we have a decode problem When we receive a netlink message from the kernel we have handler functions for when we send a netlink command, if these return a failure ( < 0 ) then we output that we had a parse issue. But if all we get is: 2018-06-21T23:47:45.298156+00:00 qct-ix1-08 zebra[1484]: netlink-cmd (NS 0) filter function error Then it is not very useful to figure out *where* the error happened. Add more error code when in a decode path to hopefully allow us to figure out where this message is coming from. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-06-22 20:22:02 +02:00
}
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
memset(tb, 0, sizeof tb);
netlink_parse_rtattr(tb, RTA_MAX, RTM_RTA(rtm), len);
if (rtm->rtm_flags & RTM_F_CLONED)
return 0;
if (rtm->rtm_protocol == RTPROT_REDIRECT)
return 0;
if (rtm->rtm_protocol == RTPROT_KERNEL)
return 0;
if (!startup && is_selfroute(rtm->rtm_protocol)
zebra: Add a breadcrumb for when we ignore a route When we receive a route that we think we own and we are not in startup conditions, then add a small debug to help debug the issue when this happens, instead of silently just ignoring the route. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-05-25 20:45:16 +02:00
&& h->nlmsg_type == RTM_NEWROUTE) {
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug("Route type: %d Received that we think we have originated, ignoring",
rtm->rtm_protocol);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return 0;
zebra: Add a breadcrumb for when we ignore a route When we receive a route that we think we own and we are not in startup conditions, then add a small debug to help debug the issue when this happens, instead of silently just ignoring the route. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-05-25 20:45:16 +02:00
}
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
/* We don't care about change notifications for the MPLS table. */
/* TODO: Revisit this. */
if (rtm->rtm_family == AF_MPLS)
return 0;
/* Table corresponding to route. */
if (tb[RTA_TABLE])
table = *(int *)RTA_DATA(tb[RTA_TABLE]);
else
table = rtm->rtm_table;
/* Map to VRF */
zebra: add a runtime flag to enable vrf with netns The netns backend is chosen by VRF if a runtime flag named vrfwnetns is selected when running zebra. In the case the NETNS backend is chosen, in some case the VRFID value is being assigned the value of the NSID. Within the perimeter of that work, this is why the vrf_lookup_by_table function is extended with a new parameter. Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
2018-01-22 09:42:53 +01:00
vrf_id = vrf_lookup_by_table(table, ns_id);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (vrf_id == VRF_DEFAULT) {
if (!is_zebra_valid_kernel_table(table)
&& !is_zebra_main_routing_table(table))
return 0;
}
/* Route which inserted by Zebra. */
zebra: Allow zebra to delete self originated routes With the change to make zebra pass routes to the kernel with the 'correct' proto name, it caused zebra to not properly recognize them on startup again the next time such that the route would not be deleted. Modify rt_netlink.c to notice that we have a self originated route and to properly mark the type of route it was. Modify rib_table_sweep to mark the nexthops as active so that when we go to delete the self originated routes it would properly delete from the kernel. Fixes: #1061 Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-09-01 16:28:19 +02:00
if (is_selfroute(rtm->rtm_protocol)) {
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
flags |= ZEBRA_FLAG_SELFROUTE;
zebra: Sweep our nexthop objects out on restart On restart, if we failed to remove any nexthop objects due to a kill -9 or such event, sweep them if we aren't using them. Add a proto field to handle this and remove the is_kernel bool. Add a dupicate flag that indicates this nexthop group is only present in our ID hashtable. It is a dupicate nexthop we received from the kernel, therefore we cannot hash on it. Make the idcounter globally accessible so that kernel updates increment it as soon as we receive them, not when we handle them. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-01 20:07:04 +02:00
proto = proto2zebra(rtm->rtm_protocol, rtm->rtm_family, false);
zebra: Allow zebra to delete self originated routes With the change to make zebra pass routes to the kernel with the 'correct' proto name, it caused zebra to not properly recognize them on startup again the next time such that the route would not be deleted. Modify rt_netlink.c to notice that we have a self originated route and to properly mark the type of route it was. Modify rib_table_sweep to mark the nexthops as active so that when we go to delete the self originated routes it would properly delete from the kernel. Fixes: #1061 Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-09-01 16:28:19 +02:00
}
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (tb[RTA_OIF])
index = *(int *)RTA_DATA(tb[RTA_OIF]);
if (tb[RTA_DST])
dest = RTA_DATA(tb[RTA_DST]);
else
dest = anyaddr;
if (tb[RTA_SRC])
src = RTA_DATA(tb[RTA_SRC]);
else
src = anyaddr;
if (tb[RTA_PREFSRC])
prefsrc = RTA_DATA(tb[RTA_PREFSRC]);
if (tb[RTA_GATEWAY])
gate = RTA_DATA(tb[RTA_GATEWAY]);
zebra: Parse in nexthop ID information from new routes Add parsing code for nexthop object ID's when we get a route. When we get a new route with the new kernel, it will come with a nexthop ID and the nexthop full info. We should just reference by ID if it exists and point to the nexthop hash entry that matches it. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-11 15:46:25 +01:00
if (tb[RTA_NH_ID])
nhe_id = *(uint32_t *)RTA_DATA(tb[RTA_NH_ID]);
zebra: Pay attention to metric from kernel When the linux kernel adds/deletes routes, the metric is important, but our routing protocols add/delete in a slightly different manner, so allow kernel metrics to match so that our rib matches the kernel's fib. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-08-09 15:13:33 +02:00
if (tb[RTA_PRIORITY])
metric = *(int *)RTA_DATA(tb[RTA_PRIORITY]);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
zebra: Add ability to support tags -> realms in linux Linux has the ability to support a concept of 'realms'. This concept allows you to mark routes with a realm id value of 1-255. If you have marked the realm of a route then you can use the tc program to apply policy to the routes. This commit adds the ability of FRR to interpret a tag from (1-255) as a realm when installing into the kernel. Please note that at this point in time there is no way to set policy from within FRR. This must be done outside of it. The normal methodology for setting tags is valid here via a route-map. Finally this is only applied if the --enable-realms configure option is applied. Signed-off-by: Kaloyan Kovachev <kkovachev@varna.net>
2017-12-08 20:44:15 +01:00
#if defined(SUPPORT_REALMS)
if (tb[RTA_FLOW])
tag = *(uint32_t *)RTA_DATA(tb[RTA_FLOW]);
#endif
zebra: Pay attention to metric from kernel When the linux kernel adds/deletes routes, the metric is important, but our routing protocols add/delete in a slightly different manner, so allow kernel metrics to match so that our rib matches the kernel's fib. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-08-09 15:13:33 +02:00
if (tb[RTA_METRICS]) {
struct rtattr *mxrta[RTAX_MAX + 1];
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
zebra: Pay attention to metric from kernel When the linux kernel adds/deletes routes, the metric is important, but our routing protocols add/delete in a slightly different manner, so allow kernel metrics to match so that our rib matches the kernel's fib. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-08-09 15:13:33 +02:00
memset(mxrta, 0, sizeof mxrta);
*: conform with COMMUNITY.md formatting rules, via 'make indent' Signed-off-by: Lou Berger <lberger@labn.net>
2018-03-06 20:02:52 +01:00
netlink_parse_rtattr(mxrta, RTAX_MAX, RTA_DATA(tb[RTA_METRICS]),
zebra: Pay attention to metric from kernel When the linux kernel adds/deletes routes, the metric is important, but our routing protocols add/delete in a slightly different manner, so allow kernel metrics to match so that our rib matches the kernel's fib. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-08-09 15:13:33 +02:00
RTA_PAYLOAD(tb[RTA_METRICS]));
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
zebra: Pay attention to metric from kernel When the linux kernel adds/deletes routes, the metric is important, but our routing protocols add/delete in a slightly different manner, so allow kernel metrics to match so that our rib matches the kernel's fib. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-08-09 15:13:33 +02:00
if (mxrta[RTAX_MTU])
*: use C99 standard fixed-width integer types The following types are nonstandard: - u_char - u_short - u_int - u_long - u_int8_t - u_int16_t - u_int32_t Replace them with the C99 standard types: - uint8_t - unsigned short - unsigned int - unsigned long - uint8_t - uint16_t - uint32_t Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-03-27 21:13:34 +02:00
mtu = *(uint32_t *)RTA_DATA(mxrta[RTAX_MTU]);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
}
if (rtm->rtm_family == AF_INET) {
p.family = AF_INET;
zebra: Add check for prefix length from kernel messages Zebra needed a check that varifies the prefix length of an address is a valid length when receiving route changes and interface address changes. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2018-07-21 05:20:28 +02:00
if (rtm->rtm_dst_len > IPV4_MAX_BITLEN) {
zebra: Make prefix length check return error Prefix length validation checks should be returning an error rather than 0. Switch to that and make them error messages. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2018-07-26 21:10:53 +02:00
zlog_err(
Change log statements to use unsigned format Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2018-07-24 17:14:07 +02:00
"Invalid destination prefix length: %u received from kernel route change",
zebra: Add check for prefix length from kernel messages Zebra needed a check that varifies the prefix length of an address is a valid length when receiving route changes and interface address changes. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2018-07-21 05:20:28 +02:00
rtm->rtm_dst_len);
zebra: Make prefix length check return error Prefix length validation checks should be returning an error rather than 0. Switch to that and make them error messages. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2018-07-26 21:10:53 +02:00
return -1;
zebra: Add check for prefix length from kernel messages Zebra needed a check that varifies the prefix length of an address is a valid length when receiving route changes and interface address changes. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2018-07-21 05:20:28 +02:00
}
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
memcpy(&p.u.prefix4, dest, 4);
p.prefixlen = rtm->rtm_dst_len;
zebra: do not ignore ipv6 srcdest routes Commit a2ca67d1d2 consolidated IPv4 and IPv6 handling. It also applied our ignorance for IPv4 srcdest routes onto IPv6. Signed-off-by: Christian Franke <chris@opensourcerouting.org>
2018-07-23 03:40:42 +02:00
if (rtm->rtm_src_len != 0) {
char buf[PREFIX_STRLEN];
zebra: flog_warn conversion Convert Zebra to user error subsystem. Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-08-16 22:10:32 +02:00
flog_warn(
zebra: ZEBRA_[ERR|WARN] -> EC_ZEBRA Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-09-13 21:21:05 +02:00
EC_ZEBRA_UNSUPPORTED_V4_SRCDEST,
zebra: flog_warn conversion Convert Zebra to user error subsystem. Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-08-16 22:10:32 +02:00
"unsupported IPv4 sourcedest route (dest %s vrf %u)",
prefix2str(&p, buf, sizeof(buf)), vrf_id);
zebra: do not ignore ipv6 srcdest routes Commit a2ca67d1d2 consolidated IPv4 and IPv6 handling. It also applied our ignorance for IPv4 srcdest routes onto IPv6. Signed-off-by: Christian Franke <chris@opensourcerouting.org>
2018-07-23 03:40:42 +02:00
return 0;
}
zebra: Add check for prefix length from kernel messages Zebra needed a check that varifies the prefix length of an address is a valid length when receiving route changes and interface address changes. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2018-07-21 05:20:28 +02:00
zebra: do not ignore ipv6 srcdest routes Commit a2ca67d1d2 consolidated IPv4 and IPv6 handling. It also applied our ignorance for IPv4 srcdest routes onto IPv6. Signed-off-by: Christian Franke <chris@opensourcerouting.org>
2018-07-23 03:40:42 +02:00
/* Force debug below to not display anything for source */
src_p.prefixlen = 0;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
} else if (rtm->rtm_family == AF_INET6) {
p.family = AF_INET6;
zebra: Add check for prefix length from kernel messages Zebra needed a check that varifies the prefix length of an address is a valid length when receiving route changes and interface address changes. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2018-07-21 05:20:28 +02:00
if (rtm->rtm_dst_len > IPV6_MAX_BITLEN) {
zebra: Make prefix length check return error Prefix length validation checks should be returning an error rather than 0. Switch to that and make them error messages. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2018-07-26 21:10:53 +02:00
zlog_err(
Change log statements to use unsigned format Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2018-07-24 17:14:07 +02:00
"Invalid destination prefix length: %u received from kernel route change",
zebra: Add check for prefix length from kernel messages Zebra needed a check that varifies the prefix length of an address is a valid length when receiving route changes and interface address changes. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2018-07-21 05:20:28 +02:00
rtm->rtm_dst_len);
zebra: Make prefix length check return error Prefix length validation checks should be returning an error rather than 0. Switch to that and make them error messages. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2018-07-26 21:10:53 +02:00
return -1;
zebra: Add check for prefix length from kernel messages Zebra needed a check that varifies the prefix length of an address is a valid length when receiving route changes and interface address changes. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2018-07-21 05:20:28 +02:00
}
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
memcpy(&p.u.prefix6, dest, 16);
p.prefixlen = rtm->rtm_dst_len;
src_p.family = AF_INET6;
zebra: Add check for prefix length from kernel messages Zebra needed a check that varifies the prefix length of an address is a valid length when receiving route changes and interface address changes. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2018-07-21 05:20:28 +02:00
if (rtm->rtm_src_len > IPV6_MAX_BITLEN) {
zebra: Make prefix length check return error Prefix length validation checks should be returning an error rather than 0. Switch to that and make them error messages. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2018-07-26 21:10:53 +02:00
zlog_err(
Change log statements to use unsigned format Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2018-07-24 17:14:07 +02:00
"Invalid source prefix length: %u received from kernel route change",
zebra: Add check for prefix length from kernel messages Zebra needed a check that varifies the prefix length of an address is a valid length when receiving route changes and interface address changes. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2018-07-21 05:20:28 +02:00
rtm->rtm_src_len);
zebra: Make prefix length check return error Prefix length validation checks should be returning an error rather than 0. Switch to that and make them error messages. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2018-07-26 21:10:53 +02:00
return -1;
zebra: Add check for prefix length from kernel messages Zebra needed a check that varifies the prefix length of an address is a valid length when receiving route changes and interface address changes. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2018-07-21 05:20:28 +02:00
}
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
memcpy(&src_p.prefix, src, 16);
src_p.prefixlen = rtm->rtm_src_len;
}
zebra: Modify metric read to be admin distance and metric For ZEBRA_ROUTE_KERNEL types: The metric/priority of the route received from the kernel is a 32 bit number. We are going to interpret the high order byte as the Admin Distance and the low order 3 bytes as the metric. This will allow us to do two things: 1) Allow the creation of kernel routes that can be overridden by zebra. 2) Allow the old behavior for 'most' kernel route types if a user enters 'ip route ...' v4 routes get a metric of 0 and v6 routes get a metric of 1024. Both of these values will end up with a admin distance of 0, which will cause them to win for the purposes of zebra. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-09-25 14:37:04 +02:00
/*
* For ZEBRA_ROUTE_KERNEL types:
*
* The metric/priority of the route received from the kernel
* is a 32 bit number. We are going to interpret the high
* order byte as the Admin Distance and the low order 3 bytes
* as the metric.
*
* This will allow us to do two things:
* 1) Allow the creation of kernel routes that can be
* overridden by zebra.
* 2) Allow the old behavior for 'most' kernel route types
* if a user enters 'ip route ...' v4 routes get a metric
* of 0 and v6 routes get a metric of 1024. Both of these
* values will end up with a admin distance of 0, which
* will cause them to win for the purposes of zebra.
*/
if (proto == ZEBRA_ROUTE_KERNEL) {
distance = (metric >> 24) & 0xFF;
*: conform with COMMUNITY.md formatting rules, via 'make indent' Signed-off-by: Lou Berger <lberger@labn.net>
2018-03-06 20:02:52 +01:00
metric = (metric & 0x00FFFFFF);
zebra: Modify metric read to be admin distance and metric For ZEBRA_ROUTE_KERNEL types: The metric/priority of the route received from the kernel is a 32 bit number. We are going to interpret the high order byte as the Admin Distance and the low order 3 bytes as the metric. This will allow us to do two things: 1) Allow the creation of kernel routes that can be overridden by zebra. 2) Allow the old behavior for 'most' kernel route types if a user enters 'ip route ...' v4 routes get a metric of 0 and v6 routes get a metric of 1024. Both of these values will end up with a admin distance of 0, which will cause them to win for the purposes of zebra. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-09-25 14:37:04 +02:00
}
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (IS_ZEBRA_DEBUG_KERNEL) {
char buf[PREFIX_STRLEN];
char buf2[PREFIX_STRLEN];
zebra: Cleanup debugs and add a bit more info This commit does 2 things: 1) When receiving a route from the kernel, display the incoming table as part of the debug, to facilatate knowing what we are talking about as part of the debug. 2) When displaying nexthop information for routes we were sending to the kernel, no need to display the route information every time Display the route then the individual nexthops for what we are doing. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-04-09 20:09:35 +02:00
zlog_debug("%s %s%s%s vrf %u(%u) metric: %d Admin Distance: %d",
*: conform with COMMUNITY.md formatting rules, via 'make indent' Signed-off-by: Lou Berger <lberger@labn.net>
2018-03-06 20:02:52 +01:00
nl_msg_type_to_str(h->nlmsg_type),
prefix2str(&p, buf, sizeof(buf)),
src_p.prefixlen ? " from " : "",
src_p.prefixlen
? prefix2str(&src_p, buf2, sizeof(buf2))
: "",
zebra: Cleanup debugs and add a bit more info This commit does 2 things: 1) When receiving a route from the kernel, display the incoming table as part of the debug, to facilatate knowing what we are talking about as part of the debug. 2) When displaying nexthop information for routes we were sending to the kernel, no need to display the route information every time Display the route then the individual nexthops for what we are doing. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-04-09 20:09:35 +02:00
vrf_id, table, metric, distance);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
}
afi_t afi = AFI_IP;
if (rtm->rtm_family == AF_INET6)
afi = AFI_IP6;
if (h->nlmsg_type == RTM_NEWROUTE) {
zebra: Add nh_vrf_id to rib_add Add to the rib_add function the ability to pass in the nexthops vrf. Additionally when we decode the netlink message from the linux kernel, properly figure out the nexthops vrf_id. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-01-08 16:21:09 +01:00
zebra: rib: use nexthop ptr in rib_add/delete This simplifies the API for the following blackhole rework. Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-08-28 01:30:16 +02:00
if (!tb[RTA_MULTIPATH]) {
zebra: Put unicast nexthop parsing into its own function Move the nexthop unicast parsing into its own function to improve code readability. It was getting a bit too indented. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-11 15:52:33 +01:00
struct nexthop nh = {0};
zebra: Add nh_vrf_id to rib_add Add to the rib_add function the ability to pass in the nexthops vrf. Additionally when we decode the netlink message from the linux kernel, properly figure out the nexthops vrf_id. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-01-08 16:21:09 +01:00
zebra: Put unicast nexthop parsing into its own function Move the nexthop unicast parsing into its own function to improve code readability. It was getting a bit too indented. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-11 15:52:33 +01:00
if (!nhe_id) {
nh = parse_nexthop_unicast(
ns_id, rtm, tb, bh_type, index, prefsrc,
zebra: Add equivalence function for nhg_depends Add a helper function to allow us to check if two nhg_hash_entry's dependency lists are equal. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-29 15:56:52 +01:00
gate, afi, vrf_id);
zebra: Add support for static encap mpls labels We were ignoring mpls labels encapped with static routes. Added support for single and multipath labels. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-08-12 00:36:22 +02:00
}
*: Track vrfs per nexthop not per route entry Track the vfrs on a per nexthop basis instead of on a per route entry basis. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-02-08 15:12:12 +01:00
rib_add(afi, SAFI_UNICAST, vrf_id, proto, 0, flags, &p,
zebra: Update rib_add to take a nexthop ID Add a parameter to the rib_add function so that it takes a nexthop ID from the kernel if one is passed along with the route. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-11 15:55:53 +01:00
&src_p, &nh, nhe_id, table, metric, mtu,
distance, tag);
zebra: rib: use nexthop ptr in rib_add/delete This simplifies the API for the following blackhole rework. Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-08-28 01:30:16 +02:00
} else {
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
/* This is a multipath route */
struct route_entry *re;
struct rtnexthop *rtnh =
(struct rtnexthop *)RTA_DATA(tb[RTA_MULTIPATH]);
re = XCALLOC(MTYPE_RE, sizeof(struct route_entry));
zebra: Allow zebra to delete self originated routes With the change to make zebra pass routes to the kernel with the 'correct' proto name, it caused zebra to not properly recognize them on startup again the next time such that the route would not be deleted. Modify rt_netlink.c to notice that we have a self originated route and to properly mark the type of route it was. Modify rib_table_sweep to mark the nexthops as active so that when we go to delete the self originated routes it would properly delete from the kernel. Fixes: #1061 Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-09-01 16:28:19 +02:00
re->type = proto;
zebra: Modify metric read to be admin distance and metric For ZEBRA_ROUTE_KERNEL types: The metric/priority of the route received from the kernel is a 32 bit number. We are going to interpret the high order byte as the Admin Distance and the low order 3 bytes as the metric. This will allow us to do two things: 1) Allow the creation of kernel routes that can be overridden by zebra. 2) Allow the old behavior for 'most' kernel route types if a user enters 'ip route ...' v4 routes get a metric of 0 and v6 routes get a metric of 1024. Both of these values will end up with a admin distance of 0, which will cause them to win for the purposes of zebra. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-09-25 14:37:04 +02:00
re->distance = distance;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
re->flags = flags;
re->metric = metric;
re->mtu = mtu;
re->vrf_id = vrf_id;
re->table = table;
zebra: Switch to using monotime(NULL) for re->uptime The re->uptime usage of time(NULL) leaves it open to timing changes from outside influence. Switching to monotime allows us to ensure that we have a timestamp that is always increasing. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-05-11 07:44:42 +02:00
re->uptime = monotime(NULL);
zebra: Add ability to support tags -> realms in linux Linux has the ability to support a concept of 'realms'. This concept allows you to mark routes with a realm id value of 1-255. If you have marked the realm of a route then you can use the tc program to apply policy to the routes. This commit adds the ability of FRR to interpret a tag from (1-255) as a realm when installing into the kernel. Please note that at this point in time there is no way to set policy from within FRR. This must be done outside of it. The normal methodology for setting tags is valid here via a route-map. Finally this is only applied if the --enable-realms configure option is applied. Signed-off-by: Kaloyan Kovachev <kkovachev@varna.net>
2017-12-08 20:44:15 +01:00
re->tag = tag;
zebra: Make route entry nexthop groups point to our hash entry Make our route entry struct's re->ng nexthop group pointer just point to the nhe->nhg nexthop hash entry nexthop group. This will allow updates to the nexthop itself to propogate to our routes immediately. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-11 15:58:05 +01:00
re->nhe_id = nhe_id;
zebra: Add check for nexthop loop to prevent hanging Bad nexthop messages from netlink were causing zebra to hang here. Added a check to verify the length of the nexthop so it doesn't keep trying to read. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2018-07-24 00:33:53 +02:00
zebra: Add equivalence function for nhg_depends Add a helper function to allow us to check if two nhg_hash_entry's dependency lists are equal. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-29 15:56:52 +01:00
if (!nhe_id) {
uint8_t nhop_num =
parse_multipath_nexthops_unicast(
ns_id, re, rtm, rtnh, tb,
prefsrc, vrf_id);
zserv_nexthop_num_warn(
__func__, (const struct prefix *)&p,
nhop_num);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
}
zebra: Add equivalence function for nhg_depends Add a helper function to allow us to check if two nhg_hash_entry's dependency lists are equal. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-29 15:56:52 +01:00
if (nhe_id || re->ng)
zebra: do not ignore ipv6 srcdest routes Commit a2ca67d1d2 consolidated IPv4 and IPv6 handling. It also applied our ignorance for IPv4 srcdest routes onto IPv6. Signed-off-by: Christian Franke <chris@opensourcerouting.org>
2018-07-23 03:40:42 +02:00
rib_add_multipath(afi, SAFI_UNICAST, &p,
&src_p, re);
zebra: Add equivalence function for nhg_depends Add a helper function to allow us to check if two nhg_hash_entry's dependency lists are equal. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-29 15:56:52 +01:00
else
XFREE(MTYPE_RE, re);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
}
} else {
zebra: Use nexthop object id on route delete When we receive a route delete from the kernel and it contains a nexthop object id, use that to match against route gateways with instead of explicit nexthops. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-01 23:36:56 +02:00
if (nhe_id) {
rib_delete(afi, SAFI_UNICAST, vrf_id, proto, 0, flags,
&p, &src_p, NULL, nhe_id, table, metric,
distance, true);
} else {
if (!tb[RTA_MULTIPATH]) {
struct nexthop nh;
size_t sz = (afi == AFI_IP) ? 4 : 16;
memset(&nh, 0, sizeof(nh));
if (bh_type == BLACKHOLE_UNSPEC) {
if (index && !gate)
nh.type = NEXTHOP_TYPE_IFINDEX;
else if (index && gate)
nh.type =
(afi == AFI_IP)
? NEXTHOP_TYPE_IPV4_IFINDEX
: NEXTHOP_TYPE_IPV6_IFINDEX;
else if (!index && gate)
nh.type =
(afi == AFI_IP)
? NEXTHOP_TYPE_IPV4
: NEXTHOP_TYPE_IPV6;
else {
nh.type =
NEXTHOP_TYPE_BLACKHOLE;
nh.bh_type = BLACKHOLE_UNSPEC;
}
} else {
zebra: allow blackhole routes from the kernel to be deleted Fixes the following bug: % ip -6 route add unreachable 4000::/64 % ip -6 route del unreachable 4000::/64 % vtysh -c "show ipv6 route" [snip] K>* 4000::/64 [0/1024] unreachable (ICMP unreachable) Signed-off-by: Renato Westphal <renato@opensourcerouting.org>
2017-09-01 02:13:23 +02:00
nh.type = NEXTHOP_TYPE_BLACKHOLE;
zebra: Use nexthop object id on route delete When we receive a route delete from the kernel and it contains a nexthop object id, use that to match against route gateways with instead of explicit nexthops. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-01 23:36:56 +02:00
nh.bh_type = bh_type;
zebra: allow blackhole routes from the kernel to be deleted Fixes the following bug: % ip -6 route add unreachable 4000::/64 % ip -6 route del unreachable 4000::/64 % vtysh -c "show ipv6 route" [snip] K>* 4000::/64 [0/1024] unreachable (ICMP unreachable) Signed-off-by: Renato Westphal <renato@opensourcerouting.org>
2017-09-01 02:13:23 +02:00
}
zebra: Use nexthop object id on route delete When we receive a route delete from the kernel and it contains a nexthop object id, use that to match against route gateways with instead of explicit nexthops. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-01 23:36:56 +02:00
nh.ifindex = index;
if (gate)
memcpy(&nh.gate, gate, sz);
rib_delete(afi, SAFI_UNICAST, vrf_id, proto, 0,
flags, &p, &src_p, &nh, 0, table,
metric, distance, true);
zebra: allow blackhole routes from the kernel to be deleted Fixes the following bug: % ip -6 route add unreachable 4000::/64 % ip -6 route del unreachable 4000::/64 % vtysh -c "show ipv6 route" [snip] K>* 4000::/64 [0/1024] unreachable (ICMP unreachable) Signed-off-by: Renato Westphal <renato@opensourcerouting.org>
2017-09-01 02:13:23 +02:00
} else {
zebra: Use nexthop object id on route delete When we receive a route delete from the kernel and it contains a nexthop object id, use that to match against route gateways with instead of explicit nexthops. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-01 23:36:56 +02:00
/* XXX: need to compare the entire list of
* nexthops here for NLM_F_APPEND stupidity */
rib_delete(afi, SAFI_UNICAST, vrf_id, proto, 0,
flags, &p, &src_p, NULL, 0, table,
metric, distance, true);
zebra: allow blackhole routes from the kernel to be deleted Fixes the following bug: % ip -6 route add unreachable 4000::/64 % ip -6 route del unreachable 4000::/64 % vtysh -c "show ipv6 route" [snip] K>* 4000::/64 [0/1024] unreachable (ICMP unreachable) Signed-off-by: Renato Westphal <renato@opensourcerouting.org>
2017-09-01 02:13:23 +02:00
}
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
}
}
return 0;
Initial revision
2002-12-13 21:15:29 +01:00
}
lib, pimd, zebra: Allow pimd to ask the kernel about mroute info When we need to lookup the mroute info for a route. Allow pimd to ask the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2016-08-11 02:04:20 +02:00
static struct mcast_route_data *mroute = NULL;
zebra: Remove unnecessary function parameter The snl variable is no longer needed to be passed around, so remove it from the calling path. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-05-23 15:37:06 +02:00
static int netlink_route_change_read_multicast(struct nlmsghdr *h,
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
ns_id_t ns_id, int startup)
zebra: Remove unused variable. Remove the rtnh variable since it is unused. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2016-08-04 01:53:42 +02:00
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
int len;
struct rtmsg *rtm;
struct rtattr *tb[RTA_MAX + 1];
struct mcast_route_data *m;
struct mcast_route_data mr;
int iif = 0;
int count;
int oif[256];
int oif_count = 0;
char sbuf[40];
char gbuf[40];
char oif_list[256] = "\0";
zebra: add a runtime flag to enable vrf with netns The netns backend is chosen by VRF if a runtime flag named vrfwnetns is selected when running zebra. In the case the NETNS backend is chosen, in some case the VRFID value is being assigned the value of the NSID. Within the perimeter of that work, this is why the vrf_lookup_by_table function is extended with a new parameter. Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
2018-01-22 09:42:53 +01:00
vrf_id_t vrf;
zebra: Allow S,G lookup to use RTNL_FAMILY_IPMR This current implementation unfortunately must ask the kernel for all mroutes because vrf's do not have the ability to request a single mroute at this time. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-06-06 16:18:17 +02:00
int table;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (mroute)
m = mroute;
else {
memset(&mr, 0, sizeof(mr));
m = &mr;
}
rtm = NLMSG_DATA(h);
len = h->nlmsg_len - NLMSG_LENGTH(sizeof(struct rtmsg));
memset(tb, 0, sizeof tb);
netlink_parse_rtattr(tb, RTA_MAX, RTM_RTA(rtm), len);
zebra: Fully decode mcast messages Fully decode mcast messages from the kernel. We are not doing anything with this at the moment, but that will change. Additionally convert over to using lookup for displaying the route type. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2016-08-10 14:54:58 +02:00
zebra: Allow S,G lookup to use RTNL_FAMILY_IPMR This current implementation unfortunately must ask the kernel for all mroutes because vrf's do not have the ability to request a single mroute at this time. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-06-06 16:18:17 +02:00
if (tb[RTA_TABLE])
table = *(int *)RTA_DATA(tb[RTA_TABLE]);
else
table = rtm->rtm_table;
zebra: add a runtime flag to enable vrf with netns The netns backend is chosen by VRF if a runtime flag named vrfwnetns is selected when running zebra. In the case the NETNS backend is chosen, in some case the VRFID value is being assigned the value of the NSID. Within the perimeter of that work, this is why the vrf_lookup_by_table function is extended with a new parameter. Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
2018-01-22 09:42:53 +01:00
vrf = vrf_lookup_by_table(table, ns_id);
zebra: Allow S,G lookup to use RTNL_FAMILY_IPMR This current implementation unfortunately must ask the kernel for all mroutes because vrf's do not have the ability to request a single mroute at this time. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-06-06 16:18:17 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (tb[RTA_IIF])
iif = *(int *)RTA_DATA(tb[RTA_IIF]);
if (tb[RTA_SRC])
zebra: Convert get_stats to use new kernel interface Convert the XXX_get_stats functions to use the new kernel interface. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-07-05 17:53:29 +02:00
m->sg.src = *(struct in_addr *)RTA_DATA(tb[RTA_SRC]);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (tb[RTA_DST])
zebra: Convert get_stats to use new kernel interface Convert the XXX_get_stats functions to use the new kernel interface. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-07-05 17:53:29 +02:00
m->sg.grp = *(struct in_addr *)RTA_DATA(tb[RTA_DST]);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
zebra: Cleanup unnecessary check We self include the enum that has the RTA_EXPIRES in it, as such this will always be true. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-01-15 14:55:34 +01:00
if (tb[RTA_EXPIRES])
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
m->lastused = *(unsigned long long *)RTA_DATA(tb[RTA_EXPIRES]);
if (tb[RTA_MULTIPATH]) {
struct rtnexthop *rtnh =
(struct rtnexthop *)RTA_DATA(tb[RTA_MULTIPATH]);
len = RTA_PAYLOAD(tb[RTA_MULTIPATH]);
for (;;) {
if (len < (int)sizeof(*rtnh) || rtnh->rtnh_len > len)
break;
oif[oif_count] = rtnh->rtnh_ifindex;
oif_count++;
zebra: Add check for nexthop loop to prevent hanging Bad nexthop messages from netlink were causing zebra to hang here. Added a check to verify the length of the nexthop so it doesn't keep trying to read. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2018-07-24 00:33:53 +02:00
if (rtnh->rtnh_len == 0)
break;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
len -= NLMSG_ALIGN(rtnh->rtnh_len);
rtnh = RTNH_NEXT(rtnh);
}
}
if (IS_ZEBRA_DEBUG_KERNEL) {
zebra: Add a conditional guard if zvrf lookup fail Add a conditional to guard against segfaulting on the debug statement when zvrf lookup fails. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-07-09 17:10:49 +02:00
struct interface *ifp = NULL;
struct zebra_vrf *zvrf = NULL;
*: fix assorted issues detected by Coverity Scan Signed-off-by: Renato Westphal <renato@opensourcerouting.org>
2017-08-25 02:43:29 +02:00
strlcpy(sbuf, inet_ntoa(m->sg.src), sizeof(sbuf));
strlcpy(gbuf, inet_ntoa(m->sg.grp), sizeof(gbuf));
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
for (count = 0; count < oif_count; count++) {
ifp = if_lookup_by_index(oif[count], vrf);
char temp[256];
zebra: Fix crash in mroute debug There exists a possibility that the ifindex we are passed does not exist and as such we should check for it not resolving as part of the debug. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-08-28 16:35:35 +02:00
sprintf(temp, "%s(%d) ", ifp ? ifp->name : "Unknown",
oif[count]);
zebra: strcat -> strlcat Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2019-05-07 00:55:59 +02:00
strlcat(oif_list, temp, sizeof(oif_list));
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
}
zebra: Add a conditional guard if zvrf lookup fail Add a conditional to guard against segfaulting on the debug statement when zvrf lookup fails. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-07-09 17:10:49 +02:00
zvrf = zebra_vrf_lookup_by_id(vrf);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
ifp = if_lookup_by_index(iif, vrf);
zebra: Add a conditional guard if zvrf lookup fail Add a conditional to guard against segfaulting on the debug statement when zvrf lookup fails. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-07-09 17:10:49 +02:00
zlog_debug(
"MCAST VRF: %s(%d) %s (%s,%s) IIF: %s(%d) OIF: %s jiffies: %lld",
(zvrf ? zvrf->vrf->name : "Unknown"), vrf,
nl_msg_type_to_str(h->nlmsg_type), sbuf, gbuf,
ifp ? ifp->name : "Unknown", iif, oif_list,
m->lastused);
zebra: Fully decode mcast messages Fully decode mcast messages from the kernel. We are not doing anything with this at the moment, but that will change. Additionally convert over to using lookup for displaying the route type. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2016-08-10 14:54:58 +02:00
}
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return 0;
zebra: Remove unused variable. Remove the rtnh variable since it is unused. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2016-08-04 01:53:42 +02:00
}
zebra: Remove unnecessary function parameter The snl variable is no longer needed to be passed around, so remove it from the calling path. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-05-23 15:37:06 +02:00
int netlink_route_change(struct nlmsghdr *h, ns_id_t ns_id, int startup)
zebra: Remove unused variable. Remove the rtnh variable since it is unused. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2016-08-04 01:53:42 +02:00
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
int len;
struct rtmsg *rtm;
rtm = NLMSG_DATA(h);
if (!(h->nlmsg_type == RTM_NEWROUTE || h->nlmsg_type == RTM_DELROUTE)) {
/* If this is not route add/delete message print warning. */
*: remove trailing newlines from zlog messages Zlog puts its own newlines on, and doing this makes logs look nasty. Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2019-03-14 19:41:15 +01:00
zlog_debug("Kernel message: %s NS %u",
zebra: Print out netlink message type instead of number When we are displaying data about a netlink message in debugs or errors, print out the message type as a string instead of a number. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-09-14 23:48:51 +02:00
nl_msg_type_to_str(h->nlmsg_type), ns_id);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return 0;
}
zebra: Newer kernels send data about mroute changes Newer linux kernels apparently send data down the netlink bus for the creation of mroutes. Add a bit of code to notice this and to handle it appropriately( ie do nothing at this point in time ) as that the correct place to do this is in the pim socket in pimd. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-09-14 23:54:11 +02:00
if (!(rtm->rtm_family == AF_INET ||
rtm->rtm_family == AF_INET6 ||
rtm->rtm_family == RTNL_FAMILY_IPMR )) {
zebra: flog_warn conversion Convert Zebra to user error subsystem. Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-08-16 22:10:32 +02:00
flog_warn(
zebra: ZEBRA_[ERR|WARN] -> EC_ZEBRA Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-09-13 21:21:05 +02:00
EC_ZEBRA_UNKNOWN_FAMILY,
zebra: Print out netlink message type instead of number When we are displaying data about a netlink message in debugs or errors, print out the message type as a string instead of a number. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-09-14 23:48:51 +02:00
"Invalid address family: %u received from kernel route change: %s",
rtm->rtm_family, nl_msg_type_to_str(h->nlmsg_type));
zebra: Add address family filters The zebra netlink socket was attempting to read netlink messages with invalid address families in a couple areas. Added filters and warn messages. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2018-07-19 23:29:16 +02:00
return 0;
}
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
/* Connected route. */
if (IS_ZEBRA_DEBUG_KERNEL)
zebra: add a runtime flag to enable vrf with netns The netns backend is chosen by VRF if a runtime flag named vrfwnetns is selected when running zebra. In the case the NETNS backend is chosen, in some case the VRFID value is being assigned the value of the NSID. Within the perimeter of that work, this is why the vrf_lookup_by_table function is extended with a new parameter. Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
2018-01-22 09:42:53 +01:00
zlog_debug("%s %s %s proto %s NS %u",
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
nl_msg_type_to_str(h->nlmsg_type),
nl_family_to_str(rtm->rtm_family),
nl_rttype_to_str(rtm->rtm_type),
zebra: add a runtime flag to enable vrf with netns The netns backend is chosen by VRF if a runtime flag named vrfwnetns is selected when running zebra. In the case the NETNS backend is chosen, in some case the VRFID value is being assigned the value of the NSID. Within the perimeter of that work, this is why the vrf_lookup_by_table function is extended with a new parameter. Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
2018-01-22 09:42:53 +01:00
nl_rtproto_to_str(rtm->rtm_protocol), ns_id);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
len = h->nlmsg_len - NLMSG_LENGTH(sizeof(struct rtmsg));
zebra: Add error messages in places where we have a decode problem When we receive a netlink message from the kernel we have handler functions for when we send a netlink command, if these return a failure ( < 0 ) then we output that we had a parse issue. But if all we get is: 2018-06-21T23:47:45.298156+00:00 qct-ix1-08 zebra[1484]: netlink-cmd (NS 0) filter function error Then it is not very useful to figure out *where* the error happened. Add more error code when in a decode path to hopefully allow us to figure out where this message is coming from. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-06-22 20:22:02 +02:00
if (len < 0) {
zlog_err("%s: Message received from netlink is of a broken size: %d %zu",
__PRETTY_FUNCTION__,
h->nlmsg_len,
(size_t)NLMSG_LENGTH(sizeof(struct rtmsg)));
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return -1;
zebra: Add error messages in places where we have a decode problem When we receive a netlink message from the kernel we have handler functions for when we send a netlink command, if these return a failure ( < 0 ) then we output that we had a parse issue. But if all we get is: 2018-06-21T23:47:45.298156+00:00 qct-ix1-08 zebra[1484]: netlink-cmd (NS 0) filter function error Then it is not very useful to figure out *where* the error happened. Add more error code when in a decode path to hopefully allow us to figure out where this message is coming from. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-06-22 20:22:02 +02:00
}
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
zebra: netlink: get blackhole routes from kernel support processing of RTN_BLACKHOLE et al. from kernel and dump them into appropriate blackhole rib entries. Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2010-02-05 04:58:46 +01:00
if (rtm->rtm_type == RTN_MULTICAST)
zebra: Remove unnecessary function parameter The snl variable is no longer needed to be passed around, so remove it from the calling path. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-05-23 15:37:06 +02:00
netlink_route_change_read_multicast(h, ns_id, startup);
zebra: netlink: get blackhole routes from kernel support processing of RTN_BLACKHOLE et al. from kernel and dump them into appropriate blackhole rib entries. Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2010-02-05 04:58:46 +01:00
else
zebra: Remove unnecessary function parameter The snl variable is no longer needed to be passed around, so remove it from the calling path. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-05-23 15:37:06 +02:00
netlink_route_change_read_unicast(h, ns_id, startup);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return 0;
zebra: Remove unused variable. Remove the rtnh variable since it is unused. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2016-08-04 01:53:42 +02:00
}
zebra: Format netlink requests correctly When zebra issues read (GET) requests to the kernel using the netlink interface, it is incorrect to format all of them in a generic manner using 'struct ifinfomsg' or 'struct rtgenmsg'. Rather, messages for a particular entity (e.g., routes) should use the corresponding structure for encoding (e.g., 'struct rtmsg'). Of course, this has to correlate with what the kernel expects. In the absence of this, there is the possibility of sending extraneous information in the request which the kernel wouldn't like. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: David Ahern <dsa@cumulusnetworks.com>
2017-05-15 07:28:32 +02:00
/* Request for specific route information from the kernel */
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
static int netlink_request_route(struct zebra_ns *zns, int family, int type)
zebra: Format netlink requests correctly When zebra issues read (GET) requests to the kernel using the netlink interface, it is incorrect to format all of them in a generic manner using 'struct ifinfomsg' or 'struct rtgenmsg'. Rather, messages for a particular entity (e.g., routes) should use the corresponding structure for encoding (e.g., 'struct rtmsg'). Of course, this has to correlate with what the kernel expects. In the absence of this, there is the possibility of sending extraneous information in the request which the kernel wouldn't like. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: David Ahern <dsa@cumulusnetworks.com>
2017-05-15 07:28:32 +02:00
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
struct {
struct nlmsghdr n;
struct rtmsg rtm;
} req;
/* Form the request, specifying filter (rtattr) if needed. */
memset(&req, 0, sizeof(req));
req.n.nlmsg_type = type;
zebra: netlink_request consolidataion Make netlink_request api generic where it can be used for dump or querying specific information request. nelink request nlm flags (NLM_F_ROOT | NLM_F_MATCH) are used to dump purpose, if client wants to query spcific MAC or IP using netlink_request does not require to set them. nlm struct is passed by the caller of netlink_request, it can also set the nlm request flags. Signed-off-by: Chirag Shah <chirag@cumulusnetworks.com>
2018-12-11 06:33:16 +01:00
req.n.nlmsg_flags = NLM_F_ROOT | NLM_F_MATCH | NLM_F_REQUEST;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
req.n.nlmsg_len = NLMSG_LENGTH(sizeof(struct rtmsg));
req.rtm.rtm_family = family;
return netlink_request(&zns->netlink_cmd, &req.n);
zebra: Format netlink requests correctly When zebra issues read (GET) requests to the kernel using the netlink interface, it is incorrect to format all of them in a generic manner using 'struct ifinfomsg' or 'struct rtgenmsg'. Rather, messages for a particular entity (e.g., routes) should use the corresponding structure for encoding (e.g., 'struct rtmsg'). Of course, this has to correlate with what the kernel expects. In the absence of this, there is the possibility of sending extraneous information in the request which the kernel wouldn't like. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: David Ahern <dsa@cumulusnetworks.com>
2017-05-15 07:28:32 +02:00
}
Initial revision
2002-12-13 21:15:29 +01:00
/* Routing table read function using netlink interface. Only called
bootstrap time. */
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
int netlink_route_read(struct zebra_ns *zns)
Initial revision
2002-12-13 21:15:29 +01:00
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
int ret;
zebra: Start abstraction of zebra_dplane_info for context passing Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Signed-off-by: Mark Stapp <mjs@voltanet.io> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-09-12 20:59:57 +02:00
struct zebra_dplane_info dp_info;
zebra_dplane_info_from_zns(&dp_info, zns, true /*is_cmd*/);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
/* Get IPv4 routing table. */
ret = netlink_request_route(zns, AF_INET, RTM_GETROUTE);
if (ret < 0)
return ret;
ret = netlink_parse_info(netlink_route_change_read_unicast,
zebra: Start abstraction of zebra_dplane_info for context passing Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Signed-off-by: Mark Stapp <mjs@voltanet.io> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-09-12 20:59:57 +02:00
&zns->netlink_cmd, &dp_info, 0, 1);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (ret < 0)
return ret;
/* Get IPv6 routing table. */
ret = netlink_request_route(zns, AF_INET6, RTM_GETROUTE);
if (ret < 0)
return ret;
ret = netlink_parse_info(netlink_route_change_read_unicast,
zebra: Start abstraction of zebra_dplane_info for context passing Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Signed-off-by: Mark Stapp <mjs@voltanet.io> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-09-12 20:59:57 +02:00
&zns->netlink_cmd, &dp_info, 0, 1);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (ret < 0)
return ret;
return 0;
Initial revision
2002-12-13 21:15:29 +01:00
}
*: use C99 standard fixed-width integer types The following types are nonstandard: - u_char - u_short - u_int - u_long - u_int8_t - u_int16_t - u_int32_t Replace them with the C99 standard types: - uint8_t - unsigned short - unsigned int - unsigned long - uint8_t - uint16_t - uint32_t Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-03-27 21:13:34 +02:00
static void _netlink_route_nl_add_gateway_info(uint8_t route_family,
uint8_t gw_family,
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
struct nlmsghdr *nlmsg,
size_t req_size, int bytelen,
zebra: use const in dplane pw nhlfe accessors Use const in the accessors for pseudowire nhlfe data; pull that through the kernel-facing apis that use that data. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-03-07 19:09:51 +01:00
const struct nexthop *nexthop)
Quagga: Install label forwarding entries for statically configured LSPs Install the statically configured LSPs into the FIB (kernel). This is done using the new attributes and definitions for MPLS in the kernel - RTA_VIA, RTA_NEWDST and AF_MPLS. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com> Ticket: CM-4804 Reviewed By: CCR-3088 Testing Done: Manual in SE-1
2016-04-16 04:19:37 +02:00
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (route_family == AF_MPLS) {
struct gw_family_t gw_fam;
gw_fam.family = gw_family;
if (gw_family == AF_INET)
memcpy(&gw_fam.gate.ipv4, &nexthop->gate.ipv4, bytelen);
else
memcpy(&gw_fam.gate.ipv6, &nexthop->gate.ipv6, bytelen);
addattr_l(nlmsg, req_size, RTA_VIA, &gw_fam.family,
bytelen + 2);
} else {
if (gw_family == AF_INET)
addattr_l(nlmsg, req_size, RTA_GATEWAY,
&nexthop->gate.ipv4, bytelen);
else
addattr_l(nlmsg, req_size, RTA_GATEWAY,
&nexthop->gate.ipv6, bytelen);
}
Quagga: Install label forwarding entries for statically configured LSPs Install the statically configured LSPs into the FIB (kernel). This is done using the new attributes and definitions for MPLS in the kernel - RTA_VIA, RTA_NEWDST and AF_MPLS. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com> Ticket: CM-4804 Reviewed By: CCR-3088 Testing Done: Manual in SE-1
2016-04-16 04:19:37 +02:00
}
*: use C99 standard fixed-width integer types The following types are nonstandard: - u_char - u_short - u_int - u_long - u_int8_t - u_int16_t - u_int32_t Replace them with the C99 standard types: - uint8_t - unsigned short - unsigned int - unsigned long - uint8_t - uint16_t - uint32_t Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-03-27 21:13:34 +02:00
static void _netlink_route_rta_add_gateway_info(uint8_t route_family,
uint8_t gw_family,
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
struct rtattr *rta,
struct rtnexthop *rtnh,
size_t req_size, int bytelen,
zebra: use const in dplane pw nhlfe accessors Use const in the accessors for pseudowire nhlfe data; pull that through the kernel-facing apis that use that data. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-03-07 19:09:51 +01:00
const struct nexthop *nexthop)
Quagga: Install label forwarding entries for statically configured LSPs Install the statically configured LSPs into the FIB (kernel). This is done using the new attributes and definitions for MPLS in the kernel - RTA_VIA, RTA_NEWDST and AF_MPLS. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com> Ticket: CM-4804 Reviewed By: CCR-3088 Testing Done: Manual in SE-1
2016-04-16 04:19:37 +02:00
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (route_family == AF_MPLS) {
struct gw_family_t gw_fam;
gw_fam.family = gw_family;
if (gw_family == AF_INET)
memcpy(&gw_fam.gate.ipv4, &nexthop->gate.ipv4, bytelen);
else
memcpy(&gw_fam.gate.ipv6, &nexthop->gate.ipv6, bytelen);
rta_addattr_l(rta, req_size, RTA_VIA, &gw_fam.family,
bytelen + 2);
rtnh->rtnh_len += RTA_LENGTH(bytelen + 2);
} else {
if (gw_family == AF_INET)
rta_addattr_l(rta, req_size, RTA_GATEWAY,
&nexthop->gate.ipv4, bytelen);
else
rta_addattr_l(rta, req_size, RTA_GATEWAY,
&nexthop->gate.ipv6, bytelen);
rtnh->rtnh_len += sizeof(struct rtattr) + bytelen;
}
Quagga: Install label forwarding entries for statically configured LSPs Install the statically configured LSPs into the FIB (kernel). This is done using the new attributes and definitions for MPLS in the kernel - RTA_VIA, RTA_NEWDST and AF_MPLS. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com> Ticket: CM-4804 Reviewed By: CCR-3088 Testing Done: Manual in SE-1
2016-04-16 04:19:37 +02:00
}
zebra: Add common netlink mpls stack building path There was some code copypasta for mpls stack building in the netlink install path. Reduced that to a common function. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-12 18:35:58 +02:00
static int build_label_stack(struct mpls_label_stack *nh_label,
mpls_lse_t *out_lse, char *label_buf,
size_t label_buf_size)
{
char label_buf1[20];
int num_labels = 0;
for (int i = 0; nh_label && i < nh_label->num_labels; i++) {
if (nh_label->label[i] == MPLS_LABEL_IMPLICIT_NULL)
continue;
if (IS_ZEBRA_DEBUG_KERNEL) {
if (!num_labels)
sprintf(label_buf, "label %u",
nh_label->label[i]);
else {
sprintf(label_buf1, "/%u", nh_label->label[i]);
strlcat(label_buf, label_buf1, label_buf_size);
}
}
out_lse[num_labels] =
mpls_lse_encode(nh_label->label[i], 0, 0, 0);
num_labels++;
}
return num_labels;
}
zebra: rework recursive route resolution Change the datastructure for recursive routes. This brings the following benefits: By using struct nexthop also to store nexthops obtained by recursive resolution, we can get rid of quite a bit of code duplication in the fib management. (rt_netlink, rt_socket, ...) With the new datastructure we can make use of all available paths when recursive routes are resolved with multipath routes. Signed-off-by: Christian Franke <chris@opensourcerouting.org> Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2013-07-05 17:35:37 +02:00
/* This function takes a nexthop as argument and adds
* the appropriate netlink attributes to an existing
* netlink message.
*
* @param routedesc: Human readable description of route type
* (direct/recursive, single-/multipath)
* @param bytelen: Length of addresses in bytes.
* @param nexthop: Nexthop information
* @param nlmsg: nlmsghdr structure to fill in.
* @param req_size: The size allocated for the message.
*/
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
static void _netlink_route_build_singlepath(const char *routedesc, int bytelen,
zebra: use const in dplane pw nhlfe accessors Use const in the accessors for pseudowire nhlfe data; pull that through the kernel-facing apis that use that data. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-03-07 19:09:51 +01:00
const struct nexthop *nexthop,
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
struct nlmsghdr *nlmsg,
struct rtmsg *rtmsg,
size_t req_size, int cmd)
zebra: rework recursive route resolution Change the datastructure for recursive routes. This brings the following benefits: By using struct nexthop also to store nexthops obtained by recursive resolution, we can get rid of quite a bit of code duplication in the fib management. (rt_netlink, rt_socket, ...) With the new datastructure we can make use of all available paths when recursive routes are resolved with multipath routes. Signed-off-by: Christian Franke <chris@opensourcerouting.org> Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2013-07-05 17:35:37 +02:00
{
zebra: Add common netlink mpls stack building path There was some code copypasta for mpls stack building in the netlink install path. Reduced that to a common function. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-12 18:35:58 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
mpls_lse_t out_lse[MPLS_MAX_LABELS];
zebra: fix logging of MPLS labels * use %u instead of %d, we don't want to print negative labels; * increase the size of label_buf to accommodate the worst case scenarios; * use strlcat() instead of strcat() as a security best practice. Signed-off-by: Renato Westphal <renato@opensourcerouting.org>
2017-09-20 05:02:01 +02:00
char label_buf[256];
zebra: Add common netlink mpls stack building path There was some code copypasta for mpls stack building in the netlink install path. Reduced that to a common function. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-12 18:35:58 +02:00
int num_labels = 0;
assert(nexthop);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
/*
* label_buf is *only* currently used within debugging.
* As such when we assign it we are guarding it inside
* a debug test. If you want to change this make sure
* you fix this assumption
*/
label_buf[0] = '\0';
zebra: Add common netlink mpls stack building path There was some code copypasta for mpls stack building in the netlink install path. Reduced that to a common function. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-12 18:35:58 +02:00
num_labels = build_label_stack(nexthop->nh_label, out_lse, label_buf,
sizeof(label_buf));
zebra: implement recursive MPLS labels When a BGP-labeled route is resolved into an LDP-labeled IGP route, zebra would install it with no labels in the kernel. This patch implements recursive MPLS labels, i.e. make zebra install all labels from the route's nexthop chain (the labels from the top-level nexthop being installed in the top of the MPLS label stack). Multiple recursion levels are supported. Signed-off-by: Renato Westphal <renato@opensourcerouting.org>
2017-09-20 05:05:25 +02:00
if (num_labels) {
/* Set the BoS bit */
out_lse[num_labels - 1] |= htonl(1 << MPLS_LS_S_SHIFT);
if (rtmsg->rtm_family == AF_MPLS)
addattr_l(nlmsg, req_size, RTA_NEWDST, &out_lse,
num_labels * sizeof(mpls_lse_t));
else {
struct rtattr *nest;
*: use C99 standard fixed-width integer types The following types are nonstandard: - u_char - u_short - u_int - u_long - u_int8_t - u_int16_t - u_int32_t Replace them with the C99 standard types: - uint8_t - unsigned short - unsigned int - unsigned long - uint8_t - uint16_t - uint32_t Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-03-27 21:13:34 +02:00
uint16_t encap = LWTUNNEL_ENCAP_MPLS;
zebra: implement recursive MPLS labels When a BGP-labeled route is resolved into an LDP-labeled IGP route, zebra would install it with no labels in the kernel. This patch implements recursive MPLS labels, i.e. make zebra install all labels from the route's nexthop chain (the labels from the top-level nexthop being installed in the top of the MPLS label stack). Multiple recursion levels are supported. Signed-off-by: Renato Westphal <renato@opensourcerouting.org>
2017-09-20 05:05:25 +02:00
addattr_l(nlmsg, req_size, RTA_ENCAP_TYPE, &encap,
*: use C99 standard fixed-width integer types The following types are nonstandard: - u_char - u_short - u_int - u_long - u_int8_t - u_int16_t - u_int32_t Replace them with the C99 standard types: - uint8_t - unsigned short - unsigned int - unsigned long - uint8_t - uint16_t - uint32_t Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-03-27 21:13:34 +02:00
sizeof(uint16_t));
zebra: implement recursive MPLS labels When a BGP-labeled route is resolved into an LDP-labeled IGP route, zebra would install it with no labels in the kernel. This patch implements recursive MPLS labels, i.e. make zebra install all labels from the route's nexthop chain (the labels from the top-level nexthop being installed in the top of the MPLS label stack). Multiple recursion levels are supported. Signed-off-by: Renato Westphal <renato@opensourcerouting.org>
2017-09-20 05:05:25 +02:00
nest = addattr_nest(nlmsg, req_size, RTA_ENCAP);
addattr_l(nlmsg, req_size, MPLS_IPTUNNEL_DST, &out_lse,
num_labels * sizeof(mpls_lse_t));
addattr_nest_end(nlmsg, nest);
zebra: Remove string operations from non-debug path Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-26 03:22:03 +02:00
}
Zebra: Add IPv6 protocol filtering support & Setting Src of IPv6 routes Ticket: Reviewed By: CCR-3335 Testing Done: bgpsmoke, ENHE tests etc. Add support for filtering routes from upper layer protocols to zebra via route-maps for IPv6. The same functionality already existed for IPv4. In addition, add support for setting source of routes via IPv6 protocol map. Signed-off-by: Dinesh G Dutt <ddutt@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com> Reviewed-by: Vivek Venkataraman <vivek@cumulusnetworks.com> Reviewed-by: Vipin Kumar <vipin@cumulusnetworks.com>
2015-09-16 08:48:00 +02:00
}
zebra: rework recursive route resolution Change the datastructure for recursive routes. This brings the following benefits: By using struct nexthop also to store nexthops obtained by recursive resolution, we can get rid of quite a bit of code duplication in the fib management. (rt_netlink, rt_socket, ...) With the new datastructure we can make use of all available paths when recursive routes are resolved with multipath routes. Signed-off-by: Christian Franke <chris@opensourcerouting.org> Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2013-07-05 17:35:37 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (CHECK_FLAG(nexthop->flags, NEXTHOP_FLAG_ONLINK))
rtmsg->rtm_flags |= RTNH_F_ONLINK;
if (rtmsg->rtm_family == AF_INET
&& (nexthop->type == NEXTHOP_TYPE_IPV6
|| nexthop->type == NEXTHOP_TYPE_IPV6_IFINDEX)) {
rtmsg->rtm_flags |= RTNH_F_ONLINK;
addattr_l(nlmsg, req_size, RTA_GATEWAY, &ipv4_ll, 4);
addattr32(nlmsg, req_size, RTA_OIF, nexthop->ifindex);
if (nexthop->rmap_src.ipv4.s_addr && (cmd == RTM_NEWROUTE))
addattr_l(nlmsg, req_size, RTA_PREFSRC,
&nexthop->rmap_src.ipv4, bytelen);
else if (nexthop->src.ipv4.s_addr && (cmd == RTM_NEWROUTE))
addattr_l(nlmsg, req_size, RTA_PREFSRC,
&nexthop->src.ipv4, bytelen);
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug(
" 5549: _netlink_route_build_singlepath() (%s): "
zebra: Add table and nexthop vrf information to netlink debug Add some additional debug information to the netlink debug messages so we can see the table we are installing to as well as the nexthop's vrf. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-03-29 14:58:45 +02:00
"nexthop via %s %s if %u(%u)",
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
routedesc, ipv4_ll_buf, label_buf,
zebra: Add table and nexthop vrf information to netlink debug Add some additional debug information to the netlink debug messages so we can see the table we are installing to as well as the nexthop's vrf. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-03-29 14:58:45 +02:00
nexthop->ifindex, nexthop->vrf_id);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return;
Zebra: Add IPv6 protocol filtering support & Setting Src of IPv6 routes Ticket: Reviewed By: CCR-3335 Testing Done: bgpsmoke, ENHE tests etc. Add support for filtering routes from upper layer protocols to zebra via route-maps for IPv6. The same functionality already existed for IPv4. In addition, add support for setting source of routes via IPv6 protocol map. Signed-off-by: Dinesh G Dutt <ddutt@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com> Reviewed-by: Vivek Venkataraman <vivek@cumulusnetworks.com> Reviewed-by: Vipin Kumar <vipin@cumulusnetworks.com>
2015-09-16 08:48:00 +02:00
}
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (nexthop->type == NEXTHOP_TYPE_IPV4
|| nexthop->type == NEXTHOP_TYPE_IPV4_IFINDEX) {
/* Send deletes to the kernel without specifying the next-hop */
if (cmd != RTM_DELROUTE)
_netlink_route_nl_add_gateway_info(
rtmsg->rtm_family, AF_INET, nlmsg, req_size,
bytelen, nexthop);
if (cmd == RTM_NEWROUTE) {
if (nexthop->rmap_src.ipv4.s_addr)
addattr_l(nlmsg, req_size, RTA_PREFSRC,
&nexthop->rmap_src.ipv4, bytelen);
else if (nexthop->src.ipv4.s_addr)
addattr_l(nlmsg, req_size, RTA_PREFSRC,
&nexthop->src.ipv4, bytelen);
}
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug(
"netlink_route_multipath() (%s): "
zebra: Add table and nexthop vrf information to netlink debug Add some additional debug information to the netlink debug messages so we can see the table we are installing to as well as the nexthop's vrf. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-03-29 14:58:45 +02:00
"nexthop via %s %s if %u(%u)",
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
routedesc, inet_ntoa(nexthop->gate.ipv4),
zebra: Add table and nexthop vrf information to netlink debug Add some additional debug information to the netlink debug messages so we can see the table we are installing to as well as the nexthop's vrf. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-03-29 14:58:45 +02:00
label_buf, nexthop->ifindex, nexthop->vrf_id);
Zebra: Add IPv6 protocol filtering support & Setting Src of IPv6 routes Ticket: Reviewed By: CCR-3335 Testing Done: bgpsmoke, ENHE tests etc. Add support for filtering routes from upper layer protocols to zebra via route-maps for IPv6. The same functionality already existed for IPv4. In addition, add support for setting source of routes via IPv6 protocol map. Signed-off-by: Dinesh G Dutt <ddutt@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com> Reviewed-by: Vivek Venkataraman <vivek@cumulusnetworks.com> Reviewed-by: Vipin Kumar <vipin@cumulusnetworks.com>
2015-09-16 08:48:00 +02:00
}
zebra: rework recursive route resolution Change the datastructure for recursive routes. This brings the following benefits: By using struct nexthop also to store nexthops obtained by recursive resolution, we can get rid of quite a bit of code duplication in the fib management. (rt_netlink, rt_socket, ...) With the new datastructure we can make use of all available paths when recursive routes are resolved with multipath routes. Signed-off-by: Christian Franke <chris@opensourcerouting.org> Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2013-07-05 17:35:37 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (nexthop->type == NEXTHOP_TYPE_IPV6
|| nexthop->type == NEXTHOP_TYPE_IPV6_IFINDEX) {
_netlink_route_nl_add_gateway_info(rtmsg->rtm_family, AF_INET6,
nlmsg, req_size, bytelen,
nexthop);
if (cmd == RTM_NEWROUTE) {
if (!IN6_IS_ADDR_UNSPECIFIED(&nexthop->rmap_src.ipv6))
addattr_l(nlmsg, req_size, RTA_PREFSRC,
&nexthop->rmap_src.ipv6, bytelen);
else if (!IN6_IS_ADDR_UNSPECIFIED(&nexthop->src.ipv6))
addattr_l(nlmsg, req_size, RTA_PREFSRC,
&nexthop->src.ipv6, bytelen);
}
zebra: rework recursive route resolution Change the datastructure for recursive routes. This brings the following benefits: By using struct nexthop also to store nexthops obtained by recursive resolution, we can get rid of quite a bit of code duplication in the fib management. (rt_netlink, rt_socket, ...) With the new datastructure we can make use of all available paths when recursive routes are resolved with multipath routes. Signed-off-by: Christian Franke <chris@opensourcerouting.org> Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2013-07-05 17:35:37 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug(
"netlink_route_multipath() (%s): "
zebra: Add table and nexthop vrf information to netlink debug Add some additional debug information to the netlink debug messages so we can see the table we are installing to as well as the nexthop's vrf. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-03-29 14:58:45 +02:00
"nexthop via %s %s if %u(%u)",
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
routedesc, inet6_ntoa(nexthop->gate.ipv6),
zebra: Add table and nexthop vrf information to netlink debug Add some additional debug information to the netlink debug messages so we can see the table we are installing to as well as the nexthop's vrf. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-03-29 14:58:45 +02:00
label_buf, nexthop->ifindex, nexthop->vrf_id);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
}
zebra: Encode the ifindex over netlink In order for routes to be leaked the ifindex must be sent down into the kernel over the netlink protocol. So send it( we always figure it out ) when we add the route. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-01-11 17:47:04 +01:00
/*
* We have the ifindex so we should always send it
* This is especially useful if we are doing route
* leaking.
*/
if (nexthop->type != NEXTHOP_TYPE_BLACKHOLE)
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
addattr32(nlmsg, req_size, RTA_OIF, nexthop->ifindex);
zebra: some v4 attributes were being written 2 times When the nexthop->type is NEXTHOP_TYPE_IPV4_IFINDEX we were writing the RTA_PREFSRC 2 times for the build_singlepath and build_multipath functions. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-01-30 02:17:36 +01:00
if (nexthop->type == NEXTHOP_TYPE_IFINDEX) {
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (cmd == RTM_NEWROUTE) {
if (nexthop->rmap_src.ipv4.s_addr)
addattr_l(nlmsg, req_size, RTA_PREFSRC,
&nexthop->rmap_src.ipv4, bytelen);
else if (nexthop->src.ipv4.s_addr)
addattr_l(nlmsg, req_size, RTA_PREFSRC,
&nexthop->src.ipv4, bytelen);
}
zebra: rework recursive route resolution Change the datastructure for recursive routes. This brings the following benefits: By using struct nexthop also to store nexthops obtained by recursive resolution, we can get rid of quite a bit of code duplication in the fib management. (rt_netlink, rt_socket, ...) With the new datastructure we can make use of all available paths when recursive routes are resolved with multipath routes. Signed-off-by: Christian Franke <chris@opensourcerouting.org> Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2013-07-05 17:35:37 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug(
"netlink_route_multipath() (%s): "
zebra: Add table and nexthop vrf information to netlink debug Add some additional debug information to the netlink debug messages so we can see the table we are installing to as well as the nexthop's vrf. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-03-29 14:58:45 +02:00
"nexthop via if %u(%u)",
routedesc, nexthop->ifindex, nexthop->vrf_id);
Zebra: Add IPv6 protocol filtering support & Setting Src of IPv6 routes Ticket: Reviewed By: CCR-3335 Testing Done: bgpsmoke, ENHE tests etc. Add support for filtering routes from upper layer protocols to zebra via route-maps for IPv6. The same functionality already existed for IPv4. In addition, add support for setting source of routes via IPv6 protocol map. Signed-off-by: Dinesh G Dutt <ddutt@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com> Reviewed-by: Vivek Venkataraman <vivek@cumulusnetworks.com> Reviewed-by: Vipin Kumar <vipin@cumulusnetworks.com>
2015-09-16 08:48:00 +02:00
}
zebra: rework recursive route resolution Change the datastructure for recursive routes. This brings the following benefits: By using struct nexthop also to store nexthops obtained by recursive resolution, we can get rid of quite a bit of code duplication in the fib management. (rt_netlink, rt_socket, ...) With the new datastructure we can make use of all available paths when recursive routes are resolved with multipath routes. Signed-off-by: Christian Franke <chris@opensourcerouting.org> Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2013-07-05 17:35:37 +02:00
}
/* This function takes a nexthop as argument and
* appends to the given rtattr/rtnexthop pair the
* representation of the nexthop. If the nexthop
* defines a preferred source, the src parameter
* will be modified to point to that src, otherwise
* it will be kept unmodified.
*
* @param routedesc: Human readable description of route type
* (direct/recursive, single-/multipath)
* @param bytelen: Length of addresses in bytes.
* @param nexthop: Nexthop information
* @param rta: rtnetlink attribute structure
* @param rtnh: pointer to an rtnetlink nexthop structure
* @param src: pointer pointing to a location where
* the prefsrc should be stored.
*/
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
static void _netlink_route_build_multipath(const char *routedesc, int bytelen,
zebra: use const in dplane pw nhlfe accessors Use const in the accessors for pseudowire nhlfe data; pull that through the kernel-facing apis that use that data. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-03-07 19:09:51 +01:00
const struct nexthop *nexthop,
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
struct rtattr *rta,
struct rtnexthop *rtnh,
struct rtmsg *rtmsg,
zebra: use const in dplane pw nhlfe accessors Use const in the accessors for pseudowire nhlfe data; pull that through the kernel-facing apis that use that data. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-03-07 19:09:51 +01:00
const union g_addr **src)
zebra: rework recursive route resolution Change the datastructure for recursive routes. This brings the following benefits: By using struct nexthop also to store nexthops obtained by recursive resolution, we can get rid of quite a bit of code duplication in the fib management. (rt_netlink, rt_socket, ...) With the new datastructure we can make use of all available paths when recursive routes are resolved with multipath routes. Signed-off-by: Christian Franke <chris@opensourcerouting.org> Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2013-07-05 17:35:37 +02:00
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
mpls_lse_t out_lse[MPLS_MAX_LABELS];
zebra: fix logging of MPLS labels * use %u instead of %d, we don't want to print negative labels; * increase the size of label_buf to accommodate the worst case scenarios; * use strlcat() instead of strcat() as a security best practice. Signed-off-by: Renato Westphal <renato@opensourcerouting.org>
2017-09-20 05:02:01 +02:00
char label_buf[256];
zebra: Add common netlink mpls stack building path There was some code copypasta for mpls stack building in the netlink install path. Reduced that to a common function. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-12 18:35:58 +02:00
int num_labels = 0;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
rtnh->rtnh_len = sizeof(*rtnh);
rtnh->rtnh_flags = 0;
rtnh->rtnh_hops = 0;
rta->rta_len += rtnh->rtnh_len;
zebra: Add common netlink mpls stack building path There was some code copypasta for mpls stack building in the netlink install path. Reduced that to a common function. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-12 18:35:58 +02:00
assert(nexthop);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
/*
* label_buf is *only* currently used within debugging.
* As such when we assign it we are guarding it inside
* a debug test. If you want to change this make sure
* you fix this assumption
*/
label_buf[0] = '\0';
zebra: Add common netlink mpls stack building path There was some code copypasta for mpls stack building in the netlink install path. Reduced that to a common function. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-12 18:35:58 +02:00
num_labels = build_label_stack(nexthop->nh_label, out_lse, label_buf,
sizeof(label_buf));
zebra: implement recursive MPLS labels When a BGP-labeled route is resolved into an LDP-labeled IGP route, zebra would install it with no labels in the kernel. This patch implements recursive MPLS labels, i.e. make zebra install all labels from the route's nexthop chain (the labels from the top-level nexthop being installed in the top of the MPLS label stack). Multiple recursion levels are supported. Signed-off-by: Renato Westphal <renato@opensourcerouting.org>
2017-09-20 05:05:25 +02:00
if (num_labels) {
/* Set the BoS bit */
out_lse[num_labels - 1] |= htonl(1 << MPLS_LS_S_SHIFT);
if (rtmsg->rtm_family == AF_MPLS) {
rta_addattr_l(rta, NL_PKT_BUF_SIZE, RTA_NEWDST,
&out_lse,
num_labels * sizeof(mpls_lse_t));
rtnh->rtnh_len +=
RTA_LENGTH(num_labels * sizeof(mpls_lse_t));
} else {
struct rtattr *nest;
*: use C99 standard fixed-width integer types The following types are nonstandard: - u_char - u_short - u_int - u_long - u_int8_t - u_int16_t - u_int32_t Replace them with the C99 standard types: - uint8_t - unsigned short - unsigned int - unsigned long - uint8_t - uint16_t - uint32_t Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-03-27 21:13:34 +02:00
uint16_t encap = LWTUNNEL_ENCAP_MPLS;
zebra: implement recursive MPLS labels When a BGP-labeled route is resolved into an LDP-labeled IGP route, zebra would install it with no labels in the kernel. This patch implements recursive MPLS labels, i.e. make zebra install all labels from the route's nexthop chain (the labels from the top-level nexthop being installed in the top of the MPLS label stack). Multiple recursion levels are supported. Signed-off-by: Renato Westphal <renato@opensourcerouting.org>
2017-09-20 05:05:25 +02:00
int len = rta->rta_len;
rta_addattr_l(rta, NL_PKT_BUF_SIZE, RTA_ENCAP_TYPE,
*: use C99 standard fixed-width integer types The following types are nonstandard: - u_char - u_short - u_int - u_long - u_int8_t - u_int16_t - u_int32_t Replace them with the C99 standard types: - uint8_t - unsigned short - unsigned int - unsigned long - uint8_t - uint16_t - uint32_t Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-03-27 21:13:34 +02:00
&encap, sizeof(uint16_t));
zebra: implement recursive MPLS labels When a BGP-labeled route is resolved into an LDP-labeled IGP route, zebra would install it with no labels in the kernel. This patch implements recursive MPLS labels, i.e. make zebra install all labels from the route's nexthop chain (the labels from the top-level nexthop being installed in the top of the MPLS label stack). Multiple recursion levels are supported. Signed-off-by: Renato Westphal <renato@opensourcerouting.org>
2017-09-20 05:05:25 +02:00
nest = rta_nest(rta, NL_PKT_BUF_SIZE, RTA_ENCAP);
rta_addattr_l(rta, NL_PKT_BUF_SIZE, MPLS_IPTUNNEL_DST,
&out_lse,
num_labels * sizeof(mpls_lse_t));
rta_nest_end(rta, nest);
rtnh->rtnh_len += rta->rta_len - len;
zebra: Remove string operations from non-debug path Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-26 03:22:03 +02:00
}
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
}
if (CHECK_FLAG(nexthop->flags, NEXTHOP_FLAG_ONLINK))
rtnh->rtnh_flags |= RTNH_F_ONLINK;
if (rtmsg->rtm_family == AF_INET
&& (nexthop->type == NEXTHOP_TYPE_IPV6
|| nexthop->type == NEXTHOP_TYPE_IPV6_IFINDEX)) {
bytelen = 4;
rtnh->rtnh_flags |= RTNH_F_ONLINK;
rta_addattr_l(rta, NL_PKT_BUF_SIZE, RTA_GATEWAY, &ipv4_ll,
bytelen);
rtnh->rtnh_len += sizeof(struct rtattr) + bytelen;
rtnh->rtnh_ifindex = nexthop->ifindex;
if (nexthop->rmap_src.ipv4.s_addr)
*src = &nexthop->rmap_src;
else if (nexthop->src.ipv4.s_addr)
*src = &nexthop->src;
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug(
" 5549: netlink_route_build_multipath() (%s): "
"nexthop via %s %s if %u",
routedesc, ipv4_ll_buf, label_buf,
nexthop->ifindex);
return;
}
if (nexthop->type == NEXTHOP_TYPE_IPV4
|| nexthop->type == NEXTHOP_TYPE_IPV4_IFINDEX) {
_netlink_route_rta_add_gateway_info(rtmsg->rtm_family, AF_INET,
rta, rtnh, NL_PKT_BUF_SIZE,
bytelen, nexthop);
if (nexthop->rmap_src.ipv4.s_addr)
*src = &nexthop->rmap_src;
else if (nexthop->src.ipv4.s_addr)
*src = &nexthop->src;
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug(
"netlink_route_multipath() (%s): "
"nexthop via %s %s if %u",
routedesc, inet_ntoa(nexthop->gate.ipv4),
label_buf, nexthop->ifindex);
}
if (nexthop->type == NEXTHOP_TYPE_IPV6
|| nexthop->type == NEXTHOP_TYPE_IPV6_IFINDEX) {
_netlink_route_rta_add_gateway_info(rtmsg->rtm_family, AF_INET6,
rta, rtnh, NL_PKT_BUF_SIZE,
bytelen, nexthop);
if (!IN6_IS_ADDR_UNSPECIFIED(&nexthop->rmap_src.ipv6))
*src = &nexthop->rmap_src;
else if (!IN6_IS_ADDR_UNSPECIFIED(&nexthop->src.ipv6))
*src = &nexthop->src;
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug(
"netlink_route_multipath() (%s): "
"nexthop via %s %s if %u",
routedesc, inet6_ntoa(nexthop->gate.ipv6),
label_buf, nexthop->ifindex);
}
zebra: Encode the ifindex over netlink In order for routes to be leaked the ifindex must be sent down into the kernel over the netlink protocol. So send it( we always figure it out ) when we add the route. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-01-11 17:47:04 +01:00
/*
* We have figured out the ifindex so we should always send it
* This is especially useful if we are doing route
* leaking.
*/
if (nexthop->type != NEXTHOP_TYPE_BLACKHOLE)
rtnh->rtnh_ifindex = nexthop->ifindex;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
/* ifindex */
zebra: some v4 attributes were being written 2 times When the nexthop->type is NEXTHOP_TYPE_IPV4_IFINDEX we were writing the RTA_PREFSRC 2 times for the build_singlepath and build_multipath functions. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-01-30 02:17:36 +01:00
if (nexthop->type == NEXTHOP_TYPE_IFINDEX) {
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (nexthop->rmap_src.ipv4.s_addr)
*src = &nexthop->rmap_src;
else if (nexthop->src.ipv4.s_addr)
*src = &nexthop->src;
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug(
"netlink_route_multipath() (%s): "
"nexthop via if %u",
routedesc, nexthop->ifindex);
}
zebra: rework recursive route resolution Change the datastructure for recursive routes. This brings the following benefits: By using struct nexthop also to store nexthops obtained by recursive resolution, we can get rid of quite a bit of code duplication in the fib management. (rt_netlink, rt_socket, ...) With the new datastructure we can make use of all available paths when recursive routes are resolved with multipath routes. Signed-off-by: Christian Franke <chris@opensourcerouting.org> Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2013-07-05 17:35:37 +02:00
}
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
static inline void _netlink_mpls_build_singlepath(const char *routedesc,
zebra: use const in dplane pw nhlfe accessors Use const in the accessors for pseudowire nhlfe data; pull that through the kernel-facing apis that use that data. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-03-07 19:09:51 +01:00
const zebra_nhlfe_t *nhlfe,
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
struct nlmsghdr *nlmsg,
struct rtmsg *rtmsg,
size_t req_size, int cmd)
Quagga: Install label forwarding entries for statically configured LSPs Install the statically configured LSPs into the FIB (kernel). This is done using the new attributes and definitions for MPLS in the kernel - RTA_VIA, RTA_NEWDST and AF_MPLS. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com> Ticket: CM-4804 Reviewed By: CCR-3088 Testing Done: Manual in SE-1
2016-04-16 04:19:37 +02:00
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
int bytelen;
*: use C99 standard fixed-width integer types The following types are nonstandard: - u_char - u_short - u_int - u_long - u_int8_t - u_int16_t - u_int32_t Replace them with the C99 standard types: - uint8_t - unsigned short - unsigned int - unsigned long - uint8_t - uint16_t - uint32_t Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-03-27 21:13:34 +02:00
uint8_t family;
Quagga: Install label forwarding entries for statically configured LSPs Install the statically configured LSPs into the FIB (kernel). This is done using the new attributes and definitions for MPLS in the kernel - RTA_VIA, RTA_NEWDST and AF_MPLS. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com> Ticket: CM-4804 Reviewed By: CCR-3088 Testing Done: Manual in SE-1
2016-04-16 04:19:37 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
family = NHLFE_FAMILY(nhlfe);
bytelen = (family == AF_INET ? 4 : 16);
_netlink_route_build_singlepath(routedesc, bytelen, nhlfe->nexthop,
nlmsg, rtmsg, req_size, cmd);
Quagga: Install label forwarding entries for statically configured LSPs Install the statically configured LSPs into the FIB (kernel). This is done using the new attributes and definitions for MPLS in the kernel - RTA_VIA, RTA_NEWDST and AF_MPLS. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com> Ticket: CM-4804 Reviewed By: CCR-3088 Testing Done: Manual in SE-1
2016-04-16 04:19:37 +02:00
}
static inline void
zebra: use const in dplane pw nhlfe accessors Use const in the accessors for pseudowire nhlfe data; pull that through the kernel-facing apis that use that data. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-03-07 19:09:51 +01:00
_netlink_mpls_build_multipath(const char *routedesc, const zebra_nhlfe_t *nhlfe,
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
struct rtattr *rta, struct rtnexthop *rtnh,
zebra: use const in dplane pw nhlfe accessors Use const in the accessors for pseudowire nhlfe data; pull that through the kernel-facing apis that use that data. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-03-07 19:09:51 +01:00
struct rtmsg *rtmsg, const union g_addr **src)
Quagga: Install label forwarding entries for statically configured LSPs Install the statically configured LSPs into the FIB (kernel). This is done using the new attributes and definitions for MPLS in the kernel - RTA_VIA, RTA_NEWDST and AF_MPLS. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com> Ticket: CM-4804 Reviewed By: CCR-3088 Testing Done: Manual in SE-1
2016-04-16 04:19:37 +02:00
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
int bytelen;
*: use C99 standard fixed-width integer types The following types are nonstandard: - u_char - u_short - u_int - u_long - u_int8_t - u_int16_t - u_int32_t Replace them with the C99 standard types: - uint8_t - unsigned short - unsigned int - unsigned long - uint8_t - uint16_t - uint32_t Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-03-27 21:13:34 +02:00
uint8_t family;
Quagga: Install label forwarding entries for statically configured LSPs Install the statically configured LSPs into the FIB (kernel). This is done using the new attributes and definitions for MPLS in the kernel - RTA_VIA, RTA_NEWDST and AF_MPLS. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com> Ticket: CM-4804 Reviewed By: CCR-3088 Testing Done: Manual in SE-1
2016-04-16 04:19:37 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
family = NHLFE_FAMILY(nhlfe);
bytelen = (family == AF_INET ? 4 : 16);
_netlink_route_build_multipath(routedesc, bytelen, nhlfe->nexthop, rta,
rtnh, rtmsg, src);
Quagga: Install label forwarding entries for statically configured LSPs Install the statically configured LSPs into the FIB (kernel). This is done using the new attributes and definitions for MPLS in the kernel - RTA_VIA, RTA_NEWDST and AF_MPLS. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com> Ticket: CM-4804 Reviewed By: CCR-3088 Testing Done: Manual in SE-1
2016-04-16 04:19:37 +02:00
}
zebra: rework recursive route resolution Change the datastructure for recursive routes. This brings the following benefits: By using struct nexthop also to store nexthops obtained by recursive resolution, we can get rid of quite a bit of code duplication in the fib management. (rt_netlink, rt_socket, ...) With the new datastructure we can make use of all available paths when recursive routes are resolved with multipath routes. Signed-off-by: Christian Franke <chris@opensourcerouting.org> Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2013-07-05 17:35:37 +02:00
/* Log debug information for netlink_route_multipath
* if debug logging is enabled.
*
* @param cmd: Netlink command which is to be processed
* @param p: Prefix for which the change is due
* @param family: Address family which the change concerns
zebra: Cleanup debugs and add a bit more info This commit does 2 things: 1) When receiving a route from the kernel, display the incoming table as part of the debug, to facilatate knowing what we are talking about as part of the debug. 2) When displaying nexthop information for routes we were sending to the kernel, no need to display the route information every time Display the route then the individual nexthops for what we are doing. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-04-09 20:09:35 +02:00
* @param zvrf: The vrf we are in
* @param tableid: The table we are working on
zebra: rework recursive route resolution Change the datastructure for recursive routes. This brings the following benefits: By using struct nexthop also to store nexthops obtained by recursive resolution, we can get rid of quite a bit of code duplication in the fib management. (rt_netlink, rt_socket, ...) With the new datastructure we can make use of all available paths when recursive routes are resolved with multipath routes. Signed-off-by: Christian Franke <chris@opensourcerouting.org> Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2013-07-05 17:35:37 +02:00
*/
zebra, libs: use const prefix ptrs in apis Add 'const' to prefix args to several zebra route update, redistribution, and route owner notification apis. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-07-10 22:02:03 +02:00
static void _netlink_route_debug(int cmd, const struct prefix *p,
int family, vrf_id_t vrfid,
zebra: Add table and nexthop vrf information to netlink debug Add some additional debug information to the netlink debug messages so we can see the table we are installing to as well as the nexthop's vrf. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-03-29 14:58:45 +02:00
uint32_t tableid)
zebra: rework recursive route resolution Change the datastructure for recursive routes. This brings the following benefits: By using struct nexthop also to store nexthops obtained by recursive resolution, we can get rid of quite a bit of code duplication in the fib management. (rt_netlink, rt_socket, ...) With the new datastructure we can make use of all available paths when recursive routes are resolved with multipath routes. Signed-off-by: Christian Franke <chris@opensourcerouting.org> Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2013-07-05 17:35:37 +02:00
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (IS_ZEBRA_DEBUG_KERNEL) {
char buf[PREFIX_STRLEN];
zlog_debug(
zebra: Cleanup debugs and add a bit more info This commit does 2 things: 1) When receiving a route from the kernel, display the incoming table as part of the debug, to facilatate knowing what we are talking about as part of the debug. 2) When displaying nexthop information for routes we were sending to the kernel, no need to display the route information every time Display the route then the individual nexthops for what we are doing. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-04-09 20:09:35 +02:00
"netlink_route_multipath(): %s %s vrf %u(%u)",
nl_msg_type_to_str(cmd),
prefix2str(p, buf, sizeof(buf)),
zebra, libs: use const prefix ptrs in apis Add 'const' to prefix args to several zebra route update, redistribution, and route owner notification apis. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-07-10 22:02:03 +02:00
vrfid, tableid);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
}
}
zebra: Add kernel debugging function for netlink nexthop messages We needed a kernel debugging function for netlink nexthop messages when people are debugging kernel zebra messages. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-06 17:13:43 +01:00
static void _netlink_nexthop_debug(int cmd, uint32_t id)
{
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug("netlink_nexthop(): %s, id=%u",
nl_msg_type_to_str(cmd), id);
}
*: use C99 standard fixed-width integer types The following types are nonstandard: - u_char - u_short - u_int - u_long - u_int8_t - u_int16_t - u_int32_t Replace them with the C99 standard types: - uint8_t - unsigned short - unsigned int - unsigned long - uint8_t - uint16_t - uint32_t Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-03-27 21:13:34 +02:00
static void _netlink_mpls_debug(int cmd, uint32_t label, const char *routedesc)
Quagga: Install label forwarding entries for statically configured LSPs Install the statically configured LSPs into the FIB (kernel). This is done using the new attributes and definitions for MPLS in the kernel - RTA_VIA, RTA_NEWDST and AF_MPLS. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com> Ticket: CM-4804 Reviewed By: CCR-3088 Testing Done: Manual in SE-1
2016-04-16 04:19:37 +02:00
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug("netlink_mpls_multipath() (%s): %s %u/20", routedesc,
nl_msg_type_to_str(cmd), label);
zebra: rework recursive route resolution Change the datastructure for recursive routes. This brings the following benefits: By using struct nexthop also to store nexthops obtained by recursive resolution, we can get rid of quite a bit of code duplication in the fib management. (rt_netlink, rt_socket, ...) With the new datastructure we can make use of all available paths when recursive routes are resolved with multipath routes. Signed-off-by: Christian Franke <chris@opensourcerouting.org> Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2013-07-05 17:35:37 +02:00
}
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
static int netlink_neigh_update(int cmd, int ifindex, uint32_t addr, char *lla,
zebra: ipv6 operations stick to namespace All ipv6 operations stick to namespace. Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
2017-12-19 12:23:32 +01:00
int llalen, ns_id_t ns_id)
2015-06-11 18:19:59 +02:00
{
zebra: Send RTPROT_ZEBRA for netlink messages missing this data Update neighbor entries and rule entries to have the RTPROT_ZEBRA protocol value. So we can tell where things come from. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-09-19 22:05:39 +02:00
uint8_t protocol = RTPROT_ZEBRA;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
struct {
struct nlmsghdr n;
struct ndmsg ndm;
char buf[256];
} req;
2015-06-11 18:19:59 +02:00
zebra: ipv6 operations stick to namespace All ipv6 operations stick to namespace. Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
2017-12-19 12:23:32 +01:00
struct zebra_ns *zns = zebra_ns_lookup(ns_id);
zebra, lib/memtypes.c: the netlink sockets work per VRF This patch lets the netlink sockets work per VRF. * The definition of "struct nlsock" is moved into zebra/rib.h. * The previous global variables "netlink" and "netlink_cmd" now become the members of "struct zebra_vrf", and are initialized in zebra_vrf_alloc(). * All relative functions now work for a specific VRF, by adding a new parameter which specifies the working VRF, except those functions in which the VRF ID can be obtained from the interface. * kernel_init(), interface_list() and route_read() are now also working per VRF, and moved from main() to zebra_vrf_enable(). * A new function kernel_terminate() is added to release the netlink sockets. It is called from zebra_vrf_disable(). * Correct VRF ID, instead of the previous VRF_DEFAULT, are now passed to the functions of processing interfaces or route entries. Signed-off-by: Feng Lu <lu.feng@6wind.com> Reviewed-by: Alain Ritoux <alain.ritoux@6wind.com> Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Acked-by: Donald Sharp <sharpd@cumulusnetworks.com> Conflicts: lib/memtypes.c zebra/rib.h zebra/rt_netlink.c Conflicts: zebra/if_netlink.c zebra/if_sysctl.c zebra/kernel_null.c zebra/rib.h zebra/rt_netlink.c zebra/rt_netlink.h
2014-07-03 12:23:09 +02:00
zebra: memset buf to prevent uninited writes into kernel Setup the buf used for extra data passed into kernel such that we are cleaning it out before writing data to it, so we can avoid writing uninited data. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-05-18 00:46:14 +02:00
memset(&req, 0, sizeof(req));
2015-06-11 18:19:59 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
req.n.nlmsg_len = NLMSG_LENGTH(sizeof(struct ndmsg));
req.n.nlmsg_flags = NLM_F_CREATE | NLM_F_REQUEST;
req.n.nlmsg_type = cmd; // RTM_NEWNEIGH or RTM_DELNEIGH
req.n.nlmsg_pid = zns->netlink_cmd.snl.nl_pid;
zebra: When calling netlink, set the nl_pid as appropriate This will allow us to further know that we originated the message. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-06-12 19:38:51 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
req.ndm.ndm_family = AF_INET;
req.ndm.ndm_state = NUD_PERMANENT;
req.ndm.ndm_ifindex = ifindex;
req.ndm.ndm_type = RTN_UNICAST;
2015-06-11 18:19:59 +02:00
zebra: Send RTPROT_ZEBRA for netlink messages missing this data Update neighbor entries and rule entries to have the RTPROT_ZEBRA protocol value. So we can tell where things come from. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-09-19 22:05:39 +02:00
addattr_l(&req.n, sizeof(req),
NDA_PROTOCOL, &protocol, sizeof(protocol));
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
addattr_l(&req.n, sizeof(req), NDA_DST, &addr, 4);
addattr_l(&req.n, sizeof(req), NDA_LLADDR, lla, llalen);
2015-06-11 18:19:59 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return netlink_talk(netlink_talk_filter, &req.n, &zns->netlink_cmd, zns,
0);
2015-06-11 18:19:59 +02:00
}
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
/*
* Routing table change via netlink interface, using a dataplane context object
*/
zebra: revise struct names to resolve review comments Use standard type naming and remove use of typedef to resolve some review comments. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-09-19 19:25:12 +02:00
static int netlink_route_multipath(int cmd, struct zebra_dplane_ctx *ctx)
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
{
int bytelen;
struct nexthop *nexthop = NULL;
unsigned int nexthop_num;
int family;
const char *routedesc;
int setsrc = 0;
union g_addr src;
const struct prefix *p, *src_p;
uint32_t table_id;
struct {
struct nlmsghdr n;
struct rtmsg r;
char buf[NL_PKT_BUF_SIZE];
} req;
p = dplane_ctx_get_dest(ctx);
src_p = dplane_ctx_get_src(ctx);
family = PREFIX_FAMILY(p);
zebra: resolve style issues in dplane commit Resolve (most) style issues in the initial zebra dataplane commit branch. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-07-11 17:08:47 +02:00
memset(&req, 0, sizeof(req) - NL_PKT_BUF_SIZE);
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
bytelen = (family == AF_INET ? 4 : 16);
req.n.nlmsg_len = NLMSG_LENGTH(sizeof(struct rtmsg));
req.n.nlmsg_flags = NLM_F_CREATE | NLM_F_REQUEST;
zebra: When installing a new route always use REPLACE When we install a new route into the kernel always use REPLACE. Else if the route is already there it can be translated into an append with the flags we are using. This is especially true for the way we handle pbr routes as that we are re-installing the same route entry from pbr at the moment. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-03-01 19:56:12 +01:00
if ((cmd == RTM_NEWROUTE) &&
((p->family == AF_INET) || v6_rr_semantics))
req.n.nlmsg_flags |= NLM_F_REPLACE;
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
req.n.nlmsg_type = cmd;
req.n.nlmsg_pid = dplane_ctx_get_ns(ctx)->nls.snl.nl_pid;
req.r.rtm_family = family;
req.r.rtm_dst_len = p->prefixlen;
req.r.rtm_src_len = src_p ? src_p->prefixlen : 0;
req.r.rtm_scope = RT_SCOPE_UNIVERSE;
zebra: resolve style issues in dplane commit Resolve (most) style issues in the initial zebra dataplane commit branch. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-07-11 17:08:47 +02:00
if (cmd == RTM_DELROUTE)
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
req.r.rtm_protocol = zebra2proto(dplane_ctx_get_old_type(ctx));
zebra: resolve style issues in dplane commit Resolve (most) style issues in the initial zebra dataplane commit branch. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-07-11 17:08:47 +02:00
else
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
req.r.rtm_protocol = zebra2proto(dplane_ctx_get_type(ctx));
/*
* blackhole routes are not RTN_UNICAST, they are
* RTN_ BLACKHOLE|UNREACHABLE|PROHIBIT
* so setting this value as a RTN_UNICAST would
* cause the route lookup of just the prefix
* to fail. So no need to specify this for
* the RTM_DELROUTE case
*/
if (cmd != RTM_DELROUTE)
req.r.rtm_type = RTN_UNICAST;
zebra: resolve style issues in dplane commit Resolve (most) style issues in the initial zebra dataplane commit branch. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-07-11 17:08:47 +02:00
addattr_l(&req.n, sizeof(req), RTA_DST, &p->u.prefix, bytelen);
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
if (src_p)
zebra: resolve style issues in dplane commit Resolve (most) style issues in the initial zebra dataplane commit branch. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-07-11 17:08:47 +02:00
addattr_l(&req.n, sizeof(req), RTA_SRC, &src_p->u.prefix,
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
bytelen);
/* Metric. */
/* Hardcode the metric for all routes coming from zebra. Metric isn't
* used
* either by the kernel or by zebra. Its purely for calculating best
* path(s)
* by the routing protocol and for communicating with protocol peers.
*/
zebra: resolve style issues in dplane commit Resolve (most) style issues in the initial zebra dataplane commit branch. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-07-11 17:08:47 +02:00
addattr32(&req.n, sizeof(req), RTA_PRIORITY, NL_DEFAULT_ROUTE_METRIC);
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
#if defined(SUPPORT_REALMS)
{
route_tag_t tag;
zebra: resolve style issues in dplane commit Resolve (most) style issues in the initial zebra dataplane commit branch. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-07-11 17:08:47 +02:00
if (cmd == RTM_DELROUTE)
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
tag = dplane_ctx_get_old_tag(ctx);
zebra: resolve style issues in dplane commit Resolve (most) style issues in the initial zebra dataplane commit branch. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-07-11 17:08:47 +02:00
else
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
tag = dplane_ctx_get_tag(ctx);
if (tag > 0 && tag <= 255)
zebra: resolve style issues in dplane commit Resolve (most) style issues in the initial zebra dataplane commit branch. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-07-11 17:08:47 +02:00
addattr32(&req.n, sizeof(req), RTA_FLOW, tag);
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
}
#endif
/* Table corresponding to this route. */
table_id = dplane_ctx_get_table(ctx);
if (table_id < 256)
req.r.rtm_table = table_id;
else {
req.r.rtm_table = RT_TABLE_UNSPEC;
zebra: resolve style issues in dplane commit Resolve (most) style issues in the initial zebra dataplane commit branch. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-07-11 17:08:47 +02:00
addattr32(&req.n, sizeof(req), RTA_TABLE, table_id);
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
}
_netlink_route_debug(cmd, p, family, dplane_ctx_get_vrf(ctx), table_id);
/*
* If we are not updating the route and we have received
* a route delete, then all we need to fill in is the
* prefix information to tell the kernel to schwack
* it.
*/
if (cmd == RTM_DELROUTE)
goto skip;
if (dplane_ctx_get_mtu(ctx) || dplane_ctx_get_nh_mtu(ctx)) {
char buf[NL_PKT_BUF_SIZE];
struct rtattr *rta = (void *)buf;
uint32_t mtu = dplane_ctx_get_mtu(ctx);
uint32_t nexthop_mtu = dplane_ctx_get_nh_mtu(ctx);
zebra: resolve style issues in dplane commit Resolve (most) style issues in the initial zebra dataplane commit branch. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-07-11 17:08:47 +02:00
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
if (!mtu || (nexthop_mtu && nexthop_mtu < mtu))
mtu = nexthop_mtu;
rta->rta_type = RTA_METRICS;
rta->rta_len = RTA_LENGTH(0);
zebra: resolve style issues in dplane commit Resolve (most) style issues in the initial zebra dataplane commit branch. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-07-11 17:08:47 +02:00
rta_addattr_l(rta, NL_PKT_BUF_SIZE,
RTAX_MTU, &mtu, sizeof(mtu));
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
addattr_l(&req.n, NL_PKT_BUF_SIZE, RTA_METRICS, RTA_DATA(rta),
RTA_PAYLOAD(rta));
}
zebra: Move the supports_nh bool to a better place Move the supports_nh bool indicating whether the kernel we are using supports nexthop objects into the netlink kernel interface itself. Since only linux and netlink support nexthop object APIs for now this is fine. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-01 20:53:06 +02:00
if (supports_nh) {
zebra: Use NHE id on kernel route sends If the kernel supports nexthop objects, send the route using an nhg_hash_entry ID instead. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 01:42:39 +02:00
/* Kernel supports nexthop objects */
addattr32(&req.n, sizeof(req), RTA_NH_ID,
dplane_ctx_get_nhe_id(ctx));
goto skip;
}
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
/* Count overall nexthops so we can decide whether to use singlepath
zebra: resolve style issues in dplane commit Resolve (most) style issues in the initial zebra dataplane commit branch. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-07-11 17:08:47 +02:00
* or multipath case.
*/
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
nexthop_num = 0;
for (ALL_NEXTHOPS_PTR(dplane_ctx_get_ng(ctx), nexthop)) {
if (CHECK_FLAG(nexthop->flags, NEXTHOP_FLAG_RECURSIVE))
continue;
if (cmd == RTM_NEWROUTE && !NEXTHOP_IS_ACTIVE(nexthop->flags))
continue;
nexthop_num++;
}
/* Singlepath case. */
zebra: Allow zebra to only mark up to multipath_num nexthops as ACTIVE NEXTHOP_FLAG_ACTIVE currently means that the nexthop is considered good enough to be installed. With current ecmp restrictions this translation from multipath_num is enforced in the data plane. The problem with this is of course that every data plane now becomes concerned about the multipath num and must enforce it independently. Currently *bsd does not honor multipath_num at all and linux marks all nexthops as being installed even when it honors a multipath_num that is less than the total. This code change moves the multipath_num enforcement from a dataplane decision to a zebra nexthop decision. Thus dataplanes now can just install those nexthops marked as NEXTHOP_FLAG_ACTIVE without having to worry about multipath_num. *BSD will now respect multipath_num and Linux now properly notes which routes are actually installed or not: sharpd@donna ~/f/t/topotests> ps -ef | grep frr frr 6261 1556 0 09:12 ? 00:00:00 /usr/lib/frr/zebra -e 2 --daemon -A 127.0.0.1 frr 6279 1556 0 09:12 ? 00:00:00 /usr/lib/frr/staticd --daemon -A 127.0.0.1 donna.cumulusnetworks.com(config)# do show ip route Codes: K - kernel route, C - connected, S - static, R - RIP, O - OSPF, I - IS-IS, B - BGP, E - EIGRP, N - NHRP, T - Table, v - VNC, V - VNC-Direct, A - Babel, D - SHARP, F - PBR, f - OpenFabric, > - selected route, * - FIB route K>* 0.0.0.0/0 [0/106] via 10.0.2.2, enp0s3, 00:00:45 S>* 4.4.4.4/32 [1/0] via 10.0.2.1, enp0s3, 00:00:02 * via 192.168.209.1, enp0s8, 00:00:02 via 192.168.210.1, enp0s9 inactive, 00:00:02 C>* 10.0.2.0/24 is directly connected, enp0s3, 00:00:45 C>* 192.168.209.0/24 is directly connected, enp0s8, 00:00:45 C>* 192.168.210.0/24 is directly connected, enp0s9, 00:00:45 donna.cumulusnetworks.com(config)# sharpd@donna ~/f/t/topotests> ip route show default via 10.0.2.2 dev enp0s3 proto dhcp metric 106 4.4.4.4 proto 196 metric 20 nexthop via 10.0.2.1 dev enp0s3 weight 1 nexthop via 192.168.209.1 dev enp0s8 weight 1 10.0.2.0/24 dev enp0s3 proto kernel scope link src 10.0.2.15 metric 106 172.17.0.0/16 dev docker0 proto kernel scope link src 172.17.0.1 linkdown 192.168.122.0/24 dev virbr0 proto kernel scope link src 192.168.122.1 linkdown 192.168.209.0/24 dev enp0s8 proto kernel scope link src 192.168.209.2 metric 105 192.168.210.0/24 dev enp0s9 proto kernel scope link src 192.168.210.2 metric 103 Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-12-13 15:21:26 +01:00
if (nexthop_num == 1) {
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
nexthop_num = 0;
for (ALL_NEXTHOPS_PTR(dplane_ctx_get_ng(ctx), nexthop)) {
/*
* So we want to cover 2 types of blackhole
* routes here:
* 1) A normal blackhole route( ala from a static
* install.
* 2) A recursively resolved blackhole route
*/
if (nexthop->type == NEXTHOP_TYPE_BLACKHOLE) {
switch (nexthop->bh_type) {
case BLACKHOLE_ADMINPROHIB:
req.r.rtm_type = RTN_PROHIBIT;
break;
case BLACKHOLE_REJECT:
req.r.rtm_type = RTN_UNREACHABLE;
break;
default:
req.r.rtm_type = RTN_BLACKHOLE;
break;
}
goto skip;
}
if (CHECK_FLAG(nexthop->flags,
NEXTHOP_FLAG_RECURSIVE)) {
zebra: resolve style issues in dplane commit Resolve (most) style issues in the initial zebra dataplane commit branch. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-07-11 17:08:47 +02:00
if (setsrc)
continue;
if (family == AF_INET) {
if (nexthop->rmap_src.ipv4.s_addr
!= 0) {
src.ipv4 =
nexthop->rmap_src.ipv4;
setsrc = 1;
} else if (nexthop->src.ipv4.s_addr
!= 0) {
src.ipv4 =
nexthop->src.ipv4;
setsrc = 1;
}
} else if (family == AF_INET6) {
if (!IN6_IS_ADDR_UNSPECIFIED(
&nexthop->rmap_src.ipv6)) {
src.ipv6 =
nexthop->rmap_src.ipv6;
setsrc = 1;
} else if (
!IN6_IS_ADDR_UNSPECIFIED(
&nexthop->src.ipv6)) {
src.ipv6 =
nexthop->src.ipv6;
setsrc = 1;
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
}
}
zebra: add handy res2str utility Add a 2str utility for dplane result codes; use it in a debug or two. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-08-17 22:53:24 +02:00
continue;
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
}
if ((cmd == RTM_NEWROUTE
&& NEXTHOP_IS_ACTIVE(nexthop->flags))) {
routedesc = nexthop->rparent
? "recursive, single-path"
: "single-path";
_netlink_route_build_singlepath(
routedesc, bytelen, nexthop, &req.n,
zebra: resolve style issues in dplane commit Resolve (most) style issues in the initial zebra dataplane commit branch. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-07-11 17:08:47 +02:00
&req.r, sizeof(req), cmd);
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
nexthop_num++;
break;
}
}
if (setsrc && (cmd == RTM_NEWROUTE)) {
if (family == AF_INET)
zebra: resolve style issues in dplane commit Resolve (most) style issues in the initial zebra dataplane commit branch. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-07-11 17:08:47 +02:00
addattr_l(&req.n, sizeof(req), RTA_PREFSRC,
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
&src.ipv4, bytelen);
else if (family == AF_INET6)
zebra: resolve style issues in dplane commit Resolve (most) style issues in the initial zebra dataplane commit branch. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-07-11 17:08:47 +02:00
addattr_l(&req.n, sizeof(req), RTA_PREFSRC,
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
&src.ipv6, bytelen);
}
} else { /* Multipath case */
char buf[NL_PKT_BUF_SIZE];
struct rtattr *rta = (void *)buf;
struct rtnexthop *rtnh;
zebra: use const in dplane pw nhlfe accessors Use const in the accessors for pseudowire nhlfe data; pull that through the kernel-facing apis that use that data. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-03-07 19:09:51 +01:00
const union g_addr *src1 = NULL;
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
rta->rta_type = RTA_MULTIPATH;
rta->rta_len = RTA_LENGTH(0);
rtnh = RTA_DATA(rta);
nexthop_num = 0;
for (ALL_NEXTHOPS_PTR(dplane_ctx_get_ng(ctx), nexthop)) {
if (CHECK_FLAG(nexthop->flags,
NEXTHOP_FLAG_RECURSIVE)) {
/* This only works for IPv4 now */
zebra: resolve style issues in dplane commit Resolve (most) style issues in the initial zebra dataplane commit branch. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-07-11 17:08:47 +02:00
if (setsrc)
continue;
if (family == AF_INET) {
if (nexthop->rmap_src.ipv4.s_addr
!= 0) {
src.ipv4 =
nexthop->rmap_src.ipv4;
setsrc = 1;
} else if (nexthop->src.ipv4.s_addr
!= 0) {
src.ipv4 =
nexthop->src.ipv4;
setsrc = 1;
}
} else if (family == AF_INET6) {
if (!IN6_IS_ADDR_UNSPECIFIED(
&nexthop->rmap_src.ipv6)) {
src.ipv6 =
nexthop->rmap_src.ipv6;
setsrc = 1;
} else if (
!IN6_IS_ADDR_UNSPECIFIED(
&nexthop->src.ipv6)) {
src.ipv6 =
nexthop->src.ipv6;
setsrc = 1;
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
}
}
zebra: Fix netlink installation of recursive ecmp routes Recursive multipath nexthops were broken by the initial async dataplane - we were trying to install an extra, invalid nexthop. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-11-13 16:21:16 +01:00
continue;
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
}
if ((cmd == RTM_NEWROUTE
&& NEXTHOP_IS_ACTIVE(nexthop->flags))) {
routedesc = nexthop->rparent
? "recursive, multipath"
: "multipath";
nexthop_num++;
_netlink_route_build_multipath(
routedesc, bytelen, nexthop, rta, rtnh,
&req.r, &src1);
rtnh = RTNH_NEXT(rtnh);
if (!setsrc && src1) {
if (family == AF_INET)
src.ipv4 = src1->ipv4;
else if (family == AF_INET6)
src.ipv6 = src1->ipv6;
setsrc = 1;
}
}
}
if (setsrc && (cmd == RTM_NEWROUTE)) {
if (family == AF_INET)
zebra: resolve style issues in dplane commit Resolve (most) style issues in the initial zebra dataplane commit branch. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-07-11 17:08:47 +02:00
addattr_l(&req.n, sizeof(req), RTA_PREFSRC,
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
&src.ipv4, bytelen);
else if (family == AF_INET6)
zebra: resolve style issues in dplane commit Resolve (most) style issues in the initial zebra dataplane commit branch. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-07-11 17:08:47 +02:00
addattr_l(&req.n, sizeof(req), RTA_PREFSRC,
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
&src.ipv6, bytelen);
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug("Setting source");
}
if (rta->rta_len > RTA_LENGTH(0))
addattr_l(&req.n, NL_PKT_BUF_SIZE, RTA_MULTIPATH,
RTA_DATA(rta), RTA_PAYLOAD(rta));
}
/* If there is no useful nexthop then return. */
if (nexthop_num == 0) {
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug(
"netlink_route_multipath(): No useful nexthop.");
return 0;
}
skip:
/* Talk to netlink socket. */
return netlink_talk_info(netlink_talk_filter, &req.n,
dplane_ctx_get_ns(ctx), 0);
}
zebra: Allow S,G lookup to use RTNL_FAMILY_IPMR This current implementation unfortunately must ask the kernel for all mroutes because vrf's do not have the ability to request a single mroute at this time. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-06-06 16:18:17 +02:00
int kernel_get_ipmr_sg_stats(struct zebra_vrf *zvrf, void *in)
lib, pimd, zebra: Allow pimd to ask the kernel about mroute info When we need to lookup the mroute info for a route. Allow pimd to ask the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2016-08-11 02:04:20 +02:00
{
zebra: Send correct default vrf tableid for MROUTE stats So the linux kernel uses the RT_TABLE_MAIN for the table id used for ip routing. The multicast routing tables use RT_TABLE_DEFAULT. We changed the internal code of zebra_vrf a few months back to use RT_TABLE_MAIN as the tableid to use. This caused the pim sg stats to stop working because of the kernel bug where it uses a different table for ip routing and ip multicast. Put a bit of a special case in to do the right thing. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-09-17 15:18:40 +02:00
uint32_t actual_table;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
int suc = 0;
struct mcast_route_data *mr = (struct mcast_route_data *)in;
zebra: Convert get_stats to use new kernel interface Convert the XXX_get_stats functions to use the new kernel interface. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-07-05 17:53:29 +02:00
struct {
struct nlmsghdr n;
struct ndmsg ndm;
char buf[256];
} req;
lib, pimd, zebra: Allow pimd to ask the kernel about mroute info When we need to lookup the mroute info for a route. Allow pimd to ask the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2016-08-11 02:04:20 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
mroute = mr;
zebra: ipv6 operations stick to namespace All ipv6 operations stick to namespace. Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
2017-12-19 12:23:32 +01:00
struct zebra_ns *zns;
zebra: Convert get_stats to use new kernel interface Convert the XXX_get_stats functions to use the new kernel interface. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-07-05 17:53:29 +02:00
zebra: retrieve zns context from zvrf when netlink discovery So as to get the correct NETNS where some discovery must be done and populated, the zns pointer is directly retrieved from zvrf, instead of checking that the VRF is a backend NETNS or not. In the case where the interfaces are discovered before the VRF is enabled ( VRF-lite populate), then the default NS is retrieved. Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
2018-02-13 10:48:48 +01:00
zns = zvrf->zns;
zebra: memset buf to prevent uninited writes into kernel Setup the buf used for extra data passed into kernel such that we are cleaning it out before writing data to it, so we can avoid writing uninited data. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-05-18 00:46:14 +02:00
memset(&req, 0, sizeof(req));
zebra: Convert get_stats to use new kernel interface Convert the XXX_get_stats functions to use the new kernel interface. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-07-05 17:53:29 +02:00
req.n.nlmsg_len = NLMSG_LENGTH(sizeof(struct ndmsg));
req.n.nlmsg_flags = NLM_F_REQUEST;
req.n.nlmsg_pid = zns->netlink_cmd.snl.nl_pid;
req.ndm.ndm_family = RTNL_FAMILY_IPMR;
req.n.nlmsg_type = RTM_GETROUTE;
addattr_l(&req.n, sizeof(req), RTA_IIF, &mroute->ifindex, 4);
addattr_l(&req.n, sizeof(req), RTA_OIF, &mroute->ifindex, 4);
addattr_l(&req.n, sizeof(req), RTA_SRC, &mroute->sg.src.s_addr, 4);
addattr_l(&req.n, sizeof(req), RTA_DST, &mroute->sg.grp.s_addr, 4);
zebra: Send correct default vrf tableid for MROUTE stats So the linux kernel uses the RT_TABLE_MAIN for the table id used for ip routing. The multicast routing tables use RT_TABLE_DEFAULT. We changed the internal code of zebra_vrf a few months back to use RT_TABLE_MAIN as the tableid to use. This caused the pim sg stats to stop working because of the kernel bug where it uses a different table for ip routing and ip multicast. Put a bit of a special case in to do the right thing. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-09-17 15:18:40 +02:00
/*
* What?
*
* So during the namespace cleanup we started storing
* the zvrf table_id for the default table as RT_TABLE_MAIN
* which is what the normal routing table for ip routing is.
* This change caused this to break our lookups of sg data
* because prior to this change the zvrf->table_id was 0
* and when the pim multicast kernel code saw a 0,
* it was auto-translated to RT_TABLE_DEFAULT. But since
* we are now passing in RT_TABLE_MAIN there is no auto-translation
* and the kernel goes screw you and the delicious cookies you
* are trying to give me. So now we have this little hack.
*/
actual_table = (zvrf->table_id == RT_TABLE_MAIN) ? RT_TABLE_DEFAULT :
zvrf->table_id;
addattr_l(&req.n, sizeof(req), RTA_TABLE, &actual_table, 4);
lib, pimd, zebra: Allow pimd to ask the kernel about mroute info When we need to lookup the mroute info for a route. Allow pimd to ask the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2016-08-11 02:04:20 +02:00
zebra: Convert get_stats to use new kernel interface Convert the XXX_get_stats functions to use the new kernel interface. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-07-05 17:53:29 +02:00
suc = netlink_talk(netlink_route_change_read_multicast, &req.n,
&zns->netlink_cmd, zns, 0);
lib, pimd, zebra: Allow pimd to ask the kernel about mroute info When we need to lookup the mroute info for a route. Allow pimd to ask the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2016-08-11 02:04:20 +02:00
zebra: Convert get_stats to use new kernel interface Convert the XXX_get_stats functions to use the new kernel interface. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-07-05 17:53:29 +02:00
mroute = NULL;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return suc;
lib, pimd, zebra: Allow pimd to ask the kernel about mroute info When we need to lookup the mroute info for a route. Allow pimd to ask the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2016-08-11 02:04:20 +02:00
}
zebra: Handle nhg_hash_entry encaps/more debugging Add code for handling nexthop group hash entry encaps and sending them to the kernel. Add some more debugging information for the encaps and groups in general. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-12 21:46:22 +02:00
/* Char length to debug ID with */
#define ID_LENGTH 10
zebra: Add functionality to send groups to kernel Add functionality to allow us to send nexthop groups to the kernel. It creates a nexthop_grp array based on the dependency list in the nhg_hash_entry and then shoves that into the netlink message. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-27 00:35:08 +01:00
static void _netlink_nexthop_build_group(struct nlmsghdr *n, size_t req_size,
zebra: Handle nhg_hash_entry encaps/more debugging Add code for handling nexthop group hash entry encaps and sending them to the kernel. Add some more debugging information for the encaps and groups in general. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-12 21:46:22 +02:00
uint32_t id,
zebra: Use a nhe context dataplane and rib metaq We will use a nhe context for dataplane interaction with nextho group hash entries. New nhe's from the kernel will be put into a group array if they are a group and queued on the rib metaq to be processed later. New nhe's sent to the kernel will be set on the dataplane context with approprate ID's in the group array if needed. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 00:03:29 +02:00
const struct nh_grp *z_grp,
zebra,lib: Refactor depends to RB tree Refactor the depends to use an RB tree instead of a list. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-14 02:10:34 +02:00
const uint8_t count)
zebra: Add functionality to send groups to kernel Add functionality to allow us to send nexthop groups to the kernel. It creates a nexthop_grp array based on the dependency list in the nhg_hash_entry and then shoves that into the netlink message. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-27 00:35:08 +01:00
{
struct nexthop_grp grp[count];
zebra: Handle nhg_hash_entry encaps/more debugging Add code for handling nexthop group hash entry encaps and sending them to the kernel. Add some more debugging information for the encaps and groups in general. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-12 21:46:22 +02:00
/* Need space for max group size, "/", and null term */
char buf[(MULTIPATH_NUM * (ID_LENGTH + 1)) + 1];
char buf1[ID_LENGTH + 2];
buf[0] = '\0';
zebra: Add functionality to send groups to kernel Add functionality to allow us to send nexthop groups to the kernel. It creates a nexthop_grp array based on the dependency list in the nhg_hash_entry and then shoves that into the netlink message. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-27 00:35:08 +01:00
memset(grp, 0, sizeof(grp));
if (count) {
zebra,lib: Refactor depends to RB tree Refactor the depends to use an RB tree instead of a list. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-14 02:10:34 +02:00
for (int i = 0; i < count; i++) {
zebra: Use a nhe context dataplane and rib metaq We will use a nhe context for dataplane interaction with nextho group hash entries. New nhe's from the kernel will be put into a group array if they are a group and queued on the rib metaq to be processed later. New nhe's sent to the kernel will be set on the dataplane context with approprate ID's in the group array if needed. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 00:03:29 +02:00
grp[i].id = z_grp[i].id;
grp[i].weight = z_grp[i].weight;
zebra: Handle nhg_hash_entry encaps/more debugging Add code for handling nexthop group hash entry encaps and sending them to the kernel. Add some more debugging information for the encaps and groups in general. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-12 21:46:22 +02:00
if (IS_ZEBRA_DEBUG_KERNEL) {
if (i == 0)
snprintf(buf, sizeof(buf1), "group %u",
grp[i].id);
else {
snprintf(buf1, sizeof(buf1), "/%u",
grp[i].id);
strlcat(buf, buf1, sizeof(buf));
}
}
zebra: Add functionality to send groups to kernel Add functionality to allow us to send nexthop groups to the kernel. It creates a nexthop_grp array based on the dependency list in the nhg_hash_entry and then shoves that into the netlink message. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-27 00:35:08 +01:00
}
zebra,lib: Refactor depends to RB tree Refactor the depends to use an RB tree instead of a list. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-14 02:10:34 +02:00
addattr_l(n, req_size, NHA_GROUP, grp, count * sizeof(*grp));
zebra: Add functionality to send groups to kernel Add functionality to allow us to send nexthop groups to the kernel. It creates a nexthop_grp array based on the dependency list in the nhg_hash_entry and then shoves that into the netlink message. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-27 00:35:08 +01:00
}
zebra: Handle nhg_hash_entry encaps/more debugging Add code for handling nexthop group hash entry encaps and sending them to the kernel. Add some more debugging information for the encaps and groups in general. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-12 21:46:22 +02:00
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug("%s: ID (%u): %s", __func__, id, buf);
zebra: Add functionality to send groups to kernel Add functionality to allow us to send nexthop groups to the kernel. It creates a nexthop_grp array based on the dependency list in the nhg_hash_entry and then shoves that into the netlink message. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-27 00:35:08 +01:00
}
zebra: Add base functionality for nexthop processing via the dataplane Add all the neccessary code to allow nexthops to be processed in separate dataplane contexts with the netlink dataplane kernel provider. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-01 23:17:00 +01:00
/**
* netlink_nexthop() - Nexthop change via the netlink interface
*
* @ctx: Dataplane ctx
*
* Return: Result status
*/
static int netlink_nexthop(int cmd, struct zebra_dplane_ctx *ctx)
{
struct {
struct nlmsghdr n;
struct nhmsg nhm;
char buf[NL_PKT_BUF_SIZE];
} req;
zebra: Handle nhg_hash_entry encaps/more debugging Add code for handling nexthop group hash entry encaps and sending them to the kernel. Add some more debugging information for the encaps and groups in general. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-12 21:46:22 +02:00
mpls_lse_t out_lse[MPLS_MAX_LABELS];
char label_buf[256];
int num_labels = 0;
size_t req_size = sizeof(req);
zebra: Move the supports_nh bool to a better place Move the supports_nh bool indicating whether the kernel we are using supports nexthop objects into the netlink kernel interface itself. Since only linux and netlink support nexthop object APIs for now this is fine. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-01 20:53:06 +02:00
/* Nothing to do if the kernel doesn't support nexthop objects */
if (!supports_nh)
return 0;
zebra: Handle nhg_hash_entry encaps/more debugging Add code for handling nexthop group hash entry encaps and sending them to the kernel. Add some more debugging information for the encaps and groups in general. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-12 21:46:22 +02:00
label_buf[0] = '\0';
memset(&req, 0, req_size);
zebra: Add base functionality for nexthop processing via the dataplane Add all the neccessary code to allow nexthops to be processed in separate dataplane contexts with the netlink dataplane kernel provider. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-01 23:17:00 +01:00
req.n.nlmsg_len = NLMSG_LENGTH(sizeof(struct nhmsg));
req.n.nlmsg_flags = NLM_F_CREATE | NLM_F_REQUEST;
zebra: Add handling for kernel del/update nexthop Add handling for delete/update nexthop object messages from the kernel. If someone deletes a nexthop object we are still using, send it back down. If the someone updates a nexthop we are using, replace that nexthop with ours. Routes are referencing this nexthop object ID and we resolved it ourselves, so we should force the other `someone` to submit to our will. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-01 20:24:35 +02:00
if (cmd == RTM_NEWNEXTHOP)
req.n.nlmsg_flags |= NLM_F_REPLACE;
zebra: Add base functionality for nexthop processing via the dataplane Add all the neccessary code to allow nexthops to be processed in separate dataplane contexts with the netlink dataplane kernel provider. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-01 23:17:00 +01:00
req.n.nlmsg_type = cmd;
zebra: Use the dest prefix for determining nexthop family Device only nexthops still need an address family associated with them. Decided to get this from the destination prefix on it. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-06 19:43:40 +01:00
req.n.nlmsg_pid = dplane_ctx_get_ns(ctx)->nls.snl.nl_pid;
zebra: Add base functionality for nexthop processing via the dataplane Add all the neccessary code to allow nexthops to be processed in separate dataplane contexts with the netlink dataplane kernel provider. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-01 23:17:00 +01:00
req.nhm.nh_family = AF_UNSPEC;
zebra: Zebra nexthop group re-work checkpatch fixes Checkpatch fixes for the zebra nexthop group re-work. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-09-03 22:12:06 +02:00
/* TODO: Scope? */
zebra: Add base functionality for nexthop processing via the dataplane Add all the neccessary code to allow nexthops to be processed in separate dataplane contexts with the netlink dataplane kernel provider. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-01 23:17:00 +01:00
zebra,lib: Refactor depends to RB tree Refactor the depends to use an RB tree instead of a list. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-14 02:10:34 +02:00
uint32_t id = dplane_ctx_get_nhe_id(ctx);
if (!id) {
zebra: Add base functionality for nexthop processing via the dataplane Add all the neccessary code to allow nexthops to be processed in separate dataplane contexts with the netlink dataplane kernel provider. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-01 23:17:00 +01:00
flog_err(
EC_ZEBRA_NHG_FIB_UPDATE,
"Failed trying to update a nexthop group in the kernel that does not have an ID");
return -1;
}
zebra: Handle nhg_hash_entry encaps/more debugging Add code for handling nexthop group hash entry encaps and sending them to the kernel. Add some more debugging information for the encaps and groups in general. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-12 21:46:22 +02:00
addattr32(&req.n, req_size, NHA_ID, id);
zebra: Add base functionality for nexthop processing via the dataplane Add all the neccessary code to allow nexthops to be processed in separate dataplane contexts with the netlink dataplane kernel provider. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-01 23:17:00 +01:00
if (cmd == RTM_NEWNEXTHOP) {
zebra: Use a nhe context dataplane and rib metaq We will use a nhe context for dataplane interaction with nextho group hash entries. New nhe's from the kernel will be put into a group array if they are a group and queued on the rib metaq to be processed later. New nhe's sent to the kernel will be set on the dataplane context with approprate ID's in the group array if needed. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 00:03:29 +02:00
if (dplane_ctx_get_nhe_nh_grp_count(ctx))
zebra,lib: Refactor depends to RB tree Refactor the depends to use an RB tree instead of a list. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-14 02:10:34 +02:00
_netlink_nexthop_build_group(
zebra: Handle nhg_hash_entry encaps/more debugging Add code for handling nexthop group hash entry encaps and sending them to the kernel. Add some more debugging information for the encaps and groups in general. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-12 21:46:22 +02:00
&req.n, req_size, id,
zebra: Use a nhe context dataplane and rib metaq We will use a nhe context for dataplane interaction with nextho group hash entries. New nhe's from the kernel will be put into a group array if they are a group and queued on the rib metaq to be processed later. New nhe's sent to the kernel will be set on the dataplane context with approprate ID's in the group array if needed. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 00:03:29 +02:00
dplane_ctx_get_nhe_nh_grp(ctx),
dplane_ctx_get_nhe_nh_grp_count(ctx));
zebra,lib: Refactor depends to RB tree Refactor the depends to use an RB tree instead of a list. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-14 02:10:34 +02:00
else {
const struct nexthop *nh =
dplane_ctx_get_nhe_ng(ctx)->nexthop;
afi_t afi = dplane_ctx_get_nhe_afi(ctx);
zebra: Add afi value for all nexthops sent/received Since nexthops are always going to need to be address family specific unless they are only a group, we have to address this when we receive and send them. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-20 18:03:22 +01:00
zebra,lib: Refactor depends to RB tree Refactor the depends to use an RB tree instead of a list. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-14 02:10:34 +02:00
if (afi == AFI_IP)
zebra: Add functionality to send groups to kernel Add functionality to allow us to send nexthop groups to the kernel. It creates a nexthop_grp array based on the dependency list in the nhg_hash_entry and then shoves that into the netlink message. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-27 00:35:08 +01:00
req.nhm.nh_family = AF_INET;
zebra,lib: Refactor depends to RB tree Refactor the depends to use an RB tree instead of a list. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-14 02:10:34 +02:00
else if (afi == AFI_IP6)
zebra: Add functionality to send groups to kernel Add functionality to allow us to send nexthop groups to the kernel. It creates a nexthop_grp array based on the dependency list in the nhg_hash_entry and then shoves that into the netlink message. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-27 00:35:08 +01:00
req.nhm.nh_family = AF_INET6;
zebra: Add base functionality for nexthop processing via the dataplane Add all the neccessary code to allow nexthops to be processed in separate dataplane contexts with the netlink dataplane kernel provider. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-01 23:17:00 +01:00
zebra: Add functionality to send groups to kernel Add functionality to allow us to send nexthop groups to the kernel. It creates a nexthop_grp array based on the dependency list in the nhg_hash_entry and then shoves that into the netlink message. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-27 00:35:08 +01:00
switch (nh->type) {
zebra: Fix nhg ifindex setting and checking We were only setting and checking the ifindex if the nexthop had an *_IFINDEX type. However, when nexthop active checking is done, the non-*_IFINDEX types can also obtain a nexthop with an ifindex and are thus valid too. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-04-22 21:42:10 +02:00
case NEXTHOP_TYPE_IPV4:
zebra: Add functionality to send groups to kernel Add functionality to allow us to send nexthop groups to the kernel. It creates a nexthop_grp array based on the dependency list in the nhg_hash_entry and then shoves that into the netlink message. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-27 00:35:08 +01:00
case NEXTHOP_TYPE_IPV4_IFINDEX:
zebra: Handle nhg_hash_entry encaps/more debugging Add code for handling nexthop group hash entry encaps and sending them to the kernel. Add some more debugging information for the encaps and groups in general. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-12 21:46:22 +02:00
addattr_l(&req.n, req_size, NHA_GATEWAY,
zebra: Add functionality to send groups to kernel Add functionality to allow us to send nexthop groups to the kernel. It creates a nexthop_grp array based on the dependency list in the nhg_hash_entry and then shoves that into the netlink message. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-27 00:35:08 +01:00
&nh->gate.ipv4, IPV4_MAX_BYTELEN);
break;
zebra: Fix nhg ifindex setting and checking We were only setting and checking the ifindex if the nexthop had an *_IFINDEX type. However, when nexthop active checking is done, the non-*_IFINDEX types can also obtain a nexthop with an ifindex and are thus valid too. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-04-22 21:42:10 +02:00
case NEXTHOP_TYPE_IPV6:
zebra: Add functionality to send groups to kernel Add functionality to allow us to send nexthop groups to the kernel. It creates a nexthop_grp array based on the dependency list in the nhg_hash_entry and then shoves that into the netlink message. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-27 00:35:08 +01:00
case NEXTHOP_TYPE_IPV6_IFINDEX:
zebra: Handle nhg_hash_entry encaps/more debugging Add code for handling nexthop group hash entry encaps and sending them to the kernel. Add some more debugging information for the encaps and groups in general. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-12 21:46:22 +02:00
addattr_l(&req.n, req_size, NHA_GATEWAY,
zebra: Add functionality to send groups to kernel Add functionality to allow us to send nexthop groups to the kernel. It creates a nexthop_grp array based on the dependency list in the nhg_hash_entry and then shoves that into the netlink message. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-27 00:35:08 +01:00
&nh->gate.ipv6, IPV6_MAX_BYTELEN);
break;
case NEXTHOP_TYPE_BLACKHOLE:
zebra: Handle nhg_hash_entry encaps/more debugging Add code for handling nexthop group hash entry encaps and sending them to the kernel. Add some more debugging information for the encaps and groups in general. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-12 21:46:22 +02:00
addattr_l(&req.n, req_size, NHA_BLACKHOLE, NULL,
0);
/* Blackhole shouldn't have anymore attributes
*/
goto nexthop_done;
zebra: Add functionality to send groups to kernel Add functionality to allow us to send nexthop groups to the kernel. It creates a nexthop_grp array based on the dependency list in the nhg_hash_entry and then shoves that into the netlink message. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-27 00:35:08 +01:00
case NEXTHOP_TYPE_IFINDEX:
/* Don't need anymore info for this */
break;
zebra: Fix nhg ifindex setting and checking We were only setting and checking the ifindex if the nexthop had an *_IFINDEX type. However, when nexthop active checking is done, the non-*_IFINDEX types can also obtain a nexthop with an ifindex and are thus valid too. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-04-22 21:42:10 +02:00
}
if (!nh->ifindex) {
zebra: Add functionality to send groups to kernel Add functionality to allow us to send nexthop groups to the kernel. It creates a nexthop_grp array based on the dependency list in the nhg_hash_entry and then shoves that into the netlink message. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-27 00:35:08 +01:00
flog_err(
EC_ZEBRA_NHG_FIB_UPDATE,
"Context received for kernel nexthop update without an interface");
return -1;
}
zebra: Handle nhg_hash_entry encaps/more debugging Add code for handling nexthop group hash entry encaps and sending them to the kernel. Add some more debugging information for the encaps and groups in general. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-12 21:46:22 +02:00
addattr32(&req.n, req_size, NHA_OIF, nh->ifindex);
num_labels =
build_label_stack(nh->nh_label, out_lse,
label_buf, sizeof(label_buf));
if (num_labels) {
/* Set the BoS bit */
out_lse[num_labels - 1] |=
htonl(1 << MPLS_LS_S_SHIFT);
/*
* TODO: MPLS unsupported for now in kernel.
*/
if (req.nhm.nh_family == AF_MPLS)
goto nexthop_done;
#if 0
addattr_l(&req.n, req_size, NHA_NEWDST,
&out_lse,
num_labels
* sizeof(mpls_lse_t));
#endif
else {
struct rtattr *nest;
uint16_t encap = LWTUNNEL_ENCAP_MPLS;
addattr_l(&req.n, req_size,
NHA_ENCAP_TYPE, &encap,
sizeof(uint16_t));
nest = addattr_nest(&req.n, req_size,
NHA_ENCAP);
addattr_l(&req.n, req_size,
MPLS_IPTUNNEL_DST, &out_lse,
num_labels
* sizeof(mpls_lse_t));
addattr_nest_end(&req.n, nest);
}
}
nexthop_done:
if (IS_ZEBRA_DEBUG_KERNEL) {
char buf[NEXTHOP_STRLEN];
snprintfrr(buf, sizeof(buf), "%pNHv", nh);
zlog_debug("%s: ID (%u): %s (%u) %s ", __func__,
id, buf, nh->vrf_id, label_buf);
}
zebra: Add base functionality for nexthop processing via the dataplane Add all the neccessary code to allow nexthops to be processed in separate dataplane contexts with the netlink dataplane kernel provider. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-01 23:17:00 +01:00
}
zebra: Sweep our nexthop objects out on restart On restart, if we failed to remove any nexthop objects due to a kill -9 or such event, sweep them if we aren't using them. Add a proto field to handle this and remove the is_kernel bool. Add a dupicate flag that indicates this nexthop group is only present in our ID hashtable. It is a dupicate nexthop we received from the kernel, therefore we cannot hash on it. Make the idcounter globally accessible so that kernel updates increment it as soon as we receive them, not when we handle them. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-01 20:07:04 +02:00
req.nhm.nh_protocol = zebra2proto(dplane_ctx_get_nhe_type(ctx));
zebra: Add base functionality for nexthop processing via the dataplane Add all the neccessary code to allow nexthops to be processed in separate dataplane contexts with the netlink dataplane kernel provider. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-01 23:17:00 +01:00
} else if (cmd != RTM_DELNEXTHOP) {
flog_err(
EC_ZEBRA_NHG_FIB_UPDATE,
"Nexthop group kernel update command (%d) does not exist",
cmd);
return -1;
}
zebra,lib: Refactor depends to RB tree Refactor the depends to use an RB tree instead of a list. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-14 02:10:34 +02:00
_netlink_nexthop_debug(cmd, id);
zebra: Add base functionality for nexthop processing via the dataplane Add all the neccessary code to allow nexthops to be processed in separate dataplane contexts with the netlink dataplane kernel provider. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-01 23:17:00 +01:00
return netlink_talk_info(netlink_talk_filter, &req.n,
dplane_ctx_get_ns(ctx), 0);
}
/**
* kernel_nexthop_update() - Update/delete a nexthop from the kernel
*
* @ctx: Dataplane context
*
* Return: Dataplane result flag
*/
enum zebra_dplane_result kernel_nexthop_update(struct zebra_dplane_ctx *ctx)
{
zebra: Add recursive functionality to NHE's Add the ability to recursively resolve nexthop group hash entries and resolve them when sending to the kernel. When copying over nexthops into an NHE, copy resolved info as well. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 00:27:40 +02:00
int cmd = 0;
int ret = 0;
zebra: Add base functionality for nexthop processing via the dataplane Add all the neccessary code to allow nexthops to be processed in separate dataplane contexts with the netlink dataplane kernel provider. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-01 23:17:00 +01:00
switch (dplane_ctx_get_op(ctx)) {
case DPLANE_OP_NH_DELETE:
cmd = RTM_DELNEXTHOP;
break;
case DPLANE_OP_NH_INSTALL:
case DPLANE_OP_NH_UPDATE:
cmd = RTM_NEWNEXTHOP;
break;
case DPLANE_OP_ROUTE_INSTALL:
case DPLANE_OP_ROUTE_UPDATE:
case DPLANE_OP_ROUTE_DELETE:
case DPLANE_OP_ROUTE_NOTIFY:
case DPLANE_OP_LSP_INSTALL:
case DPLANE_OP_LSP_UPDATE:
case DPLANE_OP_LSP_DELETE:
case DPLANE_OP_LSP_NOTIFY:
case DPLANE_OP_PW_INSTALL:
case DPLANE_OP_PW_UNINSTALL:
case DPLANE_OP_SYS_ROUTE_ADD:
case DPLANE_OP_SYS_ROUTE_DELETE:
case DPLANE_OP_ADDR_INSTALL:
case DPLANE_OP_ADDR_UNINSTALL:
case DPLANE_OP_MAC_INSTALL:
case DPLANE_OP_MAC_DELETE:
zebra: Add DPLANE_NEIGH and DPLANE_VTEP to nhg cases Add DPLANE_OP_NEIGH and DPLANE_OP_VTEP to nhg dplane handler's switch statements. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-09-25 00:38:28 +02:00
case DPLANE_OP_NEIGH_INSTALL:
case DPLANE_OP_NEIGH_UPDATE:
case DPLANE_OP_NEIGH_DELETE:
case DPLANE_OP_VTEP_ADD:
case DPLANE_OP_VTEP_DELETE:
zebra: Add base functionality for nexthop processing via the dataplane Add all the neccessary code to allow nexthops to be processed in separate dataplane contexts with the netlink dataplane kernel provider. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-01 23:17:00 +01:00
case DPLANE_OP_NONE:
flog_err(
EC_ZEBRA_NHG_FIB_UPDATE,
"Context received for kernel nexthop update with incorrect OP code (%u)",
dplane_ctx_get_op(ctx));
return ZEBRA_DPLANE_REQUEST_FAILURE;
}
ret = netlink_nexthop(cmd, ctx);
return (ret == 0 ? ZEBRA_DPLANE_REQUEST_SUCCESS
: ZEBRA_DPLANE_REQUEST_FAILURE);
}
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
/*
* Update or delete a prefix from the kernel,
* using info from a dataplane context.
*/
zebra: revise struct names to resolve review comments Use standard type naming and remove use of typedef to resolve some review comments. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-09-19 19:25:12 +02:00
enum zebra_dplane_result kernel_route_update(struct zebra_dplane_ctx *ctx)
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
{
int cmd, ret;
const struct prefix *p = dplane_ctx_get_dest(ctx);
zebra: add handy res2str utility Add a 2str utility for dplane result codes; use it in a debug or two. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-08-17 22:53:24 +02:00
struct nexthop *nexthop;
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
if (dplane_ctx_get_op(ctx) == DPLANE_OP_ROUTE_DELETE) {
cmd = RTM_DELROUTE;
} else if (dplane_ctx_get_op(ctx) == DPLANE_OP_ROUTE_INSTALL) {
cmd = RTM_NEWROUTE;
} else if (dplane_ctx_get_op(ctx) == DPLANE_OP_ROUTE_UPDATE) {
if (p->family == AF_INET || v6_rr_semantics) {
/* Single 'replace' operation */
cmd = RTM_NEWROUTE;
zebra: Fix route replace v4 semantics with new system route When a new system route comes in and we have a pre-existing non-system route we are not deleting the current system route from the linux kernel. Modify the code such that when a route replace is sent to the kernel with a new route as a system route and the old route as a non-system route do a delete of the old route so it is no longer in the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-07-29 17:10:25 +02:00
/*
* With route replace semantics in place
* for v4 routes and the new route is a system
* route we do not install anything.
* The problem here is that the new system
* route should cause us to withdraw from
* the kernel the old non-system route
*/
if (RSYSTEM_ROUTE(dplane_ctx_get_type(ctx)) &&
!RSYSTEM_ROUTE(dplane_ctx_get_old_type(ctx)))
(void)netlink_route_multipath(RTM_DELROUTE,
ctx);
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
} else {
/*
* So v6 route replace semantics are not in
* the kernel at this point as I understand it.
* so let's do a delete then an add.
* In the future once v6 route replace semantics
* are in we can figure out what to do here to
* allow working with old and new kernels.
*
* I'm also intentionally ignoring the failure case
* of the route delete. If that happens yeah we're
* screwed.
*/
zebra: Modify code so that dplane is responsible for indicating success/fail of install We have several route types KERNEL and CONNECT that are handled via special case in the code. This was causing a lot of work keeping the two different classes of route types as special(SYSTEM OR NOT). Put the dplane in charge of the code that sets the bits for signalling route install/failure. This greatly simplifies the code calling path and makes all route types be handled exactly the same. Additionaly code that we want to run post data plane install can just work as per normal then, instead of having to know we need to run it when we have a special type of route. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com.
2019-02-08 21:01:41 +01:00
if (!RSYSTEM_ROUTE(dplane_ctx_get_old_type(ctx)))
(void)netlink_route_multipath(RTM_DELROUTE,
ctx);
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
cmd = RTM_NEWROUTE;
}
} else {
return ZEBRA_DPLANE_REQUEST_FAILURE;
}
zebra: Modify code so that dplane is responsible for indicating success/fail of install We have several route types KERNEL and CONNECT that are handled via special case in the code. This was causing a lot of work keeping the two different classes of route types as special(SYSTEM OR NOT). Put the dplane in charge of the code that sets the bits for signalling route install/failure. This greatly simplifies the code calling path and makes all route types be handled exactly the same. Additionaly code that we want to run post data plane install can just work as per normal then, instead of having to know we need to run it when we have a special type of route. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com.
2019-02-08 21:01:41 +01:00
if (!RSYSTEM_ROUTE(dplane_ctx_get_type(ctx)))
ret = netlink_route_multipath(cmd, ctx);
else
ret = 0;
zebra: add handy res2str utility Add a 2str utility for dplane result codes; use it in a debug or two. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-08-17 22:53:24 +02:00
if ((cmd == RTM_NEWROUTE) && (ret == 0)) {
/* Update installed nexthops to signal which have been
* installed.
*/
for (ALL_NEXTHOPS_PTR(dplane_ctx_get_ng(ctx), nexthop)) {
if (CHECK_FLAG(nexthop->flags, NEXTHOP_FLAG_RECURSIVE))
continue;
if (CHECK_FLAG(nexthop->flags, NEXTHOP_FLAG_ACTIVE)) {
SET_FLAG(nexthop->flags, NEXTHOP_FLAG_FIB);
}
}
}
zebra: start dataplane layer work Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Slide netlink_talk_info into place to remove dependency on core zebra structs; add accessors for dplane context block Start init of route context from zebra core re and rn structs; start queueing and event handling for incoming route updates. Expose netlink apis that don't rely on zebra core structs; add parallel route-update code path using the dplane ctx; simplest possible event loop to process queued route' updates. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-05-23 18:20:43 +02:00
return (ret == 0 ?
ZEBRA_DPLANE_REQUEST_SUCCESS : ZEBRA_DPLANE_REQUEST_FAILURE);
}
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
/**
* netlink_nexthop_process_nh() - Parse the gatway/if info from a new nexthop
*
* @tb: Netlink RTA data
* @family: Address family in the nhmsg
zebra: Add new nexthops to the interface nexthop hash entry list When we get a new nexthop and find the interface associated with it, add this nexthop to the interface's zebra interface info nexthop hash entry list. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-11 21:31:44 +01:00
* @ifp: Interface connected - this should be NULL, we fill it in
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
* @ns_id: Namspace id
*
* Return: New nexthop
*/
zebra: Use a nhe context dataplane and rib metaq We will use a nhe context for dataplane interaction with nextho group hash entries. New nhe's from the kernel will be put into a group array if they are a group and queued on the rib metaq to be processed later. New nhe's sent to the kernel will be set on the dataplane context with approprate ID's in the group array if needed. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 00:03:29 +02:00
static struct nexthop netlink_nexthop_process_nh(struct rtattr **tb,
unsigned char family,
struct interface **ifp,
ns_id_t ns_id)
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
{
zebra: Use a nhe context dataplane and rib metaq We will use a nhe context for dataplane interaction with nextho group hash entries. New nhe's from the kernel will be put into a group array if they are a group and queued on the rib metaq to be processed later. New nhe's sent to the kernel will be set on the dataplane context with approprate ID's in the group array if needed. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 00:03:29 +02:00
struct nexthop nh = {};
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
void *gate = NULL;
zebra: Refactor nexthop group creation code to use allocated memory Simplify the code for nexthop hash entry creation. I made nexthop hash entry creation expect the nexthop group and depends to always be allocated before lookup. Before, it was only allocated if it had dependencies. I think it makes the code a bit more readable to go ahead an allocate even for single nexthops as well. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-22 18:11:07 +01:00
enum nexthop_types_t type = 0;
zebra: Use a nhe context dataplane and rib metaq We will use a nhe context for dataplane interaction with nextho group hash entries. New nhe's from the kernel will be put into a group array if they are a group and queued on the rib metaq to be processed later. New nhe's sent to the kernel will be set on the dataplane context with approprate ID's in the group array if needed. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 00:03:29 +02:00
int if_index = 0;
size_t sz = 0;
zebra: Fix some nhg SA issues found in latest Coverity Fix 2 Coverity issues: 1) zebra_nhg.c -> all paths in nhg_ctx_process_finish have already deref'ed the ctx pointer no need for a test of it 2) the **ifp pointer passed in may be NULL. Prevent an accidental deref if calling function does not pass in a ifp pointer. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-10-29 01:30:06 +01:00
struct interface *ifp_lookup;
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
if_index = *(int *)RTA_DATA(tb[NHA_OIF]);
zebra: Refactor nexthop group creation code to use allocated memory Simplify the code for nexthop hash entry creation. I made nexthop hash entry creation expect the nexthop group and depends to always be allocated before lookup. Before, it was only allocated if it had dependencies. I think it makes the code a bit more readable to go ahead an allocate even for single nexthops as well. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-22 18:11:07 +01:00
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
if (tb[NHA_GATEWAY]) {
switch (family) {
case AF_INET:
zebra: Refactor nexthop group creation code to use allocated memory Simplify the code for nexthop hash entry creation. I made nexthop hash entry creation expect the nexthop group and depends to always be allocated before lookup. Before, it was only allocated if it had dependencies. I think it makes the code a bit more readable to go ahead an allocate even for single nexthops as well. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-22 18:11:07 +01:00
type = NEXTHOP_TYPE_IPV4_IFINDEX;
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
sz = 4;
break;
case AF_INET6:
zebra: Refactor nexthop group creation code to use allocated memory Simplify the code for nexthop hash entry creation. I made nexthop hash entry creation expect the nexthop group and depends to always be allocated before lookup. Before, it was only allocated if it had dependencies. I think it makes the code a bit more readable to go ahead an allocate even for single nexthops as well. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-22 18:11:07 +01:00
type = NEXTHOP_TYPE_IPV6_IFINDEX;
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
sz = 16;
break;
default:
flog_warn(
EC_ZEBRA_BAD_NHG_MESSAGE,
zebra: Make bad address family log message more clear The message for an invalid address family on a nexthop gateway did not specify that is what for the gateway specifically. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-19 21:43:27 +01:00
"Nexthop gateway with bad address family (%d) received from kernel",
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
family);
zebra: Use a nhe context dataplane and rib metaq We will use a nhe context for dataplane interaction with nextho group hash entries. New nhe's from the kernel will be put into a group array if they are a group and queued on the rib metaq to be processed later. New nhe's sent to the kernel will be set on the dataplane context with approprate ID's in the group array if needed. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 00:03:29 +02:00
return nh;
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
}
gate = RTA_DATA(tb[NHA_GATEWAY]);
zebra: Use a nhe context dataplane and rib metaq We will use a nhe context for dataplane interaction with nextho group hash entries. New nhe's from the kernel will be put into a group array if they are a group and queued on the rib metaq to be processed later. New nhe's sent to the kernel will be set on the dataplane context with approprate ID's in the group array if needed. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 00:03:29 +02:00
} else
zebra: Refactor nexthop group creation code to use allocated memory Simplify the code for nexthop hash entry creation. I made nexthop hash entry creation expect the nexthop group and depends to always be allocated before lookup. Before, it was only allocated if it had dependencies. I think it makes the code a bit more readable to go ahead an allocate even for single nexthops as well. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-22 18:11:07 +01:00
type = NEXTHOP_TYPE_IFINDEX;
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
zebra: Refactor nexthop group creation code to use allocated memory Simplify the code for nexthop hash entry creation. I made nexthop hash entry creation expect the nexthop group and depends to always be allocated before lookup. Before, it was only allocated if it had dependencies. I think it makes the code a bit more readable to go ahead an allocate even for single nexthops as well. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-22 18:11:07 +01:00
if (type)
zebra: Use a nhe context dataplane and rib metaq We will use a nhe context for dataplane interaction with nextho group hash entries. New nhe's from the kernel will be put into a group array if they are a group and queued on the rib metaq to be processed later. New nhe's sent to the kernel will be set on the dataplane context with approprate ID's in the group array if needed. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 00:03:29 +02:00
nh.type = type;
zebra: Refactor nexthop group creation code to use allocated memory Simplify the code for nexthop hash entry creation. I made nexthop hash entry creation expect the nexthop group and depends to always be allocated before lookup. Before, it was only allocated if it had dependencies. I think it makes the code a bit more readable to go ahead an allocate even for single nexthops as well. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-22 18:11:07 +01:00
if (gate)
zebra: Use a nhe context dataplane and rib metaq We will use a nhe context for dataplane interaction with nextho group hash entries. New nhe's from the kernel will be put into a group array if they are a group and queued on the rib metaq to be processed later. New nhe's sent to the kernel will be set on the dataplane context with approprate ID's in the group array if needed. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 00:03:29 +02:00
memcpy(&(nh.gate), gate, sz);
zebra: Refactor nexthop group creation code to use allocated memory Simplify the code for nexthop hash entry creation. I made nexthop hash entry creation expect the nexthop group and depends to always be allocated before lookup. Before, it was only allocated if it had dependencies. I think it makes the code a bit more readable to go ahead an allocate even for single nexthops as well. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-22 18:11:07 +01:00
if (if_index)
zebra: Use a nhe context dataplane and rib metaq We will use a nhe context for dataplane interaction with nextho group hash entries. New nhe's from the kernel will be put into a group array if they are a group and queued on the rib metaq to be processed later. New nhe's sent to the kernel will be set on the dataplane context with approprate ID's in the group array if needed. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 00:03:29 +02:00
nh.ifindex = if_index;
zebra: Refactor nexthop group creation code to use allocated memory Simplify the code for nexthop hash entry creation. I made nexthop hash entry creation expect the nexthop group and depends to always be allocated before lookup. Before, it was only allocated if it had dependencies. I think it makes the code a bit more readable to go ahead an allocate even for single nexthops as well. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-22 18:11:07 +01:00
zebra: Fix some nhg SA issues found in latest Coverity Fix 2 Coverity issues: 1) zebra_nhg.c -> all paths in nhg_ctx_process_finish have already deref'ed the ctx pointer no need for a test of it 2) the **ifp pointer passed in may be NULL. Prevent an accidental deref if calling function does not pass in a ifp pointer. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-10-29 01:30:06 +01:00
ifp_lookup =
if_lookup_by_index_per_ns(zebra_ns_lookup(ns_id), nh.ifindex);
zebra: Use a nhe context dataplane and rib metaq We will use a nhe context for dataplane interaction with nextho group hash entries. New nhe's from the kernel will be put into a group array if they are a group and queued on the rib metaq to be processed later. New nhe's sent to the kernel will be set on the dataplane context with approprate ID's in the group array if needed. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 00:03:29 +02:00
if (ifp)
zebra: Fix some nhg SA issues found in latest Coverity Fix 2 Coverity issues: 1) zebra_nhg.c -> all paths in nhg_ctx_process_finish have already deref'ed the ctx pointer no need for a test of it 2) the **ifp pointer passed in may be NULL. Prevent an accidental deref if calling function does not pass in a ifp pointer. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-10-29 01:30:06 +01:00
*ifp = ifp_lookup;
if (ifp_lookup)
nh.vrf_id = ifp_lookup->vrf_id;
zebra: Use a nhe context dataplane and rib metaq We will use a nhe context for dataplane interaction with nextho group hash entries. New nhe's from the kernel will be put into a group array if they are a group and queued on the rib metaq to be processed later. New nhe's sent to the kernel will be set on the dataplane context with approprate ID's in the group array if needed. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 00:03:29 +02:00
else {
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
flog_warn(
EC_ZEBRA_UNKNOWN_INTERFACE,
"%s: Unknown nexthop interface %u received, defaulting to VRF_DEFAULT",
zebra: Use a nhe context dataplane and rib metaq We will use a nhe context for dataplane interaction with nextho group hash entries. New nhe's from the kernel will be put into a group array if they are a group and queued on the rib metaq to be processed later. New nhe's sent to the kernel will be set on the dataplane context with approprate ID's in the group array if needed. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 00:03:29 +02:00
__PRETTY_FUNCTION__, nh.ifindex);
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
zebra: Use a nhe context dataplane and rib metaq We will use a nhe context for dataplane interaction with nextho group hash entries. New nhe's from the kernel will be put into a group array if they are a group and queued on the rib metaq to be processed later. New nhe's sent to the kernel will be set on the dataplane context with approprate ID's in the group array if needed. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 00:03:29 +02:00
nh.vrf_id = VRF_DEFAULT;
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
}
if (tb[NHA_ENCAP] && tb[NHA_ENCAP_TYPE]) {
uint16_t encap_type = *(uint16_t *)RTA_DATA(tb[NHA_ENCAP_TYPE]);
int num_labels = 0;
zebra: TODO for hanlding blackhole attr exclusive Add a TODO statement for handling the exclusiveness of blackhole attributes. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-26 23:06:10 +02:00
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
mpls_label_t labels[MPLS_MAX_LABELS] = {0};
zebra: Use a nhe context dataplane and rib metaq We will use a nhe context for dataplane interaction with nextho group hash entries. New nhe's from the kernel will be put into a group array if they are a group and queued on the rib metaq to be processed later. New nhe's sent to the kernel will be set on the dataplane context with approprate ID's in the group array if needed. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 00:03:29 +02:00
if (encap_type == LWTUNNEL_ENCAP_MPLS)
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
num_labels = parse_encap_mpls(tb[NHA_ENCAP], labels);
zebra: Use a nhe context dataplane and rib metaq We will use a nhe context for dataplane interaction with nextho group hash entries. New nhe's from the kernel will be put into a group array if they are a group and queued on the rib metaq to be processed later. New nhe's sent to the kernel will be set on the dataplane context with approprate ID's in the group array if needed. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 00:03:29 +02:00
if (num_labels)
nexthop_add_labels(&nh, ZEBRA_LSP_STATIC, num_labels,
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
labels);
}
return nh;
}
zebra: Read in nexthop dependencies from the kernel Add functionality to read in a group from the kernel, create a hash entry for it, and add its nexthops to its dependency list. Further, we create its nhg struct separtely from this, copying over any nexthops it should reference directly into it. Thus, we have two types for representation of the nexthop group: nhe->nhg_depends->[nhe, nhe, nhe] nhe->nhg->nexthop->nexthop->nexthop Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-21 15:47:19 +01:00
static int netlink_nexthop_process_group(struct rtattr **tb,
zebra: Guard nexthop group overflow read Guard against an overflow read when processing nexthop groups from netlink. Add a check to ensure we don't try to write passed the array size. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-10-23 19:08:10 +02:00
struct nh_grp *z_grp, int z_grp_size)
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
{
zebra: Use a nhe context dataplane and rib metaq We will use a nhe context for dataplane interaction with nextho group hash entries. New nhe's from the kernel will be put into a group array if they are a group and queued on the rib metaq to be processed later. New nhe's sent to the kernel will be set on the dataplane context with approprate ID's in the group array if needed. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 00:03:29 +02:00
uint8_t count = 0;
/* linux/nexthop.h group struct */
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
struct nexthop_grp *n_grp = NULL;
zebra: Read in nexthop dependencies from the kernel Add functionality to read in a group from the kernel, create a hash entry for it, and add its nexthops to its dependency list. Further, we create its nhg struct separtely from this, copying over any nexthops it should reference directly into it. Thus, we have two types for representation of the nexthop group: nhe->nhg_depends->[nhe, nhe, nhe] nhe->nhg->nexthop->nexthop->nexthop Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-21 15:47:19 +01:00
n_grp = (struct nexthop_grp *)RTA_DATA(tb[NHA_GROUP]);
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
count = (RTA_PAYLOAD(tb[NHA_GROUP]) / sizeof(*n_grp));
if (!count || (count * sizeof(*n_grp)) != RTA_PAYLOAD(tb[NHA_GROUP])) {
flog_warn(EC_ZEBRA_BAD_NHG_MESSAGE,
"Invalid nexthop group received from the kernel");
zebra: Read in nexthop dependencies from the kernel Add functionality to read in a group from the kernel, create a hash entry for it, and add its nexthops to its dependency list. Further, we create its nhg struct separtely from this, copying over any nexthops it should reference directly into it. Thus, we have two types for representation of the nexthop group: nhe->nhg_depends->[nhe, nhe, nhe] nhe->nhg->nexthop->nexthop->nexthop Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-21 15:47:19 +01:00
return count;
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
}
zebra: Sweep our nexthop objects out on restart On restart, if we failed to remove any nexthop objects due to a kill -9 or such event, sweep them if we aren't using them. Add a proto field to handle this and remove the is_kernel bool. Add a dupicate flag that indicates this nexthop group is only present in our ID hashtable. It is a dupicate nexthop we received from the kernel, therefore we cannot hash on it. Make the idcounter globally accessible so that kernel updates increment it as soon as we receive them, not when we handle them. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-01 20:07:04 +02:00
#if 0
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
// TODO: Need type for something?
zebra: Read in nexthop dependencies from the kernel Add functionality to read in a group from the kernel, create a hash entry for it, and add its nexthops to its dependency list. Further, we create its nhg struct separtely from this, copying over any nexthops it should reference directly into it. Thus, we have two types for representation of the nexthop group: nhe->nhg_depends->[nhe, nhe, nhe] nhe->nhg->nexthop->nexthop->nexthop Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-21 15:47:19 +01:00
zlog_debug("Nexthop group type: %d",
*((uint16_t *)RTA_DATA(tb[NHA_GROUP_TYPE])));
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
zebra: Sweep our nexthop objects out on restart On restart, if we failed to remove any nexthop objects due to a kill -9 or such event, sweep them if we aren't using them. Add a proto field to handle this and remove the is_kernel bool. Add a dupicate flag that indicates this nexthop group is only present in our ID hashtable. It is a dupicate nexthop we received from the kernel, therefore we cannot hash on it. Make the idcounter globally accessible so that kernel updates increment it as soon as we receive them, not when we handle them. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-01 20:07:04 +02:00
#endif
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
zebra: Guard nexthop group overflow read Guard against an overflow read when processing nexthop groups from netlink. Add a check to ensure we don't try to write passed the array size. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-10-23 19:08:10 +02:00
for (int i = 0; ((i < count) && (i < z_grp_size)); i++) {
zebra: Use a nhe context dataplane and rib metaq We will use a nhe context for dataplane interaction with nextho group hash entries. New nhe's from the kernel will be put into a group array if they are a group and queued on the rib metaq to be processed later. New nhe's sent to the kernel will be set on the dataplane context with approprate ID's in the group array if needed. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 00:03:29 +02:00
z_grp[i].id = n_grp[i].id;
z_grp[i].weight = n_grp[i].weight;
zebra: Read in nexthop dependencies from the kernel Add functionality to read in a group from the kernel, create a hash entry for it, and add its nexthops to its dependency list. Further, we create its nhg struct separtely from this, copying over any nexthops it should reference directly into it. Thus, we have two types for representation of the nexthop group: nhe->nhg_depends->[nhe, nhe, nhe] nhe->nhg->nexthop->nexthop->nexthop Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-21 15:47:19 +01:00
}
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
return count;
}
/**
* netlink_nexthop_change() - Read in change about nexthops from the kernel
*
* @h: Netlink message header
* @ns_id: Namspace id
* @startup: Are we reading under startup conditions?
*
* Return: Result status
*/
int netlink_nexthop_change(struct nlmsghdr *h, ns_id_t ns_id, int startup)
{
int len;
/* nexthop group id */
uint32_t id;
unsigned char family;
zebra: Sweep our nexthop objects out on restart On restart, if we failed to remove any nexthop objects due to a kill -9 or such event, sweep them if we aren't using them. Add a proto field to handle this and remove the is_kernel bool. Add a dupicate flag that indicates this nexthop group is only present in our ID hashtable. It is a dupicate nexthop we received from the kernel, therefore we cannot hash on it. Make the idcounter globally accessible so that kernel updates increment it as soon as we receive them, not when we handle them. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-01 20:07:04 +02:00
int type;
zebra: Add afi value for all nexthops sent/received Since nexthops are always going to need to be address family specific unless they are only a group, we have to address this when we receive and send them. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-20 18:03:22 +01:00
afi_t afi = AFI_UNSPEC;
zebra: VRF ID should be null if a nexthop group A nexthop group should not have a VRF ID. Only individual nexthops need to be using a VRF. Fixed this both kernel and proto side. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-04-11 18:00:51 +02:00
vrf_id_t vrf_id = 0;
zebra: Add new nexthops to the interface nexthop hash entry list When we get a new nexthop and find the interface associated with it, add this nexthop to the interface's zebra interface info nexthop hash entry list. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-11 21:31:44 +01:00
struct interface *ifp = NULL;
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
struct nhmsg *nhm = NULL;
zebra: Use a nhe context dataplane and rib metaq We will use a nhe context for dataplane interaction with nextho group hash entries. New nhe's from the kernel will be put into a group array if they are a group and queued on the rib metaq to be processed later. New nhe's sent to the kernel will be set on the dataplane context with approprate ID's in the group array if needed. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 00:03:29 +02:00
struct nexthop nh = {};
struct nh_grp grp[MULTIPATH_NUM] = {};
zebra: Read in nexthop dependencies from the kernel Add functionality to read in a group from the kernel, create a hash entry for it, and add its nexthops to its dependency list. Further, we create its nhg struct separtely from this, copying over any nexthops it should reference directly into it. Thus, we have two types for representation of the nexthop group: nhe->nhg_depends->[nhe, nhe, nhe] nhe->nhg->nexthop->nexthop->nexthop Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-21 15:47:19 +01:00
/* Count of nexthops in group array */
zebra: Use a nhe context dataplane and rib metaq We will use a nhe context for dataplane interaction with nextho group hash entries. New nhe's from the kernel will be put into a group array if they are a group and queued on the rib metaq to be processed later. New nhe's sent to the kernel will be set on the dataplane context with approprate ID's in the group array if needed. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 00:03:29 +02:00
uint8_t grp_count = 0;
struct rtattr *tb[NHA_MAX + 1] = {};
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
nhm = NLMSG_DATA(h);
if (startup && h->nlmsg_type != RTM_NEWNEXTHOP)
return 0;
len = h->nlmsg_len - NLMSG_LENGTH(sizeof(struct nhmsg));
if (len < 0) {
zlog_warn(
"%s: Message received from netlink is of a broken size %d %zu",
__PRETTY_FUNCTION__, h->nlmsg_len,
(size_t)NLMSG_LENGTH(sizeof(struct nhmsg)));
return -1;
}
netlink_parse_rtattr(tb, NHA_MAX, RTM_NHA(nhm), len);
if (!tb[NHA_ID]) {
flog_warn(
EC_ZEBRA_BAD_NHG_MESSAGE,
"Nexthop group without an ID received from the kernel");
return -1;
}
/* We use the ID key'd nhg table for kernel updates */
id = *((uint32_t *)RTA_DATA(tb[NHA_ID]));
zebra: Add afi value for all nexthops sent/received Since nexthops are always going to need to be address family specific unless they are only a group, we have to address this when we receive and send them. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-20 18:03:22 +01:00
family = nhm->nh_family;
afi = family2afi(family);
zebra: Sweep our nexthop objects out on restart On restart, if we failed to remove any nexthop objects due to a kill -9 or such event, sweep them if we aren't using them. Add a proto field to handle this and remove the is_kernel bool. Add a dupicate flag that indicates this nexthop group is only present in our ID hashtable. It is a dupicate nexthop we received from the kernel, therefore we cannot hash on it. Make the idcounter globally accessible so that kernel updates increment it as soon as we receive them, not when we handle them. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-01 20:07:04 +02:00
type = proto2zebra(nhm->nh_protocol, 0, true);
zebra: Make kernel debug nexthop standardized Make the the kernel debug zlog for nexthop messages from the kernel more aligned with the route message kernel debug zlog. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-28 20:18:28 +01:00
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug("%s ID (%u) %s NS %u",
nl_msg_type_to_str(h->nlmsg_type), id,
nl_family_to_str(family), ns_id);
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
if (h->nlmsg_type == RTM_NEWNEXTHOP) {
if (tb[NHA_GROUP]) {
/**
* If this is a group message its only going to have
* an array of nexthop IDs associated with it
*/
zebra: Guard nexthop group overflow read Guard against an overflow read when processing nexthop groups from netlink. Add a check to ensure we don't try to write passed the array size. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-10-23 19:08:10 +02:00
grp_count = netlink_nexthop_process_group(
tb, grp, array_size(grp));
zebra: Read in nexthop dependencies from the kernel Add functionality to read in a group from the kernel, create a hash entry for it, and add its nexthops to its dependency list. Further, we create its nhg struct separtely from this, copying over any nexthops it should reference directly into it. Thus, we have two types for representation of the nexthop group: nhe->nhg_depends->[nhe, nhe, nhe] nhe->nhg->nexthop->nexthop->nexthop Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-21 15:47:19 +01:00
} else {
if (tb[NHA_BLACKHOLE]) {
/**
* This nexthop is just for blackhole-ing
* traffic, it should not have an OIF, GATEWAY,
* or ENCAP
*/
zebra: Use a nhe context dataplane and rib metaq We will use a nhe context for dataplane interaction with nextho group hash entries. New nhe's from the kernel will be put into a group array if they are a group and queued on the rib metaq to be processed later. New nhe's sent to the kernel will be set on the dataplane context with approprate ID's in the group array if needed. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 00:03:29 +02:00
nh.type = NEXTHOP_TYPE_BLACKHOLE;
nh.bh_type = BLACKHOLE_UNSPEC;
} else if (tb[NHA_OIF])
zebra: Read in nexthop dependencies from the kernel Add functionality to read in a group from the kernel, create a hash entry for it, and add its nexthops to its dependency list. Further, we create its nhg struct separtely from this, copying over any nexthops it should reference directly into it. Thus, we have two types for representation of the nexthop group: nhe->nhg_depends->[nhe, nhe, nhe] nhe->nhg->nexthop->nexthop->nexthop Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-21 15:47:19 +01:00
/**
* This is a true new nexthop, so we need
* to parse the gateway and device info
*/
nh = netlink_nexthop_process_nh(tb, family,
&ifp, ns_id);
zebra: Use a nhe context dataplane and rib metaq We will use a nhe context for dataplane interaction with nextho group hash entries. New nhe's from the kernel will be put into a group array if they are a group and queued on the rib metaq to be processed later. New nhe's sent to the kernel will be set on the dataplane context with approprate ID's in the group array if needed. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 00:03:29 +02:00
else {
zebra: Refactor nexthop group creation code to use allocated memory Simplify the code for nexthop hash entry creation. I made nexthop hash entry creation expect the nexthop group and depends to always be allocated before lookup. Before, it was only allocated if it had dependencies. I think it makes the code a bit more readable to go ahead an allocate even for single nexthops as well. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-22 18:11:07 +01:00
flog_warn(
EC_ZEBRA_BAD_NHG_MESSAGE,
"Invalid Nexthop message received from the kernel with ID (%u)",
id);
return -1;
}
zebra: Use a nhe context dataplane and rib metaq We will use a nhe context for dataplane interaction with nextho group hash entries. New nhe's from the kernel will be put into a group array if they are a group and queued on the rib metaq to be processed later. New nhe's sent to the kernel will be set on the dataplane context with approprate ID's in the group array if needed. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 00:03:29 +02:00
SET_FLAG(nh.flags, NEXTHOP_FLAG_ACTIVE);
if (nhm->nh_flags & RTNH_F_ONLINK)
SET_FLAG(nh.flags, NEXTHOP_FLAG_ONLINK);
vrf_id = nh.vrf_id;
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
}
zebra: Sweep our nexthop objects out on restart On restart, if we failed to remove any nexthop objects due to a kill -9 or such event, sweep them if we aren't using them. Add a proto field to handle this and remove the is_kernel bool. Add a dupicate flag that indicates this nexthop group is only present in our ID hashtable. It is a dupicate nexthop we received from the kernel, therefore we cannot hash on it. Make the idcounter globally accessible so that kernel updates increment it as soon as we receive them, not when we handle them. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-01 20:07:04 +02:00
if (zebra_nhg_kernel_find(id, &nh, grp, grp_count, vrf_id, afi,
type, startup))
zebra: Use a nhe context dataplane and rib metaq We will use a nhe context for dataplane interaction with nextho group hash entries. New nhe's from the kernel will be put into a group array if they are a group and queued on the rib metaq to be processed later. New nhe's sent to the kernel will be set on the dataplane context with approprate ID's in the group array if needed. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-05-15 00:03:29 +02:00
return -1;
zebra: Refactor nexthop group creation code to use allocated memory Simplify the code for nexthop hash entry creation. I made nexthop hash entry creation expect the nexthop group and depends to always be allocated before lookup. Before, it was only allocated if it had dependencies. I think it makes the code a bit more readable to go ahead an allocate even for single nexthops as well. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-22 18:11:07 +01:00
zebra: Add handling for kernel del/update nexthop Add handling for delete/update nexthop object messages from the kernel. If someone deletes a nexthop object we are still using, send it back down. If the someone updates a nexthop we are using, replace that nexthop with ours. Routes are referencing this nexthop object ID and we resolved it ourselves, so we should force the other `someone` to submit to our will. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-01 20:24:35 +02:00
} else if (h->nlmsg_type == RTM_DELNEXTHOP)
zebra_nhg_kernel_del(id);
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
return 0;
}
/**
* netlink_request_nexthop() - Request nextop information from the kernel
* @zns: Zebra namespace
* @family: AF_* netlink family
* @type: RTM_* route type
*
* Return: Result status
*/
static int netlink_request_nexthop(struct zebra_ns *zns, int family, int type)
{
struct {
struct nlmsghdr n;
struct nhmsg nhm;
} req;
/* Form the request, specifying filter (rtattr) if needed. */
memset(&req, 0, sizeof(req));
req.n.nlmsg_type = type;
req.n.nlmsg_flags = NLM_F_ROOT | NLM_F_MATCH | NLM_F_REQUEST;
req.n.nlmsg_len = NLMSG_LENGTH(sizeof(struct nhmsg));
req.nhm.nh_family = family;
return netlink_request(&zns->netlink_cmd, &req.n);
}
zebra: Force off kernel nexthop group API for now Force off kernel nexthop group API for now. Will re-enable after suffient testing. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-28 22:15:17 +02:00
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
/**
* netlink_nexthop_read() - Nexthop read function using netlink interface
*
* @zns: Zebra name space
*
* Return: Result status
* Only called at bootstrap time.
*/
int netlink_nexthop_read(struct zebra_ns *zns)
{
int ret;
struct zebra_dplane_info dp_info;
zebra_dplane_info_from_zns(&dp_info, zns, true /*is_cmd*/);
/* Get nexthop objects */
ret = netlink_request_nexthop(zns, AF_UNSPEC, RTM_GETNEXTHOP);
if (ret < 0)
return ret;
ret = netlink_parse_info(netlink_nexthop_change, &zns->netlink_cmd,
&dp_info, 0, 1);
zebra: Move the supports_nh bool to a better place Move the supports_nh bool indicating whether the kernel we are using supports nexthop objects into the netlink kernel interface itself. Since only linux and netlink support nexthop object APIs for now this is fine. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-08-01 20:53:06 +02:00
if (!ret)
/* If we succesfully read in nexthop objects,
* this kernel must support them.
*/
supports_nh = true;
else if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug("Nexthop objects not supported on this kernel");
zebra: Return proper status result We were ignoring the status result interger from the netlink request and message parsing and just returning 0. Fixed this to return the result of the last one. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-03-06 20:10:08 +01:00
return ret;
zebra: Add functionality to parse RTM_NEWNEXTHOP and RTM_DELNEXTHOP messages Add the functionality to parse new nexthop group messages from the kernel and insert them into the appropriate hash tables. Parsing is done at startup between interface and interface address lookup. Add functionality to parse changes to nexthops we already have. Add functionality to parse delete nexthop messages from the kernel and remove them from our table. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-02-26 00:18:07 +01:00
}
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
int kernel_neigh_update(int add, int ifindex, uint32_t addr, char *lla,
zebra: ipv6 operations stick to namespace All ipv6 operations stick to namespace. Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
2017-12-19 12:23:32 +01:00
int llalen, ns_id_t ns_id)
zebra: create an abstraction layer for netlink_neigh_update() We shouldn't have platform-agnostic code (e.g. zebra/interface.c) calling platform-specific functions (e.g. netlink_neigh_update). This commit introduces the kernel_neigh_update() function, which then has to be implemented by all supported platforms. Currently only Linux implements this function, which is only used by the RTADV code. Signed-off-by: Renato Westphal <renato@opensourcerouting.org>
2016-08-04 15:07:32 +02:00
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return netlink_neigh_update(add ? RTM_NEWNEIGH : RTM_DELNEIGH, ifindex,
zebra: ipv6 operations stick to namespace All ipv6 operations stick to namespace. Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
2017-12-19 12:23:32 +01:00
addr, lla, llalen, ns_id);
zebra: create an abstraction layer for netlink_neigh_update() We shouldn't have platform-agnostic code (e.g. zebra/interface.c) calling platform-specific functions (e.g. netlink_neigh_update). This commit introduces the kernel_neigh_update() function, which then has to be implemented by all supported platforms. Currently only Linux implements this function, which is only used by the RTADV code. Signed-off-by: Renato Westphal <renato@opensourcerouting.org>
2016-08-04 15:07:32 +02:00
}
Initial revision
2002-12-13 21:15:29 +01:00
zebra: VNI and VTEP handling Implement fundamental handling for VNIs and VTEPs: - Handle EVPN enable/disable by client (advertise-all-vni) - Create/update/delete VNIs based on VxLAN interface events and inform client - Handle VTEP add/delete from client and install into kernel - New debug command for VxLAN/EVPN - kernel interface (Linux/netlink only) Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:38:26 +02:00
/*
* Add remote VTEP to the flood list for this VxLAN interface (VNI). This
* is done by adding an FDB entry with a MAC of 00:00:00:00:00:00.
*/
zebra: move EVPN VTEP programming to dataplane Move VTEP install/uninstall to the zebra dataplane. Remove synch kernel-facing apis and helper functions. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-08-26 21:44:54 +02:00
static int netlink_vxlan_flood_update_ctx(const struct zebra_dplane_ctx *ctx,
int cmd)
zebra: VNI and VTEP handling Implement fundamental handling for VNIs and VTEPs: - Handle EVPN enable/disable by client (advertise-all-vni) - Create/update/delete VNIs based on VxLAN interface events and inform client - Handle VTEP add/delete from client and install into kernel - New debug command for VxLAN/EVPN - kernel interface (Linux/netlink only) Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:38:26 +02:00
{
zebra: Send RTPROT_ZEBRA for netlink messages missing this data Update neighbor entries and rule entries to have the RTPROT_ZEBRA protocol value. So we can tell where things come from. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-09-19 22:05:39 +02:00
uint8_t protocol = RTPROT_ZEBRA;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
struct {
struct nlmsghdr n;
struct ndmsg ndm;
char buf[256];
} req;
*: use C99 standard fixed-width integer types The following types are nonstandard: - u_char - u_short - u_int - u_long - u_int8_t - u_int16_t - u_int32_t Replace them with the C99 standard types: - uint8_t - unsigned short - unsigned int - unsigned long - uint8_t - uint16_t - uint32_t Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-03-27 21:13:34 +02:00
uint8_t dst_mac[6] = {0x0, 0x0, 0x0, 0x0, 0x0, 0x0};
zebra: move EVPN VTEP programming to dataplane Move VTEP install/uninstall to the zebra dataplane. Remove synch kernel-facing apis and helper functions. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-08-26 21:44:54 +02:00
const struct ipaddr *addr;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
zebra: memset buf to prevent uninited writes into kernel Setup the buf used for extra data passed into kernel such that we are cleaning it out before writing data to it, so we can avoid writing uninited data. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-05-18 00:46:14 +02:00
memset(&req, 0, sizeof(req));
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
req.n.nlmsg_len = NLMSG_LENGTH(sizeof(struct ndmsg));
req.n.nlmsg_flags = NLM_F_REQUEST;
if (cmd == RTM_NEWNEIGH)
req.n.nlmsg_flags |= (NLM_F_CREATE | NLM_F_APPEND);
req.n.nlmsg_type = cmd;
req.ndm.ndm_family = PF_BRIDGE;
req.ndm.ndm_state = NUD_NOARP | NUD_PERMANENT;
zebra: Zebra nexthop group re-work checkpatch fixes Checkpatch fixes for the zebra nexthop group re-work. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-09-03 22:12:06 +02:00
req.ndm.ndm_flags |= NTF_SELF; /* Handle by "self", not "master" */
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
zebra: Send RTPROT_ZEBRA for netlink messages missing this data Update neighbor entries and rule entries to have the RTPROT_ZEBRA protocol value. So we can tell where things come from. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-09-19 22:05:39 +02:00
addattr_l(&req.n, sizeof(req),
NDA_PROTOCOL, &protocol, sizeof(protocol));
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
addattr_l(&req.n, sizeof(req), NDA_LLADDR, &dst_mac, 6);
zebra: move EVPN VTEP programming to dataplane Move VTEP install/uninstall to the zebra dataplane. Remove synch kernel-facing apis and helper functions. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-08-26 21:44:54 +02:00
req.ndm.ndm_ifindex = dplane_ctx_get_ifindex(ctx);
zebra: VNI and VTEP handling Implement fundamental handling for VNIs and VTEPs: - Handle EVPN enable/disable by client (advertise-all-vni) - Create/update/delete VNIs based on VxLAN interface events and inform client - Handle VTEP add/delete from client and install into kernel - New debug command for VxLAN/EVPN - kernel interface (Linux/netlink only) Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:38:26 +02:00
zebra: move EVPN VTEP programming to dataplane Move VTEP install/uninstall to the zebra dataplane. Remove synch kernel-facing apis and helper functions. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-08-26 21:44:54 +02:00
addr = dplane_ctx_neigh_get_ipaddr(ctx);
zebra: VNI and VTEP handling Implement fundamental handling for VNIs and VTEPs: - Handle EVPN enable/disable by client (advertise-all-vni) - Create/update/delete VNIs based on VxLAN interface events and inform client - Handle VTEP add/delete from client and install into kernel - New debug command for VxLAN/EVPN - kernel interface (Linux/netlink only) Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:38:26 +02:00
zebra: move EVPN VTEP programming to dataplane Move VTEP install/uninstall to the zebra dataplane. Remove synch kernel-facing apis and helper functions. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-08-26 21:44:54 +02:00
addattr_l(&req.n, sizeof(req), NDA_DST, &(addr->ipaddr_v4), 4);
zebra: VNI and VTEP handling Implement fundamental handling for VNIs and VTEPs: - Handle EVPN enable/disable by client (advertise-all-vni) - Create/update/delete VNIs based on VxLAN interface events and inform client - Handle VTEP add/delete from client and install into kernel - New debug command for VxLAN/EVPN - kernel interface (Linux/netlink only) Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:38:26 +02:00
zebra: move EVPN VTEP programming to dataplane Move VTEP install/uninstall to the zebra dataplane. Remove synch kernel-facing apis and helper functions. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-08-26 21:44:54 +02:00
return netlink_talk_info(netlink_talk_filter, &req.n,
dplane_ctx_get_ns(ctx), 0);
zebra: VNI and VTEP handling Implement fundamental handling for VNIs and VTEPs: - Handle EVPN enable/disable by client (advertise-all-vni) - Create/update/delete VNIs based on VxLAN interface events and inform client - Handle VTEP add/delete from client and install into kernel - New debug command for VxLAN/EVPN - kernel interface (Linux/netlink only) Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:38:26 +02:00
}
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
#ifndef NDA_RTA
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
#define NDA_RTA(r) \
((struct rtattr *)(((char *)(r)) + NLMSG_ALIGN(sizeof(struct ndmsg))))
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
#endif
zebra: Remove unnecessary function parameter The snl variable is no longer needed to be passed around, so remove it from the calling path. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-05-23 15:37:06 +02:00
static int netlink_macfdb_change(struct nlmsghdr *h, int len, ns_id_t ns_id)
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
struct ndmsg *ndm;
struct interface *ifp;
struct zebra_if *zif;
struct rtattr *tb[NDA_MAX + 1];
struct interface *br_if;
struct ethaddr mac;
vlanid_t vid = 0;
struct prefix vtep_ip;
int vid_present = 0, dst_present = 0;
char buf[ETHER_ADDR_STRLEN];
char vid_buf[20];
char dst_buf[30];
zebra: Use boolean for certain VxLAN-EVPN flags Use boolean variables instead of unsigned int for certain VxLAN-EVPN flags which are really used as boolean. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Anuradha Karuppiah <anuradhak@cumulusnetworks.com> Reviewed-by: Chirag Shah <chirag@cumulusnetworks.com> Ticket: CM-22288 Reviewed By: CCR-7832 Testing Done: Along with a subsequent, related commit
2018-09-10 19:13:20 +02:00
bool sticky;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
ndm = NLMSG_DATA(h);
zebra: Handle L3 multi-tenancy correctly for EVPN Fix following flaws that resulted in EVPN with L3 multi-tenancy (i.e., EVPN dealing with VxLAN routing in the presence of tenant VRFs) not working properly: 1. EVPN enable ("advertise-all-vni") is a global command, ensure it is accordingly processed. The config is maintained against the default VRF. 2. There was an incorrect attempt to derive the L3 VRF for L2 interfaces - the VRF only applies for L3 interfaces, though the code may initialize to the default value in other cases. 3. Functions to map (port, VLAN) to SVI or vice versa were incorrect - particularly, zvni_map_svi() since it was looking in the L3 VRF for "matching" L2 interface which it would never find. Fix. In addition, since the 'zebra_vrf *' parameter is not relevant in most places, it has been removed. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Mitesh Kanjariya <mitesh@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com> Ticket: CM-17840 Reviewed By: CCR-6685 Testing Done: evpn-smoke, various manual tests
2017-09-18 23:45:34 +02:00
/* We only process macfdb notifications if EVPN is enabled */
if (!is_evpn_enabled())
return 0;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
/* The interface should exist. */
zebra: ipv6 operations stick to namespace All ipv6 operations stick to namespace. Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
2017-12-19 12:23:32 +01:00
ifp = if_lookup_by_index_per_ns(zebra_ns_lookup(ns_id),
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
ndm->ndm_ifindex);
zebra: Add some debugs to neighbor entry processing When we get a neighbor entry in zebra we start processing it. Let's add some additional debugs to the processing so that when it bails out and we don't use the data, we know the reason. This should help in debugging the problems from why bgp does not appear to have data associated with a neighbor entry in the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-03-08 16:46:55 +01:00
if (!ifp || !ifp->info) {
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug("\t%s without associated interface: %u",
__PRETTY_FUNCTION__, ndm->ndm_ifindex);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return 0;
zebra: Add some debugs to neighbor entry processing When we get a neighbor entry in zebra we start processing it. Let's add some additional debugs to the processing so that when it bails out and we don't use the data, we know the reason. This should help in debugging the problems from why bgp does not appear to have data associated with a neighbor entry in the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-03-08 16:46:55 +01:00
}
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
/* The interface should be something we're interested in. */
zebra: Add some debugs to neighbor entry processing When we get a neighbor entry in zebra we start processing it. Let's add some additional debugs to the processing so that when it bails out and we don't use the data, we know the reason. This should help in debugging the problems from why bgp does not appear to have data associated with a neighbor entry in the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-03-08 16:46:55 +01:00
if (!IS_ZEBRA_IF_BRIDGE_SLAVE(ifp)) {
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug("\t%s Not interested in %s, not a slave",
__PRETTY_FUNCTION__, ifp->name);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return 0;
zebra: Add some debugs to neighbor entry processing When we get a neighbor entry in zebra we start processing it. Let's add some additional debugs to the processing so that when it bails out and we don't use the data, we know the reason. This should help in debugging the problems from why bgp does not appear to have data associated with a neighbor entry in the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-03-08 16:46:55 +01:00
}
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
/* Drop "permanent" entries. */
zebra: Add some debugs to neighbor entry processing When we get a neighbor entry in zebra we start processing it. Let's add some additional debugs to the processing so that when it bails out and we don't use the data, we know the reason. This should help in debugging the problems from why bgp does not appear to have data associated with a neighbor entry in the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-03-08 16:46:55 +01:00
if (ndm->ndm_state & NUD_PERMANENT) {
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug("\t%s Entry is PERMANENT, dropping",
__PRETTY_FUNCTION__);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return 0;
zebra: Add some debugs to neighbor entry processing When we get a neighbor entry in zebra we start processing it. Let's add some additional debugs to the processing so that when it bails out and we don't use the data, we know the reason. This should help in debugging the problems from why bgp does not appear to have data associated with a neighbor entry in the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-03-08 16:46:55 +01:00
}
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
zif = (struct zebra_if *)ifp->info;
if ((br_if = zif->brslave_info.br_if) == NULL) {
zebra: Add some debugs to neighbor entry processing When we get a neighbor entry in zebra we start processing it. Let's add some additional debugs to the processing so that when it bails out and we don't use the data, we know the reason. This should help in debugging the problems from why bgp does not appear to have data associated with a neighbor entry in the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-03-08 16:46:55 +01:00
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug(
"%s family %s IF %s(%u) brIF %u - no bridge master",
nl_msg_type_to_str(h->nlmsg_type),
nl_family_to_str(ndm->ndm_family), ifp->name,
ndm->ndm_ifindex,
zif->brslave_info.bridge_ifindex);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return 0;
}
/* Parse attributes and extract fields of interest. */
memset(tb, 0, sizeof tb);
netlink_parse_rtattr(tb, NDA_MAX, NDA_RTA(ndm), len);
if (!tb[NDA_LLADDR]) {
zebra: Add some debugs to neighbor entry processing When we get a neighbor entry in zebra we start processing it. Let's add some additional debugs to the processing so that when it bails out and we don't use the data, we know the reason. This should help in debugging the problems from why bgp does not appear to have data associated with a neighbor entry in the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-03-08 16:46:55 +01:00
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug("%s family %s IF %s(%u) brIF %u - no LLADDR",
nl_msg_type_to_str(h->nlmsg_type),
nl_family_to_str(ndm->ndm_family), ifp->name,
ndm->ndm_ifindex,
zif->brslave_info.bridge_ifindex);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return 0;
}
zebra: Switch to using ETH_ALEN Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-08-03 14:43:56 +02:00
if (RTA_PAYLOAD(tb[NDA_LLADDR]) != ETH_ALEN) {
zebra: Add some debugs to neighbor entry processing When we get a neighbor entry in zebra we start processing it. Let's add some additional debugs to the processing so that when it bails out and we don't use the data, we know the reason. This should help in debugging the problems from why bgp does not appear to have data associated with a neighbor entry in the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-03-08 16:46:55 +01:00
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug(
"%s family %s IF %s(%u) brIF %u - LLADDR is not MAC, len %lu",
nl_msg_type_to_str(h->nlmsg_type),
nl_family_to_str(ndm->ndm_family), ifp->name,
ndm->ndm_ifindex,
zif->brslave_info.bridge_ifindex,
(unsigned long)RTA_PAYLOAD(tb[NDA_LLADDR]));
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return 0;
}
zebra: Switch to using ETH_ALEN Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-08-03 14:43:56 +02:00
memcpy(&mac, RTA_DATA(tb[NDA_LLADDR]), ETH_ALEN);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if ((NDA_VLAN <= NDA_MAX) && tb[NDA_VLAN]) {
vid_present = 1;
*: use C99 standard fixed-width integer types The following types are nonstandard: - u_char - u_short - u_int - u_long - u_int8_t - u_int16_t - u_int32_t Replace them with the C99 standard types: - uint8_t - unsigned short - unsigned int - unsigned long - uint8_t - uint16_t - uint32_t Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-03-27 21:13:34 +02:00
vid = *(uint16_t *)RTA_DATA(tb[NDA_VLAN]);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
sprintf(vid_buf, " VLAN %u", vid);
}
if (tb[NDA_DST]) {
/* TODO: Only IPv4 supported now. */
dst_present = 1;
vtep_ip.family = AF_INET;
vtep_ip.prefixlen = IPV4_MAX_BITLEN;
memcpy(&(vtep_ip.u.prefix4.s_addr), RTA_DATA(tb[NDA_DST]),
IPV4_MAX_BYTELEN);
sprintf(dst_buf, " dst %s", inet_ntoa(vtep_ip.u.prefix4));
}
zebra: Use boolean for certain VxLAN-EVPN flags Use boolean variables instead of unsigned int for certain VxLAN-EVPN flags which are really used as boolean. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Anuradha Karuppiah <anuradhak@cumulusnetworks.com> Reviewed-by: Chirag Shah <chirag@cumulusnetworks.com> Ticket: CM-22288 Reviewed By: CCR-7832 Testing Done: Along with a subsequent, related commit
2018-09-10 19:13:20 +02:00
sticky = !!(ndm->ndm_state & NUD_NOARP);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug("Rx %s family %s IF %s(%u)%s %sMAC %s%s",
nl_msg_type_to_str(h->nlmsg_type),
nl_family_to_str(ndm->ndm_family), ifp->name,
ndm->ndm_ifindex, vid_present ? vid_buf : "",
sticky ? "sticky " : "",
prefix_mac2str(&mac, buf, sizeof(buf)),
dst_present ? dst_buf : "");
zebra: Add some debugs to neighbor entry processing When we get a neighbor entry in zebra we start processing it. Let's add some additional debugs to the processing so that when it bails out and we don't use the data, we know the reason. This should help in debugging the problems from why bgp does not appear to have data associated with a neighbor entry in the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-03-08 16:46:55 +01:00
if (filter_vlan && vid != filter_vlan) {
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug("\tFiltered due to filter vlan: %d",
filter_vlan);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return 0;
zebra: Add some debugs to neighbor entry processing When we get a neighbor entry in zebra we start processing it. Let's add some additional debugs to the processing so that when it bails out and we don't use the data, we know the reason. This should help in debugging the problems from why bgp does not appear to have data associated with a neighbor entry in the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-03-08 16:46:55 +01:00
}
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
/* If add or update, do accordingly if learnt on a "local" interface; if
* the notification is over VxLAN, this has to be related to
* multi-homing,
* so perform an implicit delete of any local entry (if it exists).
*/
if (h->nlmsg_type == RTM_NEWNEIGH) {
if (IS_ZEBRA_IF_VXLAN(ifp))
return zebra_vxlan_check_del_local_mac(ifp, br_if, &mac,
vid);
return zebra_vxlan_local_mac_add_update(ifp, br_if, &mac, vid,
sticky);
}
/* This is a delete notification.
* 1. For a MAC over VxLan, check if it needs to be refreshed(readded)
* 2. For a MAC over "local" interface, delete the mac
* Note: We will get notifications from both bridge driver and VxLAN
* driver.
* Ignore the notification from VxLan driver as it is also generated
* when mac moves from remote to local.
*/
zebra: Add some debugs to neighbor entry processing When we get a neighbor entry in zebra we start processing it. Let's add some additional debugs to the processing so that when it bails out and we don't use the data, we know the reason. This should help in debugging the problems from why bgp does not appear to have data associated with a neighbor entry in the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-03-08 16:46:55 +01:00
if (dst_present) {
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug("\tNo Destination Present");
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return 0;
zebra: Add some debugs to neighbor entry processing When we get a neighbor entry in zebra we start processing it. Let's add some additional debugs to the processing so that when it bails out and we don't use the data, we know the reason. This should help in debugging the problems from why bgp does not appear to have data associated with a neighbor entry in the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-03-08 16:46:55 +01:00
}
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (IS_ZEBRA_IF_VXLAN(ifp))
return zebra_vxlan_check_readd_remote_mac(ifp, br_if, &mac,
vid);
return zebra_vxlan_local_mac_del(ifp, br_if, &mac, vid);
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
}
zebra: Remove unnecessary function parameter The snl variable is no longer needed to be passed around, so remove it from the calling path. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-05-23 15:37:06 +02:00
static int netlink_macfdb_table(struct nlmsghdr *h, ns_id_t ns_id, int startup)
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
int len;
struct ndmsg *ndm;
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (h->nlmsg_type != RTM_NEWNEIGH)
return 0;
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
/* Length validity. */
len = h->nlmsg_len - NLMSG_LENGTH(sizeof(struct ndmsg));
if (len < 0)
return -1;
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
/* We are interested only in AF_BRIDGE notifications. */
ndm = NLMSG_DATA(h);
if (ndm->ndm_family != AF_BRIDGE)
return 0;
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
zebra: Remove unnecessary function parameter The snl variable is no longer needed to be passed around, so remove it from the calling path. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-05-23 15:37:06 +02:00
return netlink_macfdb_change(h, len, ns_id);
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
}
/* Request for MAC FDB information from the kernel */
zebra: Start abstraction of zebra_dplane_info for context passing Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Signed-off-by: Mark Stapp <mjs@voltanet.io> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-09-12 20:59:57 +02:00
static int netlink_request_macs(struct nlsock *netlink_cmd, int family,
int type, ifindex_t master_ifindex)
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
struct {
struct nlmsghdr n;
struct ifinfomsg ifm;
char buf[256];
} req;
/* Form the request, specifying filter (rtattr) if needed. */
memset(&req, 0, sizeof(req));
req.n.nlmsg_type = type;
zebra: netlink_request consolidataion Make netlink_request api generic where it can be used for dump or querying specific information request. nelink request nlm flags (NLM_F_ROOT | NLM_F_MATCH) are used to dump purpose, if client wants to query spcific MAC or IP using netlink_request does not require to set them. nlm struct is passed by the caller of netlink_request, it can also set the nlm request flags. Signed-off-by: Chirag Shah <chirag@cumulusnetworks.com>
2018-12-11 06:33:16 +01:00
req.n.nlmsg_flags = NLM_F_ROOT | NLM_F_MATCH | NLM_F_REQUEST;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
req.n.nlmsg_len = NLMSG_LENGTH(sizeof(struct ifinfomsg));
req.ifm.ifi_family = family;
if (master_ifindex)
addattr32(&req.n, sizeof(req), IFLA_MASTER, master_ifindex);
zebra: Start abstraction of zebra_dplane_info for context passing Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Signed-off-by: Mark Stapp <mjs@voltanet.io> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-09-12 20:59:57 +02:00
return netlink_request(netlink_cmd, &req.n);
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
}
/*
* MAC forwarding database read using netlink interface. This is invoked
* at startup.
*/
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
int netlink_macfdb_read(struct zebra_ns *zns)
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
int ret;
zebra: Start abstraction of zebra_dplane_info for context passing Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Signed-off-by: Mark Stapp <mjs@voltanet.io> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-09-12 20:59:57 +02:00
struct zebra_dplane_info dp_info;
zebra_dplane_info_from_zns(&dp_info, zns, true /*is_cmd*/);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
/* Get bridge FDB table. */
zebra: Start abstraction of zebra_dplane_info for context passing Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Signed-off-by: Mark Stapp <mjs@voltanet.io> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-09-12 20:59:57 +02:00
ret = netlink_request_macs(&zns->netlink_cmd, AF_BRIDGE, RTM_GETNEIGH,
0);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (ret < 0)
return ret;
/* We are reading entire table. */
filter_vlan = 0;
zebra: Start abstraction of zebra_dplane_info for context passing Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Signed-off-by: Mark Stapp <mjs@voltanet.io> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-09-12 20:59:57 +02:00
ret = netlink_parse_info(netlink_macfdb_table, &zns->netlink_cmd,
&dp_info, 0, 1);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return ret;
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
}
/*
* MAC forwarding database read using netlink interface. This is for a
* specific bridge and matching specific access VLAN (if VLAN-aware bridge).
*/
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
int netlink_macfdb_read_for_bridge(struct zebra_ns *zns, struct interface *ifp,
struct interface *br_if)
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
struct zebra_if *br_zif;
struct zebra_if *zif;
struct zebra_l2info_vxlan *vxl;
zebra: Start abstraction of zebra_dplane_info for context passing Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Signed-off-by: Mark Stapp <mjs@voltanet.io> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-09-12 20:59:57 +02:00
struct zebra_dplane_info dp_info;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
int ret = 0;
zebra: Start abstraction of zebra_dplane_info for context passing Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Signed-off-by: Mark Stapp <mjs@voltanet.io> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-09-12 20:59:57 +02:00
zebra_dplane_info_from_zns(&dp_info, zns, true /*is_cmd*/);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
/* Save VLAN we're filtering on, if needed. */
br_zif = (struct zebra_if *)br_if->info;
zif = (struct zebra_if *)ifp->info;
vxl = &zif->l2info.vxl;
if (IS_ZEBRA_IF_BRIDGE_VLAN_AWARE(br_zif))
filter_vlan = vxl->access_vlan;
/* Get bridge FDB table for specific bridge - we do the VLAN filtering.
*/
zebra: Start abstraction of zebra_dplane_info for context passing Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Signed-off-by: Mark Stapp <mjs@voltanet.io> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-09-12 20:59:57 +02:00
ret = netlink_request_macs(&zns->netlink_cmd, AF_BRIDGE, RTM_GETNEIGH,
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
br_if->ifindex);
if (ret < 0)
return ret;
zebra: Start abstraction of zebra_dplane_info for context passing Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Signed-off-by: Mark Stapp <mjs@voltanet.io> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-09-12 20:59:57 +02:00
ret = netlink_parse_info(netlink_macfdb_table, &zns->netlink_cmd,
&dp_info, 0, 0);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
/* Reset VLAN filter. */
filter_vlan = 0;
return ret;
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
}
zebra: evpn remote delete fetch local entry An EVPN type-2 entry is in freeze state during remote update, remote VTEP can send typ-2 withdraw update, upon receiving an entry delete (withdraw), first check kernel has in local reachable state. Upon unfreeze use the local entry to advertise to peers. Fetch is for both MAC and IP, delete can come for only MAC or MAC-IP combined route. The specific entry fetch only required request flag to be set, dump flag is not required. Testing Done: Simulate two VTEPs to do M1, IP1 mobility sequence, freeze MAC during remote MAC update, subsequently send withdraw type-2 route from origintating VTEP. This results in read apis to invoke for local reachable entry. Zebra updates its cache and upon unfreeze originates type-2. Signed-off-by: Chirag Shah <chirag@cumulusnetworks.com>
2018-12-11 06:33:16 +01:00
/* Request for MAC FDB for a specific MAC address in VLAN from the kernel */
static int netlink_request_specific_mac_in_bridge(struct zebra_ns *zns,
int family,
int type,
struct interface *br_if,
struct ethaddr *mac,
vlanid_t vid)
{
struct {
struct nlmsghdr n;
struct ndmsg ndm;
char buf[256];
} req;
struct zebra_if *br_zif;
char buf[ETHER_ADDR_STRLEN];
memset(&req, 0, sizeof(req));
req.n.nlmsg_len = NLMSG_LENGTH(sizeof(struct ndmsg));
req.n.nlmsg_type = type; /* RTM_GETNEIGH */
req.n.nlmsg_flags = NLM_F_REQUEST;
req.ndm.ndm_family = family; /* AF_BRIDGE */
/* req.ndm.ndm_state = NUD_REACHABLE; */
addattr_l(&req.n, sizeof(req), NDA_LLADDR, mac, 6);
br_zif = (struct zebra_if *)br_if->info;
if (IS_ZEBRA_IF_BRIDGE_VLAN_AWARE(br_zif) && vid > 0)
addattr16(&req.n, sizeof(req), NDA_VLAN, vid);
addattr32(&req.n, sizeof(req), NDA_MASTER, br_if->ifindex);
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug("%s: Tx family %s IF %s(%u) MAC %s vid %u",
__PRETTY_FUNCTION__,
nl_family_to_str(req.ndm.ndm_family), br_if->name,
br_if->ifindex,
prefix_mac2str(mac, buf, sizeof(buf)), vid);
return netlink_request(&zns->netlink_cmd, &req.n);
}
int netlink_macfdb_read_specific_mac(struct zebra_ns *zns,
struct interface *br_if,
struct ethaddr *mac, vlanid_t vid)
{
int ret = 0;
struct zebra_dplane_info dp_info;
zebra_dplane_info_from_zns(&dp_info, zns, true /*is_cmd*/);
/* Get bridge FDB table for specific bridge - we do the VLAN filtering.
*/
ret = netlink_request_specific_mac_in_bridge(zns, AF_BRIDGE,
RTM_GETNEIGH,
br_if, mac, vid);
if (ret < 0)
return ret;
ret = netlink_parse_info(netlink_macfdb_table, &zns->netlink_cmd,
&dp_info, 1, 0);
return ret;
}
zebra: use dataplane for vxlan remote mac programming Move vxlan remote MAC install and uninstall to the async dataplane. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-07-30 17:54:07 +02:00
/*
* Netlink-specific handler for MAC updates using dataplane context object.
*/
static enum zebra_dplane_result
netlink_macfdb_update_ctx(struct zebra_dplane_ctx *ctx)
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
{
zebra: Send RTPROT_ZEBRA for netlink messages missing this data Update neighbor entries and rule entries to have the RTPROT_ZEBRA protocol value. So we can tell where things come from. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-09-19 22:05:39 +02:00
uint8_t protocol = RTPROT_ZEBRA;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
struct {
struct nlmsghdr n;
struct ndmsg ndm;
char buf[256];
} req;
zebra: use dataplane for vxlan remote mac programming Move vxlan remote MAC install and uninstall to the async dataplane. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-07-30 17:54:07 +02:00
int ret;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
int dst_alen;
zebra: dead code (Coverity 1473593) Conditional code in netlink_macfdb_update() introduced in 2232a77c used the 'dst_present' variable because not all cases were covered. Now it is not necessary. Signed-off-by: F. Aragon <paco@voltanet.io>
2018-09-25 16:49:34 +02:00
int vid_present = 0;
zebra: use dataplane for vxlan remote mac programming Move vxlan remote MAC install and uninstall to the async dataplane. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-07-30 17:54:07 +02:00
int cmd;
struct in_addr vtep_ip;
vlanid_t vid;
if (dplane_ctx_get_op(ctx) == DPLANE_OP_MAC_INSTALL)
cmd = RTM_NEWNEIGH;
else
cmd = RTM_DELNEIGH;
zebra: memset buf to prevent uninited writes into kernel Setup the buf used for extra data passed into kernel such that we are cleaning it out before writing data to it, so we can avoid writing uninited data. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-05-18 00:46:14 +02:00
memset(&req, 0, sizeof(req));
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
req.n.nlmsg_len = NLMSG_LENGTH(sizeof(struct ndmsg));
req.n.nlmsg_flags = NLM_F_REQUEST;
if (cmd == RTM_NEWNEIGH)
req.n.nlmsg_flags |= (NLM_F_CREATE | NLM_F_REPLACE);
req.n.nlmsg_type = cmd;
req.ndm.ndm_family = AF_BRIDGE;
req.ndm.ndm_flags |= NTF_SELF | NTF_MASTER;
req.ndm.ndm_state = NUD_REACHABLE;
zebra: use dataplane for vxlan remote mac programming Move vxlan remote MAC install and uninstall to the async dataplane. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-07-30 17:54:07 +02:00
if (dplane_ctx_mac_is_sticky(ctx))
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
req.ndm.ndm_state |= NUD_NOARP;
else
req.ndm.ndm_flags |= NTF_EXT_LEARNED;
zebra: Send RTPROT_ZEBRA for netlink messages missing this data Update neighbor entries and rule entries to have the RTPROT_ZEBRA protocol value. So we can tell where things come from. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-09-19 22:05:39 +02:00
addattr_l(&req.n, sizeof(req),
NDA_PROTOCOL, &protocol, sizeof(protocol));
zebra: use dataplane for vxlan remote mac programming Move vxlan remote MAC install and uninstall to the async dataplane. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-07-30 17:54:07 +02:00
addattr_l(&req.n, sizeof(req), NDA_LLADDR,
dplane_ctx_mac_get_addr(ctx), 6);
zebra: avoid using zebra datastructs in evpn dataplane path Some netlink-facing code used for evpn/vxlan programming was being run in the dataplane pthread, but accessing zebra core datastructs. Move some additional data into the dataplane context, and use it in the netlink path instead. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-09-05 18:58:58 +02:00
req.ndm.ndm_ifindex = dplane_ctx_get_ifindex(ctx);
zebra: use dataplane for vxlan remote mac programming Move vxlan remote MAC install and uninstall to the async dataplane. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-07-30 17:54:07 +02:00
zebra: Remove unused parameter in MAC delete When a MAC moves from local to remote, a replace is allowed, EVPN no longer has to delete the local MAC before installing the remote MAC. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Anuradha Karuppiah <anuradhak@cumulusnetworks.com> Reviewed-by: Chirag Shah <chirag@cumulusnetworks.com>
2018-08-29 02:02:40 +02:00
dst_alen = 4; // TODO: hardcoded
zebra: use dataplane for vxlan remote mac programming Move vxlan remote MAC install and uninstall to the async dataplane. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-07-30 17:54:07 +02:00
vtep_ip = *(dplane_ctx_mac_get_vtep_ip(ctx));
zebra: Remove unused parameter in MAC delete When a MAC moves from local to remote, a replace is allowed, EVPN no longer has to delete the local MAC before installing the remote MAC. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Anuradha Karuppiah <anuradhak@cumulusnetworks.com> Reviewed-by: Chirag Shah <chirag@cumulusnetworks.com>
2018-08-29 02:02:40 +02:00
addattr_l(&req.n, sizeof(req), NDA_DST, &vtep_ip, dst_alen);
zebra: use dataplane for vxlan remote mac programming Move vxlan remote MAC install and uninstall to the async dataplane. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-07-30 17:54:07 +02:00
zebra: avoid using zebra datastructs in evpn dataplane path Some netlink-facing code used for evpn/vxlan programming was being run in the dataplane pthread, but accessing zebra core datastructs. Move some additional data into the dataplane context, and use it in the netlink path instead. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-09-05 18:58:58 +02:00
vid = dplane_ctx_mac_get_vlan(ctx);
if (vid > 0) {
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
addattr16(&req.n, sizeof(req), NDA_VLAN, vid);
vid_present = 1;
}
zebra: avoid using zebra datastructs in evpn dataplane path Some netlink-facing code used for evpn/vxlan programming was being run in the dataplane pthread, but accessing zebra core datastructs. Move some additional data into the dataplane context, and use it in the netlink path instead. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-09-05 18:58:58 +02:00
addattr32(&req.n, sizeof(req), NDA_MASTER,
dplane_ctx_mac_get_br_ifindex(ctx));
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
zebra: use dataplane for vxlan remote mac programming Move vxlan remote MAC install and uninstall to the async dataplane. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-07-30 17:54:07 +02:00
if (IS_ZEBRA_DEBUG_KERNEL) {
char ipbuf[PREFIX_STRLEN];
char buf[ETHER_ADDR_STRLEN];
char dst_buf[PREFIX_STRLEN + 10];
zebra: avoid using zebra datastructs in evpn dataplane path Some netlink-facing code used for evpn/vxlan programming was being run in the dataplane pthread, but accessing zebra core datastructs. Move some additional data into the dataplane context, and use it in the netlink path instead. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-09-05 18:58:58 +02:00
char vid_buf[20];
if (vid_present)
snprintf(vid_buf, sizeof(vid_buf), " VLAN %u", vid);
else
vid_buf[0] = '\0';
zebra: use dataplane for vxlan remote mac programming Move vxlan remote MAC install and uninstall to the async dataplane. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-07-30 17:54:07 +02:00
inet_ntop(AF_INET, &vtep_ip, ipbuf, sizeof(ipbuf));
snprintf(dst_buf, sizeof(dst_buf), " dst %s", ipbuf);
prefix_mac2str(dplane_ctx_mac_get_addr(ctx), buf, sizeof(buf));
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
zlog_debug("Tx %s family %s IF %s(%u)%s %sMAC %s%s",
nl_msg_type_to_str(cmd),
zebra: avoid using zebra datastructs in evpn dataplane path Some netlink-facing code used for evpn/vxlan programming was being run in the dataplane pthread, but accessing zebra core datastructs. Move some additional data into the dataplane context, and use it in the netlink path instead. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-09-05 18:58:58 +02:00
nl_family_to_str(req.ndm.ndm_family),
dplane_ctx_get_ifname(ctx),
dplane_ctx_get_ifindex(ctx), vid_buf,
zebra: use dataplane for vxlan remote mac programming Move vxlan remote MAC install and uninstall to the async dataplane. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-07-30 17:54:07 +02:00
dplane_ctx_mac_is_sticky(ctx) ? "sticky " : "",
buf, dst_buf);
}
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
zebra: use dataplane for vxlan remote mac programming Move vxlan remote MAC install and uninstall to the async dataplane. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-07-30 17:54:07 +02:00
ret = netlink_talk_info(netlink_talk_filter, &req.n,
dplane_ctx_get_ns(ctx), 0);
if (ret == 0)
return ZEBRA_DPLANE_REQUEST_SUCCESS;
else
return ZEBRA_DPLANE_REQUEST_FAILURE;
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
}
zebra: handle neighbor delete from kernel for ipv4 link-local Problem reported that kernel neighbor entries could end up in "FAILED" state when the neighbor entry was deleted. This fix handles the notification of the event from netlink messages and re-inserts the deleted entry. Signed-off-by: Don Slice <dslice@cumulusnetworks.com>
2018-11-28 16:57:01 +01:00
/*
* In the event the kernel deletes ipv4 link-local neighbor entries created for
* 5549 support, re-install them.
*/
static void netlink_handle_5549(struct ndmsg *ndm, struct zebra_if *zif,
zebra: failed neighbor event logging was a bit too aggresive The failed neighbor event logging that was recently added in commit: 3acae086bacea16fdcdd3987f092d678d56bc356 cast a bit too broad of a stroke. We should only inform the user that we were ignoring the RTM_NEWNEIGH FAIL callback when we believe it was one of our own 5549 entries. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-06-20 11:37:17 +02:00
struct interface *ifp, struct ipaddr *ip,
bool handle_failed)
zebra: handle neighbor delete from kernel for ipv4 link-local Problem reported that kernel neighbor entries could end up in "FAILED" state when the neighbor entry was deleted. This fix handles the notification of the event from netlink messages and re-inserts the deleted entry. Signed-off-by: Don Slice <dslice@cumulusnetworks.com>
2018-11-28 16:57:01 +01:00
{
if (ndm->ndm_family != AF_INET)
return;
if (!zif->v6_2_v4_ll_neigh_entry)
return;
if (ipv4_ll.s_addr != ip->ip._v4_addr.s_addr)
return;
zebra: failed neighbor event logging was a bit too aggresive The failed neighbor event logging that was recently added in commit: 3acae086bacea16fdcdd3987f092d678d56bc356 cast a bit too broad of a stroke. We should only inform the user that we were ignoring the RTM_NEWNEIGH FAIL callback when we believe it was one of our own 5549 entries. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-06-20 11:37:17 +02:00
if (handle_failed && ndm->ndm_state & NUD_FAILED) {
zlog_info("Neighbor Entry for %s has entered a failed state, not reinstalling",
ifp->name);
return;
}
zebra: handle neighbor delete from kernel for ipv4 link-local Problem reported that kernel neighbor entries could end up in "FAILED" state when the neighbor entry was deleted. This fix handles the notification of the event from netlink messages and re-inserts the deleted entry. Signed-off-by: Don Slice <dslice@cumulusnetworks.com>
2018-11-28 16:57:01 +01:00
if_nbr_ipv6ll_to_ipv4ll_neigh_update(ifp, &zif->v6_2_v4_ll_addr6, true);
}
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
#define NUD_VALID \
(NUD_PERMANENT | NUD_NOARP | NUD_REACHABLE | NUD_PROBE | NUD_STALE \
| NUD_DELAY)
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
zebra: Remove unnecessary function parameter The snl variable is no longer needed to be passed around, so remove it from the calling path. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-05-23 15:37:06 +02:00
static int netlink_ipneigh_change(struct nlmsghdr *h, int len, ns_id_t ns_id)
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
struct ndmsg *ndm;
struct interface *ifp;
struct zebra_if *zif;
struct rtattr *tb[NDA_MAX + 1];
struct interface *link_if;
struct ethaddr mac;
struct ipaddr ip;
char buf[ETHER_ADDR_STRLEN];
char buf2[INET6_ADDRSTRLEN];
int mac_present = 0;
zebra: Use boolean for certain VxLAN-EVPN flags Use boolean variables instead of unsigned int for certain VxLAN-EVPN flags which are really used as boolean. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Anuradha Karuppiah <anuradhak@cumulusnetworks.com> Reviewed-by: Chirag Shah <chirag@cumulusnetworks.com> Ticket: CM-22288 Reviewed By: CCR-7832 Testing Done: Along with a subsequent, related commit
2018-09-10 19:13:20 +02:00
bool is_ext;
bool is_router;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
ndm = NLMSG_DATA(h);
/* The interface should exist. */
zebra: ipv6 operations stick to namespace All ipv6 operations stick to namespace. Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
2017-12-19 12:23:32 +01:00
ifp = if_lookup_by_index_per_ns(zebra_ns_lookup(ns_id),
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
ndm->ndm_ifindex);
zebra: Handle L3 multi-tenancy correctly for EVPN Fix following flaws that resulted in EVPN with L3 multi-tenancy (i.e., EVPN dealing with VxLAN routing in the presence of tenant VRFs) not working properly: 1. EVPN enable ("advertise-all-vni") is a global command, ensure it is accordingly processed. The config is maintained against the default VRF. 2. There was an incorrect attempt to derive the L3 VRF for L2 interfaces - the VRF only applies for L3 interfaces, though the code may initialize to the default value in other cases. 3. Functions to map (port, VLAN) to SVI or vice versa were incorrect - particularly, zvni_map_svi() since it was looking in the L3 VRF for "matching" L2 interface which it would never find. Fix. In addition, since the 'zebra_vrf *' parameter is not relevant in most places, it has been removed. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Mitesh Kanjariya <mitesh@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com> Ticket: CM-17840 Reviewed By: CCR-6685 Testing Done: evpn-smoke, various manual tests
2017-09-18 23:45:34 +02:00
if (!ifp || !ifp->info)
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return 0;
zebra: Notice when our neighbor entry is removed and fight back Notice when someone deletes a neighbor entry we've put in for rfc-5549 gets deleted by some evil evil person. When this happens notice and push it back in, immediately. Ticket: CM-18612 Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-04-09 14:04:39 +02:00
zif = (struct zebra_if *)ifp->info;
/* Parse attributes and extract fields of interest. */
memset(tb, 0, sizeof tb);
netlink_parse_rtattr(tb, NDA_MAX, NDA_RTA(ndm), len);
if (!tb[NDA_DST]) {
zebra: flog_warn conversion Convert Zebra to user error subsystem. Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-08-16 22:10:32 +02:00
zlog_debug("%s family %s IF %s(%u) - no DST",
nl_msg_type_to_str(h->nlmsg_type),
nl_family_to_str(ndm->ndm_family), ifp->name,
ndm->ndm_ifindex);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return 0;
zebra: Notice when our neighbor entry is removed and fight back Notice when someone deletes a neighbor entry we've put in for rfc-5549 gets deleted by some evil evil person. When this happens notice and push it back in, immediately. Ticket: CM-18612 Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-04-09 14:04:39 +02:00
}
memset(&ip, 0, sizeof(struct ipaddr));
ip.ipa_type = (ndm->ndm_family == AF_INET) ? IPADDR_V4 : IPADDR_V6;
memcpy(&ip.ip.addr, RTA_DATA(tb[NDA_DST]), RTA_PAYLOAD(tb[NDA_DST]));
zebra: handle neighbor delete from kernel for ipv4 link-local Problem reported that kernel neighbor entries could end up in "FAILED" state when the neighbor entry was deleted. This fix handles the notification of the event from netlink messages and re-inserts the deleted entry. Signed-off-by: Don Slice <dslice@cumulusnetworks.com>
2018-11-28 16:57:01 +01:00
/* if kernel deletes our rfc5549 neighbor entry, re-install it */
if (h->nlmsg_type == RTM_DELNEIGH && (ndm->ndm_state & NUD_PERMANENT)) {
zebra: failed neighbor event logging was a bit too aggresive The failed neighbor event logging that was recently added in commit: 3acae086bacea16fdcdd3987f092d678d56bc356 cast a bit too broad of a stroke. We should only inform the user that we were ignoring the RTM_NEWNEIGH FAIL callback when we believe it was one of our own 5549 entries. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-06-20 11:37:17 +02:00
netlink_handle_5549(ndm, zif, ifp, &ip, false);
zebra: Add some debugs to neighbor entry processing When we get a neighbor entry in zebra we start processing it. Let's add some additional debugs to the processing so that when it bails out and we don't use the data, we know the reason. This should help in debugging the problems from why bgp does not appear to have data associated with a neighbor entry in the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-03-08 16:46:55 +01:00
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug(
"\tNeighbor Entry Received is a 5549 entry, finished");
zebra: Notice when our neighbor entry is removed and fight back Notice when someone deletes a neighbor entry we've put in for rfc-5549 gets deleted by some evil evil person. When this happens notice and push it back in, immediately. Ticket: CM-18612 Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-04-09 14:04:39 +02:00
return 0;
}
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
zebra: handle neighbor delete from kernel for ipv4 link-local Problem reported that kernel neighbor entries could end up in "FAILED" state when the neighbor entry was deleted. This fix handles the notification of the event from netlink messages and re-inserts the deleted entry. Signed-off-by: Don Slice <dslice@cumulusnetworks.com>
2018-11-28 16:57:01 +01:00
/* if kernel marks our rfc5549 neighbor entry invalid, re-install it */
zebra: failed neighbor event logging was a bit too aggresive The failed neighbor event logging that was recently added in commit: 3acae086bacea16fdcdd3987f092d678d56bc356 cast a bit too broad of a stroke. We should only inform the user that we were ignoring the RTM_NEWNEIGH FAIL callback when we believe it was one of our own 5549 entries. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-06-20 11:37:17 +02:00
if (h->nlmsg_type == RTM_NEWNEIGH && !(ndm->ndm_state & NUD_VALID))
netlink_handle_5549(ndm, zif, ifp, &ip, true);
zebra: handle neighbor delete from kernel for ipv4 link-local Problem reported that kernel neighbor entries could end up in "FAILED" state when the neighbor entry was deleted. This fix handles the notification of the event from netlink messages and re-inserts the deleted entry. Signed-off-by: Don Slice <dslice@cumulusnetworks.com>
2018-11-28 16:57:01 +01:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
/* The neighbor is present on an SVI. From this, we locate the
* underlying
* bridge because we're only interested in neighbors on a VxLAN bridge.
* The bridge is located based on the nature of the SVI:
* (a) In the case of a VLAN-aware bridge, the SVI is a L3 VLAN
* interface
* and is linked to the bridge
* (b) In the case of a VLAN-unaware bridge, the SVI is the bridge
* inteface
* itself
*/
if (IS_ZEBRA_IF_VLAN(ifp)) {
zebra: ipv6 operations stick to namespace All ipv6 operations stick to namespace. Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
2017-12-19 12:23:32 +01:00
link_if = if_lookup_by_index_per_ns(zebra_ns_lookup(ns_id),
zebra: lookup link by ifindex Frr has an assumption that when interface A links to B, we already know about B. But that might be true always. It is probably purely depends on the configuration and how the interfaces are hashed in Kernel. FRR seems to sometimes get "A is linked to B" before it knows about B, in that case, the linkage between the data structure for A & B won't be proper. Ticket: CM-17679 Review: ccr-6628 Testing: Manual Signed-off-by: Mitesh Kanjariya <mitesh@cumulusnetworks.com>
2017-08-20 02:28:58 +02:00
zif->link_ifindex);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (!link_if)
return 0;
} else if (IS_ZEBRA_IF_BRIDGE(ifp))
link_if = ifp;
zebra: Add some debugs to neighbor entry processing When we get a neighbor entry in zebra we start processing it. Let's add some additional debugs to the processing so that when it bails out and we don't use the data, we know the reason. This should help in debugging the problems from why bgp does not appear to have data associated with a neighbor entry in the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-03-08 16:46:55 +01:00
else {
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug(
"\tNeighbor Entry received is not on a VLAN or a BRIDGE, ignoring");
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return 0;
zebra: Add some debugs to neighbor entry processing When we get a neighbor entry in zebra we start processing it. Let's add some additional debugs to the processing so that when it bails out and we don't use the data, we know the reason. This should help in debugging the problems from why bgp does not appear to have data associated with a neighbor entry in the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-03-08 16:46:55 +01:00
}
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
memset(&mac, 0, sizeof(struct ethaddr));
if (h->nlmsg_type == RTM_NEWNEIGH) {
if (tb[NDA_LLADDR]) {
zebra: Switch to using ETH_ALEN Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-08-03 14:43:56 +02:00
if (RTA_PAYLOAD(tb[NDA_LLADDR]) != ETH_ALEN) {
zebra: Add some debugs to neighbor entry processing When we get a neighbor entry in zebra we start processing it. Let's add some additional debugs to the processing so that when it bails out and we don't use the data, we know the reason. This should help in debugging the problems from why bgp does not appear to have data associated with a neighbor entry in the kernel. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-03-08 16:46:55 +01:00
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug(
"%s family %s IF %s(%u) - LLADDR is not MAC, len %lu",
nl_msg_type_to_str(
h->nlmsg_type),
nl_family_to_str(
ndm->ndm_family),
ifp->name, ndm->ndm_ifindex,
(unsigned long)RTA_PAYLOAD(
tb[NDA_LLADDR]));
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return 0;
}
mac_present = 1;
zebra: Switch to using ETH_ALEN Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-08-03 14:43:56 +02:00
memcpy(&mac, RTA_DATA(tb[NDA_LLADDR]), ETH_ALEN);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
}
zebra: Use boolean for certain VxLAN-EVPN flags Use boolean variables instead of unsigned int for certain VxLAN-EVPN flags which are really used as boolean. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Anuradha Karuppiah <anuradhak@cumulusnetworks.com> Reviewed-by: Chirag Shah <chirag@cumulusnetworks.com> Ticket: CM-22288 Reviewed By: CCR-7832 Testing Done: Along with a subsequent, related commit
2018-09-10 19:13:20 +02:00
is_ext = !!(ndm->ndm_flags & NTF_EXT_LEARNED);
is_router = !!(ndm->ndm_flags & NTF_ROUTER);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug(
"Rx %s family %s IF %s(%u) IP %s MAC %s state 0x%x flags 0x%x",
nl_msg_type_to_str(h->nlmsg_type),
nl_family_to_str(ndm->ndm_family), ifp->name,
ndm->ndm_ifindex,
ipaddr2str(&ip, buf2, sizeof(buf2)),
mac_present
? prefix_mac2str(&mac, buf, sizeof(buf))
: "",
ndm->ndm_state, ndm->ndm_flags);
/* If the neighbor state is valid for use, process as an add or
* update
* else process as a delete. Note that the delete handling may
* result
* in re-adding the neighbor if it is a valid "remote" neighbor.
*/
if (ndm->ndm_state & NUD_VALID)
zebra: act on kernel notifications for remote neighbors as well There can be a race condition between kernel and frr as follows. Frr sends remote neigh notification. At the (almost) same time kernel might send a notification saying neigh is local. After processing this notifications, the state in frr is local while state in kernel is remote. This causes kernel and frr to be out of sync. This problem will be avoided if FRR acts on the kernel notifications for remote neighbors. When FRR sees a remote neighbor notification for a neighbor which it thinks is local, FRR will change the neigh state to remote. Ticket: CM-19923/CM-18830 Review: CCR-7222 Testing: Manual Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com>
2018-03-03 00:28:33 +01:00
return zebra_vxlan_handle_kernel_neigh_update(
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
ifp, link_if, &ip, &mac, ndm->ndm_state,
zebra: Use boolean for certain VxLAN-EVPN flags Use boolean variables instead of unsigned int for certain VxLAN-EVPN flags which are really used as boolean. Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Anuradha Karuppiah <anuradhak@cumulusnetworks.com> Reviewed-by: Chirag Shah <chirag@cumulusnetworks.com> Ticket: CM-22288 Reviewed By: CCR-7832 Testing Done: Along with a subsequent, related commit
2018-09-10 19:13:20 +02:00
is_ext, is_router);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
zebra: act on kernel notifications for remote neighbors as well There can be a race condition between kernel and frr as follows. Frr sends remote neigh notification. At the (almost) same time kernel might send a notification saying neigh is local. After processing this notifications, the state in frr is local while state in kernel is remote. This causes kernel and frr to be out of sync. This problem will be avoided if FRR acts on the kernel notifications for remote neighbors. When FRR sees a remote neighbor notification for a neighbor which it thinks is local, FRR will change the neigh state to remote. Ticket: CM-19923/CM-18830 Review: CCR-7222 Testing: Manual Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com>
2018-03-03 00:28:33 +01:00
return zebra_vxlan_handle_kernel_neigh_del(ifp, link_if, &ip);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
}
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug("Rx %s family %s IF %s(%u) IP %s",
nl_msg_type_to_str(h->nlmsg_type),
nl_family_to_str(ndm->ndm_family), ifp->name,
ndm->ndm_ifindex,
ipaddr2str(&ip, buf2, sizeof(buf2)));
/* Process the delete - it may result in re-adding the neighbor if it is
* a valid "remote" neighbor.
*/
zebra: act on kernel notifications for remote neighbors as well There can be a race condition between kernel and frr as follows. Frr sends remote neigh notification. At the (almost) same time kernel might send a notification saying neigh is local. After processing this notifications, the state in frr is local while state in kernel is remote. This causes kernel and frr to be out of sync. This problem will be avoided if FRR acts on the kernel notifications for remote neighbors. When FRR sees a remote neighbor notification for a neighbor which it thinks is local, FRR will change the neigh state to remote. Ticket: CM-19923/CM-18830 Review: CCR-7222 Testing: Manual Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com>
2018-03-03 00:28:33 +01:00
return zebra_vxlan_handle_kernel_neigh_del(ifp, link_if, &ip);
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
}
zebra: Remove unnecessary function parameter The snl variable is no longer needed to be passed around, so remove it from the calling path. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-05-23 15:37:06 +02:00
static int netlink_neigh_table(struct nlmsghdr *h, ns_id_t ns_id, int startup)
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
int len;
struct ndmsg *ndm;
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (h->nlmsg_type != RTM_NEWNEIGH)
return 0;
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
/* Length validity. */
len = h->nlmsg_len - NLMSG_LENGTH(sizeof(struct ndmsg));
if (len < 0)
return -1;
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
/* We are interested only in AF_INET or AF_INET6 notifications. */
ndm = NLMSG_DATA(h);
if (ndm->ndm_family != AF_INET && ndm->ndm_family != AF_INET6)
return 0;
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
zebra: Remove unnecessary function parameter The snl variable is no longer needed to be passed around, so remove it from the calling path. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-05-23 15:37:06 +02:00
return netlink_neigh_change(h, len);
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
}
/* Request for IP neighbor information from the kernel */
zebra: Start abstraction of zebra_dplane_info for context passing Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Signed-off-by: Mark Stapp <mjs@voltanet.io> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-09-12 20:59:57 +02:00
static int netlink_request_neigh(struct nlsock *netlink_cmd, int family,
int type, ifindex_t ifindex)
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
struct {
struct nlmsghdr n;
struct ndmsg ndm;
char buf[256];
} req;
/* Form the request, specifying filter (rtattr) if needed. */
memset(&req, 0, sizeof(req));
req.n.nlmsg_type = type;
zebra: netlink_request consolidataion Make netlink_request api generic where it can be used for dump or querying specific information request. nelink request nlm flags (NLM_F_ROOT | NLM_F_MATCH) are used to dump purpose, if client wants to query spcific MAC or IP using netlink_request does not require to set them. nlm struct is passed by the caller of netlink_request, it can also set the nlm request flags. Signed-off-by: Chirag Shah <chirag@cumulusnetworks.com>
2018-12-11 06:33:16 +01:00
req.n.nlmsg_flags = NLM_F_ROOT | NLM_F_MATCH | NLM_F_REQUEST;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
req.n.nlmsg_len = NLMSG_LENGTH(sizeof(struct ndmsg));
req.ndm.ndm_family = family;
if (ifindex)
addattr32(&req.n, sizeof(req), NDA_IFINDEX, ifindex);
zebra: Start abstraction of zebra_dplane_info for context passing Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Signed-off-by: Mark Stapp <mjs@voltanet.io> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-09-12 20:59:57 +02:00
return netlink_request(netlink_cmd, &req.n);
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
}
/*
* IP Neighbor table read using netlink interface. This is invoked
* at startup.
*/
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
int netlink_neigh_read(struct zebra_ns *zns)
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
int ret;
zebra: Start abstraction of zebra_dplane_info for context passing Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Signed-off-by: Mark Stapp <mjs@voltanet.io> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-09-12 20:59:57 +02:00
struct zebra_dplane_info dp_info;
zebra_dplane_info_from_zns(&dp_info, zns, true /*is_cmd*/);
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
/* Get IP neighbor table. */
zebra: Start abstraction of zebra_dplane_info for context passing Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Signed-off-by: Mark Stapp <mjs@voltanet.io> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-09-12 20:59:57 +02:00
ret = netlink_request_neigh(&zns->netlink_cmd, AF_UNSPEC, RTM_GETNEIGH,
0);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (ret < 0)
return ret;
zebra: Start abstraction of zebra_dplane_info for context passing Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Signed-off-by: Mark Stapp <mjs@voltanet.io> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-09-12 20:59:57 +02:00
ret = netlink_parse_info(netlink_neigh_table, &zns->netlink_cmd,
&dp_info, 0, 1);
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return ret;
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
}
/*
* IP Neighbor table read using netlink interface. This is for a specific
* VLAN device.
*/
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
int netlink_neigh_read_for_vlan(struct zebra_ns *zns, struct interface *vlan_if)
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
int ret = 0;
zebra: Start abstraction of zebra_dplane_info for context passing Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Signed-off-by: Mark Stapp <mjs@voltanet.io> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-09-12 20:59:57 +02:00
struct zebra_dplane_info dp_info;
zebra_dplane_info_from_zns(&dp_info, zns, true /*is_cmd*/);
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
zebra: Start abstraction of zebra_dplane_info for context passing Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Signed-off-by: Mark Stapp <mjs@voltanet.io> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-09-12 20:59:57 +02:00
ret = netlink_request_neigh(&zns->netlink_cmd, AF_UNSPEC, RTM_GETNEIGH,
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
vlan_if->ifindex);
if (ret < 0)
return ret;
zebra: Start abstraction of zebra_dplane_info for context passing Reduce or eliminate use of global zebra_ns structs in a couple of netlink/kernel code paths, so that those paths can potentially be made asynch eventually. Signed-off-by: Mark Stapp <mjs@voltanet.io> Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-09-12 20:59:57 +02:00
ret = netlink_parse_info(netlink_neigh_table, &zns->netlink_cmd,
&dp_info, 0, 0);
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return ret;
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
}
zebra: evpn remote delete fetch local entry An EVPN type-2 entry is in freeze state during remote update, remote VTEP can send typ-2 withdraw update, upon receiving an entry delete (withdraw), first check kernel has in local reachable state. Upon unfreeze use the local entry to advertise to peers. Fetch is for both MAC and IP, delete can come for only MAC or MAC-IP combined route. The specific entry fetch only required request flag to be set, dump flag is not required. Testing Done: Simulate two VTEPs to do M1, IP1 mobility sequence, freeze MAC during remote MAC update, subsequently send withdraw type-2 route from origintating VTEP. This results in read apis to invoke for local reachable entry. Zebra updates its cache and upon unfreeze originates type-2. Signed-off-by: Chirag Shah <chirag@cumulusnetworks.com>
2018-12-11 06:33:16 +01:00
/*
* Request for a specific IP in VLAN (SVI) device from IP Neighbor table,
* read using netlink interface.
*/
static int netlink_request_specific_neigh_in_vlan(struct zebra_ns *zns,
int type, struct ipaddr *ip,
ifindex_t ifindex)
{
struct {
struct nlmsghdr n;
struct ndmsg ndm;
char buf[256];
} req;
int ipa_len;
/* Form the request, specifying filter (rtattr) if needed. */
memset(&req, 0, sizeof(req));
req.n.nlmsg_len = NLMSG_LENGTH(sizeof(struct ndmsg));
req.n.nlmsg_flags = NLM_F_REQUEST;
req.n.nlmsg_type = type; /* RTM_GETNEIGH */
req.ndm.ndm_ifindex = ifindex;
if (IS_IPADDR_V4(ip)) {
ipa_len = IPV4_MAX_BYTELEN;
req.ndm.ndm_family = AF_INET;
} else {
ipa_len = IPV6_MAX_BYTELEN;
req.ndm.ndm_family = AF_INET6;
}
addattr_l(&req.n, sizeof(req), NDA_DST, &ip->ip.addr, ipa_len);
return netlink_request(&zns->netlink_cmd, &req.n);
}
int netlink_neigh_read_specific_ip(struct ipaddr *ip,
struct interface *vlan_if)
{
int ret = 0;
struct zebra_ns *zns;
struct zebra_vrf *zvrf = zebra_vrf_lookup_by_id(vlan_if->vrf_id);
char buf[INET6_ADDRSTRLEN];
struct zebra_dplane_info dp_info;
zns = zvrf->zns;
zebra_dplane_info_from_zns(&dp_info, zns, true /*is_cmd*/);
if (IS_ZEBRA_DEBUG_KERNEL)
zlog_debug("%s: neigh request IF %s(%u) IP %s vrf_id %u",
__PRETTY_FUNCTION__, vlan_if->name,
vlan_if->ifindex,
ipaddr2str(ip, buf, sizeof(buf)),
vlan_if->vrf_id);
ret = netlink_request_specific_neigh_in_vlan(zns, RTM_GETNEIGH, ip,
vlan_if->ifindex);
if (ret < 0)
return ret;
ret = netlink_parse_info(netlink_neigh_table, &zns->netlink_cmd,
&dp_info, 1, 0);
return ret;
}
zebra: Remove unnecessary function parameter The snl variable is no longer needed to be passed around, so remove it from the calling path. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-05-23 15:37:06 +02:00
int netlink_neigh_change(struct nlmsghdr *h, ns_id_t ns_id)
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
{
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
int len;
struct ndmsg *ndm;
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (!(h->nlmsg_type == RTM_NEWNEIGH || h->nlmsg_type == RTM_DELNEIGH))
return 0;
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
/* Length validity. */
len = h->nlmsg_len - NLMSG_LENGTH(sizeof(struct ndmsg));
zebra: Add error messages in places where we have a decode problem When we receive a netlink message from the kernel we have handler functions for when we send a netlink command, if these return a failure ( < 0 ) then we output that we had a parse issue. But if all we get is: 2018-06-21T23:47:45.298156+00:00 qct-ix1-08 zebra[1484]: netlink-cmd (NS 0) filter function error Then it is not very useful to figure out *where* the error happened. Add more error code when in a decode path to hopefully allow us to figure out where this message is coming from. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-06-22 20:22:02 +02:00
if (len < 0) {
zlog_err("%s: Message received from netlink is of a broken size %d %zu",
__PRETTY_FUNCTION__, h->nlmsg_len,
(size_t)NLMSG_LENGTH(sizeof(struct ndmsg)));
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return -1;
zebra: Add error messages in places where we have a decode problem When we receive a netlink message from the kernel we have handler functions for when we send a netlink command, if these return a failure ( < 0 ) then we output that we had a parse issue. But if all we get is: 2018-06-21T23:47:45.298156+00:00 qct-ix1-08 zebra[1484]: netlink-cmd (NS 0) filter function error Then it is not very useful to figure out *where* the error happened. Add more error code when in a decode path to hopefully allow us to figure out where this message is coming from. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-06-22 20:22:02 +02:00
}
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
/* Is this a notification for the MAC FDB or IP neighbor table? */
ndm = NLMSG_DATA(h);
if (ndm->ndm_family == AF_BRIDGE)
zebra: Remove unnecessary function parameter The snl variable is no longer needed to be passed around, so remove it from the calling path. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-05-23 15:37:06 +02:00
return netlink_macfdb_change(h, len, ns_id);
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (ndm->ndm_type != RTN_UNICAST)
return 0;
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
if (ndm->ndm_family == AF_INET || ndm->ndm_family == AF_INET6)
zebra: Remove unnecessary function parameter The snl variable is no longer needed to be passed around, so remove it from the calling path. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-05-23 15:37:06 +02:00
return netlink_ipneigh_change(h, len, ns_id);
zebra: Add address family filters The zebra netlink socket was attempting to read netlink messages with invalid address families in a couple areas. Added filters and warn messages. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2018-07-19 23:29:16 +02:00
else {
zebra: flog_warn conversion Convert Zebra to user error subsystem. Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-08-16 22:10:32 +02:00
flog_warn(
zebra: ZEBRA_[ERR|WARN] -> EC_ZEBRA Signed-off-by: Quentin Young <qlyoung@cumulusnetworks.com>
2018-09-13 21:21:05 +02:00
EC_ZEBRA_UNKNOWN_FAMILY,
zebra: Print out netlink message type instead of number When we are displaying data about a netlink message in debugs or errors, print out the message type as a string instead of a number. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-09-14 23:48:51 +02:00
"Invalid address family: %u received from kernel neighbor change: %s",
ndm->ndm_family, nl_msg_type_to_str(h->nlmsg_type));
zebra: Add address family filters The zebra netlink socket was attempting to read netlink messages with invalid address families in a couple areas. Added filters and warn messages. Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2018-07-19 23:29:16 +02:00
return 0;
}
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
return 0;
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
}
zebra: Use dataplane for evpn neighbor changes Move neighbor programming to the dataplane; remove old apis; remove some ifdef'd use of direct netlink code points, using neutral values outside of the netlink- specific files. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-08-23 19:59:10 +02:00
/*
* Utility neighbor-update function, using info from dplane context.
*/
static int netlink_neigh_update_ctx(const struct zebra_dplane_ctx *ctx,
int cmd)
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
{
zebra: Send RTPROT_ZEBRA for netlink messages missing this data Update neighbor entries and rule entries to have the RTPROT_ZEBRA protocol value. So we can tell where things come from. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-09-19 22:05:39 +02:00
uint8_t protocol = RTPROT_ZEBRA;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
struct {
struct nlmsghdr n;
struct ndmsg ndm;
char buf[256];
} req;
int ipa_len;
char buf[INET6_ADDRSTRLEN];
char buf2[ETHER_ADDR_STRLEN];
zebra: Use dataplane for evpn neighbor changes Move neighbor programming to the dataplane; remove old apis; remove some ifdef'd use of direct netlink code points, using neutral values outside of the netlink- specific files. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-08-23 19:59:10 +02:00
const struct ipaddr *ip;
const struct ethaddr *mac;
uint8_t flags;
uint16_t state;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
zebra: memset buf to prevent uninited writes into kernel Setup the buf used for extra data passed into kernel such that we are cleaning it out before writing data to it, so we can avoid writing uninited data. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2018-05-18 00:46:14 +02:00
memset(&req, 0, sizeof(req));
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
zebra: Use dataplane for evpn neighbor changes Move neighbor programming to the dataplane; remove old apis; remove some ifdef'd use of direct netlink code points, using neutral values outside of the netlink- specific files. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-08-23 19:59:10 +02:00
ip = dplane_ctx_neigh_get_ipaddr(ctx);
mac = dplane_ctx_neigh_get_mac(ctx);
if (is_zero_mac(mac))
mac = NULL;
flags = neigh_flags_to_netlink(dplane_ctx_neigh_get_flags(ctx));
state = neigh_state_to_netlink(dplane_ctx_neigh_get_state(ctx));
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
req.n.nlmsg_len = NLMSG_LENGTH(sizeof(struct ndmsg));
req.n.nlmsg_flags = NLM_F_REQUEST;
if (cmd == RTM_NEWNEIGH)
req.n.nlmsg_flags |= (NLM_F_CREATE | NLM_F_REPLACE);
req.n.nlmsg_type = cmd; // RTM_NEWNEIGH or RTM_DELNEIGH
req.ndm.ndm_family = IS_IPADDR_V4(ip) ? AF_INET : AF_INET6;
bgpd: support evpn nd ext community EVPN ND ext community support NA flag R-bit, to have proxy ND. Set R-bit in EVPN NA if a given router is default gateway or there is a local router attached, which can be determine based on local neighbor entry. Implement BGP ext community attribute to generate and parse R-bit and pass along zebra to program neigh entry in kernel. Upon receiving MAC/IP update with community type 0x06 and sub_type 0x08, pass the R-bit to zebra to program neigh entry. Set NTF_ROUTER in neigh entry and inform kernel to do proxy NA for EVPN. Ref: https://tools.ietf.org/html/draft-ietf-bess-evpn-na-flags-01 Ticket:CM-21712, CM-21711 Reviewed By: Testing Done: Configure Local vni enabled L3 Gateway, which would act as router, checked show evpn arp-cache vni x ip <ip of svi> on originated and remote VTEPs. "Router" flag is set. Signed-off-by: Chirag Shah <chirag@cumulusnetworks.com>
2018-07-07 06:46:46 +02:00
req.ndm.ndm_state = state;
zebra: Use dataplane for evpn neighbor changes Move neighbor programming to the dataplane; remove old apis; remove some ifdef'd use of direct netlink code points, using neutral values outside of the netlink- specific files. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-08-23 19:59:10 +02:00
req.ndm.ndm_ifindex = dplane_ctx_get_ifindex(ctx);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
req.ndm.ndm_type = RTN_UNICAST;
bgpd: support evpn nd ext community EVPN ND ext community support NA flag R-bit, to have proxy ND. Set R-bit in EVPN NA if a given router is default gateway or there is a local router attached, which can be determine based on local neighbor entry. Implement BGP ext community attribute to generate and parse R-bit and pass along zebra to program neigh entry in kernel. Upon receiving MAC/IP update with community type 0x06 and sub_type 0x08, pass the R-bit to zebra to program neigh entry. Set NTF_ROUTER in neigh entry and inform kernel to do proxy NA for EVPN. Ref: https://tools.ietf.org/html/draft-ietf-bess-evpn-na-flags-01 Ticket:CM-21712, CM-21711 Reviewed By: Testing Done: Configure Local vni enabled L3 Gateway, which would act as router, checked show evpn arp-cache vni x ip <ip of svi> on originated and remote VTEPs. "Router" flag is set. Signed-off-by: Chirag Shah <chirag@cumulusnetworks.com>
2018-07-07 06:46:46 +02:00
req.ndm.ndm_flags = flags;
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
zebra: Send RTPROT_ZEBRA for netlink messages missing this data Update neighbor entries and rule entries to have the RTPROT_ZEBRA protocol value. So we can tell where things come from. Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
2019-09-19 22:05:39 +02:00
addattr_l(&req.n, sizeof(req),
NDA_PROTOCOL, &protocol, sizeof(protocol));
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
ipa_len = IS_IPADDR_V4(ip) ? IPV4_MAX_BYTELEN : IPV6_MAX_BYTELEN;
addattr_l(&req.n, sizeof(req), NDA_DST, &ip->ip.addr, ipa_len);
if (mac)
addattr_l(&req.n, sizeof(req), NDA_LLADDR, mac, 6);
if (IS_ZEBRA_DEBUG_KERNEL)
zebra: add kernel neigh update api The kernel neigh update api helps update neighbor entry, using changing state and flags parameters. Ticket:CM-22864 Reviewed By: Testing Done: Signed-off-by:Chirag Shah <chirag@cumulusnetworks.com>
2019-01-29 00:32:45 +01:00
zlog_debug("Tx %s family %s IF %s(%u) Neigh %s MAC %s flags 0x%x state 0x%x",
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
nl_msg_type_to_str(cmd),
zebra: Use dataplane for evpn neighbor changes Move neighbor programming to the dataplane; remove old apis; remove some ifdef'd use of direct netlink code points, using neutral values outside of the netlink- specific files. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-08-23 19:59:10 +02:00
nl_family_to_str(req.ndm.ndm_family),
dplane_ctx_get_ifname(ctx),
dplane_ctx_get_ifindex(ctx),
ipaddr2str(ip, buf, sizeof(buf)),
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
mac ? prefix_mac2str(mac, buf2, sizeof(buf2))
zebra: Use dataplane for evpn neighbor changes Move neighbor programming to the dataplane; remove old apis; remove some ifdef'd use of direct netlink code points, using neutral values outside of the netlink- specific files. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-08-23 19:59:10 +02:00
: "null",
flags, state);
*: reindent indent.py `git ls-files | pcregrep '\.[ch]$' | pcregrep -v '^(ldpd|babeld|nhrpd)/'` Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-17 14:03:14 +02:00
zebra: Use dataplane for evpn neighbor changes Move neighbor programming to the dataplane; remove old apis; remove some ifdef'd use of direct netlink code points, using neutral values outside of the netlink- specific files. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-08-23 19:59:10 +02:00
return netlink_talk_info(netlink_talk_filter, &req.n,
dplane_ctx_get_ns(ctx), 0);
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
}
zebra: use dataplane for vxlan remote mac programming Move vxlan remote MAC install and uninstall to the async dataplane. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-07-30 17:54:07 +02:00
/*
* Update MAC, using dataplane context object.
*/
enum zebra_dplane_result kernel_mac_update_ctx(struct zebra_dplane_ctx *ctx)
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
{
zebra: use dataplane for vxlan remote mac programming Move vxlan remote MAC install and uninstall to the async dataplane. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-07-30 17:54:07 +02:00
return netlink_macfdb_update_ctx(ctx);
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
}
zebra: Use dataplane for evpn neighbor changes Move neighbor programming to the dataplane; remove old apis; remove some ifdef'd use of direct netlink code points, using neutral values outside of the netlink- specific files. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-08-23 19:59:10 +02:00
enum zebra_dplane_result kernel_neigh_update_ctx(struct zebra_dplane_ctx *ctx)
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
{
zebra: Use dataplane for evpn neighbor changes Move neighbor programming to the dataplane; remove old apis; remove some ifdef'd use of direct netlink code points, using neutral values outside of the netlink- specific files. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-08-23 19:59:10 +02:00
int ret = -1;
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
zebra: Use dataplane for evpn neighbor changes Move neighbor programming to the dataplane; remove old apis; remove some ifdef'd use of direct netlink code points, using neutral values outside of the netlink- specific files. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-08-23 19:59:10 +02:00
switch (dplane_ctx_get_op(ctx)) {
case DPLANE_OP_NEIGH_INSTALL:
case DPLANE_OP_NEIGH_UPDATE:
ret = netlink_neigh_update_ctx(ctx, RTM_NEWNEIGH);
break;
case DPLANE_OP_NEIGH_DELETE:
ret = netlink_neigh_update_ctx(ctx, RTM_DELNEIGH);
break;
zebra: move EVPN VTEP programming to dataplane Move VTEP install/uninstall to the zebra dataplane. Remove synch kernel-facing apis and helper functions. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-08-26 21:44:54 +02:00
case DPLANE_OP_VTEP_ADD:
ret = netlink_vxlan_flood_update_ctx(ctx, RTM_NEWNEIGH);
break;
case DPLANE_OP_VTEP_DELETE:
ret = netlink_vxlan_flood_update_ctx(ctx, RTM_DELNEIGH);
break;
zebra: Use dataplane for evpn neighbor changes Move neighbor programming to the dataplane; remove old apis; remove some ifdef'd use of direct netlink code points, using neutral values outside of the netlink- specific files. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-08-23 19:59:10 +02:00
default:
break;
}
zebra: MAC and Neighbor (ARP/ND) handling Implement handling of MACs and Neighbors (ARP/ND entries) in zebra: - MAC and Neighbor database handlers - Read MACs and Neighbors from the kernel, when needed and create entries in zebra's MAC and Neighbor databases. - Handle add/update/delete notifications from the kernel for MACs and Neighbors and update zebra's database appropriately - Inform locally learnt MACs and Neighbors to client - Handle MACIP add/delete from client and install appriporiate entries into the kernel - Since Neighbor entries will be installed on an SVI, implement the needed mappings NOTE: kernel interface is only implemented for Linux/netlink Signed-off-by: Vivek Venkatraman <vivek@cumulusnetworks.com> Reviewed-by: Donald Sharp <sharpd@cumulusnetworks.com>
2017-05-15 07:44:13 +02:00
zebra: Use dataplane for evpn neighbor changes Move neighbor programming to the dataplane; remove old apis; remove some ifdef'd use of direct netlink code points, using neutral values outside of the netlink- specific files. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-08-23 19:59:10 +02:00
return (ret == 0 ?
ZEBRA_DPLANE_REQUEST_SUCCESS : ZEBRA_DPLANE_REQUEST_FAILURE);
zebra: add kernel neigh update api The kernel neigh update api helps update neighbor entry, using changing state and flags parameters. Ticket:CM-22864 Reviewed By: Testing Done: Signed-off-by:Chirag Shah <chirag@cumulusnetworks.com>
2019-01-29 00:32:45 +01:00
}
zebra: infra for LSP updates using dplane Adding infra to zebra dplane to support LSP updates. Add kernel api for LSP updates that uses a dataplane context; add stub apis for netlink, bsd, and 'null' kernel paths. Add version of netlink mpls update code that takes a dplane context struct instead of a zebra lsp struct. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-10-23 16:57:01 +02:00
/*
* MPLS label forwarding table change via netlink interface, using dataplane
* context information.
*/
zebra: openbsd LSP update code Finish the LSP update code for the async dataplane for the openbsd platform. Remove synch apis now that we've converted to the async code path. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-11-26 22:08:55 +01:00
int netlink_mpls_multipath(int cmd, struct zebra_dplane_ctx *ctx)
zebra: infra for LSP updates using dplane Adding infra to zebra dplane to support LSP updates. Add kernel api for LSP updates that uses a dataplane context; add stub apis for netlink, bsd, and 'null' kernel paths. Add version of netlink mpls update code that takes a dplane context struct instead of a zebra lsp struct. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-10-23 16:57:01 +02:00
{
mpls_lse_t lse;
zebra: use const in dplane pw nhlfe accessors Use const in the accessors for pseudowire nhlfe data; pull that through the kernel-facing apis that use that data. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-03-07 19:09:51 +01:00
const zebra_nhlfe_t *nhlfe;
zebra: infra for LSP updates using dplane Adding infra to zebra dplane to support LSP updates. Add kernel api for LSP updates that uses a dataplane context; add stub apis for netlink, bsd, and 'null' kernel paths. Add version of netlink mpls update code that takes a dplane context struct instead of a zebra lsp struct. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-10-23 16:57:01 +02:00
struct nexthop *nexthop = NULL;
unsigned int nexthop_num;
const char *routedesc;
int route_type;
struct {
struct nlmsghdr n;
struct rtmsg r;
char buf[NL_PKT_BUF_SIZE];
} req;
memset(&req, 0, sizeof(req) - NL_PKT_BUF_SIZE);
/*
* Count # nexthops so we can decide whether to use singlepath
* or multipath case.
*/
nexthop_num = 0;
for (nhlfe = dplane_ctx_get_nhlfe(ctx); nhlfe; nhlfe = nhlfe->next) {
nexthop = nhlfe->nexthop;
if (!nexthop)
continue;
if (cmd == RTM_NEWROUTE) {
/* Count all selected NHLFEs */
if (CHECK_FLAG(nhlfe->flags, NHLFE_FLAG_SELECTED)
&& CHECK_FLAG(nexthop->flags, NEXTHOP_FLAG_ACTIVE))
nexthop_num++;
} else { /* DEL */
/* Count all installed NHLFEs */
if (CHECK_FLAG(nhlfe->flags, NHLFE_FLAG_INSTALLED)
&& CHECK_FLAG(nexthop->flags, NEXTHOP_FLAG_FIB))
nexthop_num++;
}
}
if ((nexthop_num == 0) ||
(!dplane_ctx_get_best_nhlfe(ctx) && (cmd != RTM_DELROUTE)))
return 0;
req.n.nlmsg_len = NLMSG_LENGTH(sizeof(struct rtmsg));
req.n.nlmsg_flags = NLM_F_CREATE | NLM_F_REQUEST;
req.n.nlmsg_type = cmd;
req.n.nlmsg_pid = dplane_ctx_get_ns(ctx)->nls.snl.nl_pid;
req.r.rtm_family = AF_MPLS;
req.r.rtm_table = RT_TABLE_MAIN;
req.r.rtm_dst_len = MPLS_LABEL_LEN_BITS;
req.r.rtm_scope = RT_SCOPE_UNIVERSE;
req.r.rtm_type = RTN_UNICAST;
if (cmd == RTM_NEWROUTE) {
/* We do a replace to handle update. */
req.n.nlmsg_flags |= NLM_F_REPLACE;
/* set the protocol value if installing */
route_type = re_type_from_lsp_type(
dplane_ctx_get_best_nhlfe(ctx)->type);
req.r.rtm_protocol = zebra2proto(route_type);
}
/* Fill destination */
lse = mpls_lse_encode(dplane_ctx_get_in_label(ctx), 0, 0, 1);
addattr_l(&req.n, sizeof(req), RTA_DST, &lse, sizeof(mpls_lse_t));
/* Fill nexthops (paths) based on single-path or multipath. The paths
* chosen depend on the operation.
*/
zebra: openbsd LSP update code Finish the LSP update code for the async dataplane for the openbsd platform. Remove synch apis now that we've converted to the async code path. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-11-26 22:08:55 +01:00
if (nexthop_num == 1) {
zebra: infra for LSP updates using dplane Adding infra to zebra dplane to support LSP updates. Add kernel api for LSP updates that uses a dataplane context; add stub apis for netlink, bsd, and 'null' kernel paths. Add version of netlink mpls update code that takes a dplane context struct instead of a zebra lsp struct. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-10-23 16:57:01 +02:00
routedesc = "single-path";
_netlink_mpls_debug(cmd, dplane_ctx_get_in_label(ctx),
routedesc);
nexthop_num = 0;
for (nhlfe = dplane_ctx_get_nhlfe(ctx);
nhlfe; nhlfe = nhlfe->next) {
nexthop = nhlfe->nexthop;
if (!nexthop)
continue;
if ((cmd == RTM_NEWROUTE
&& (CHECK_FLAG(nhlfe->flags, NHLFE_FLAG_SELECTED)
&& CHECK_FLAG(nexthop->flags,
NEXTHOP_FLAG_ACTIVE)))
|| (cmd == RTM_DELROUTE
&& (CHECK_FLAG(nhlfe->flags,
NHLFE_FLAG_INSTALLED)
&& CHECK_FLAG(nexthop->flags,
NEXTHOP_FLAG_FIB)))) {
/* Add the gateway */
_netlink_mpls_build_singlepath(
routedesc, nhlfe,
&req.n, &req.r,
sizeof(req), cmd);
nexthop_num++;
break;
}
}
} else { /* Multipath case */
char buf[NL_PKT_BUF_SIZE];
struct rtattr *rta = (void *)buf;
struct rtnexthop *rtnh;
zebra: use const in dplane pw nhlfe accessors Use const in the accessors for pseudowire nhlfe data; pull that through the kernel-facing apis that use that data. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2019-03-07 19:09:51 +01:00
const union g_addr *src1 = NULL;
zebra: infra for LSP updates using dplane Adding infra to zebra dplane to support LSP updates. Add kernel api for LSP updates that uses a dataplane context; add stub apis for netlink, bsd, and 'null' kernel paths. Add version of netlink mpls update code that takes a dplane context struct instead of a zebra lsp struct. Signed-off-by: Mark Stapp <mjs@voltanet.io>
2018-10-23 16:57:01 +02:00
rta->rta_type = RTA_MULTIPATH;
rta->rta_len = RTA_LENGTH(0);
rtnh = RTA_DATA(rta);
routedesc = "multipath";
_netlink_mpls_debug(cmd, dplane_ctx_get_in_label(ctx),
routedesc);
nexthop_num = 0;
for (nhlfe = dplane_ctx_get_nhlfe(ctx);
nhlfe; nhlfe = nhlfe->next) {
nexthop = nhlfe->nexthop;
if (!nexthop)
continue;
if ((cmd == RTM_NEWROUTE
&& (CHECK_FLAG(nhlfe->flags, NHLFE_FLAG_SELECTED)
&& CHECK_FLAG(nexthop->flags,
NEXTHOP_FLAG_ACTIVE)))
|| (cmd == RTM_DELROUTE
&& (CHECK_FLAG(nhlfe->flags,
NHLFE_FLAG_INSTALLED)
&& CHECK_FLAG(nexthop->flags,
NEXTHOP_FLAG_FIB)))) {
nexthop_num++;
/* Build the multipath */
_netlink_mpls_build_multipath(routedesc, nhlfe,
rta, rtnh, &req.r,
&src1);
rtnh = RTNH_NEXT(rtnh);
}
}
/* Add the multipath */
if (rta->rta_len > RTA_LENGTH(0))
addattr_l(&req.n, NL_PKT_BUF_SIZE, RTA_MULTIPATH,
RTA_DATA(rta), RTA_PAYLOAD(rta));
}
/* Talk to netlink socket. */
return netlink_talk_info(netlink_talk_filter, &req.n,
dplane_ctx_get_ns(ctx), 0);
}
build: zebra: remove *_method Makefile hacks replace with preprocessor checks in source files. Much simpler... Signed-off-by: David Lamparter <equinox@opensourcerouting.org>
2017-07-26 19:49:15 +02:00
#endif /* HAVE_NETLINK */
Reference in a new issue Copy permalink