]> git.puffer.fish Git - mirror/frr.git/commit
bgpd, tests: don't send local nexthop from rr client 17073/head
authorLouis Scalbert <louis.scalbert@6wind.com>
Fri, 11 Oct 2024 04:59:16 +0000 (06:59 +0200)
committerLouis Scalbert <louis.scalbert@6wind.com>
Mon, 14 Oct 2024 08:06:35 +0000 (10:06 +0200)
commitc4a82636282c059a32ff9b19396c547c46cbf82b
tree8a60feccfc235104f03ba8bed299d904ef5c6067
parent4ae65cc21eccd6d324949d33d36a328ff6b45cfe
bgpd, tests: don't send local nexthop from rr client

AS 65000  | AS 65001
          |
      RR  |
       |  |
R1 --- | --- R2
          |

When r1 peer is an iBGP route reflector client of rr and r2 peer is a
eBGP neighbor of rr, and all three routers shares the same network, r2
receives announcements coming from r1 with a IPv6 link-local nexthop
from rr. This is incorrect as r2 should send traffic to r1 without
involving rr.

Do not send an IPv6 link-local nexthop if the originating peer is a
route-reflector client.

Link: https://github.com/FRRouting/frr/pull/16219#issuecomment-2397425505
Link: https://github.com/FRRouting/frr/pull/17037#discussion_r1792529683
Signed-off-by: Louis Scalbert <louis.scalbert@6wind.com>
bgpd/bgp_route.c
tests/topotests/bgp_nexthop_ipv6/r4/show_bgp_ipv6_step2.json
tests/topotests/bgp_nexthop_ipv6/r5/show_bgp_ipv6_step2.json