| commit | 9e7a53c179f6897128b24435452b5d3d0f8c715a | [log] [tgz] |
|---|---|---|
| author | Timo Teräs <timo.teras@iki.fi> | Thu Apr 24 10:22:37 2014 +0300 |
| committer | Timo Teräs <timo.teras@iki.fi> | Wed Jun 25 21:20:20 2014 +0300 |
| tree | 80a5cc6e9e6dc411587764d865a2bf5f489000ba | |
| parent | 2aa640bd78b64821bde9a53ecdd1e96e91b20ae4 [diff] [blame] |
bgpd: implement "next-hop-self all" As specified in: http://www.cisco.com/c/en/us/td/docs/ios-xml/ios/iproute_bgp/command/irg-cr-book/bgp-m1.html#wp4972925610 This allows overriding next-hop for ibgp learned routes on an RR for reflected routes. Especially useful for using iBGP in DMVPN setups. See: http://blog.ipspace.net/2014/04/changes-in-ibgp-next-hop-processing.html Signed-off-by: Timo Teräs <timo.teras@iki.fi>
diff --git a/bgpd/bgp_route.c b/bgpd/bgp_route.c index f421ca5..232a6a1 100644 --- a/bgpd/bgp_route.c +++ b/bgpd/bgp_route.c
@@ -976,7 +976,8 @@ } /* next-hop-set */ - if (transparent || reflect + if (transparent + || (reflect && ! CHECK_FLAG (peer->af_flags[afi][safi], PEER_FLAG_NEXTHOP_SELF_ALL)) || (CHECK_FLAG (peer->af_flags[afi][safi], PEER_FLAG_NEXTHOP_UNCHANGED) && ((p->family == AF_INET && attr->nexthop.s_addr) #ifdef HAVE_IPV6