diff options
| author | Claudio Jeker <claudio@cvs.openbsd.org> | 2005-04-12 14:32:02 +0000 |
|---|---|---|
| committer | Claudio Jeker <claudio@cvs.openbsd.org> | 2005-04-12 14:32:02 +0000 |
| commit | cf15ea6afd9afb2f756e819126b4db8ed244adf9 (patch) | |
| tree | 44a92901de0895b80ab82145c4c9cc1aca114a48 /usr.sbin/bgpd/rde_decide.c | |
| parent | 69eb085e352c4a2ce2742b3774e248665379d0ac (diff) | |
Introduce a per prefix weight. The weight is used to tip prefixes with equal
long AS pathes in one or the other direction. It weights a prefix at a very
late stage in the decision process. This is a nice bgpd feature to traffic
engineer networks where most AS pathes are equally long.
OK henning@
Diffstat (limited to 'usr.sbin/bgpd/rde_decide.c')
| -rw-r--r-- | usr.sbin/bgpd/rde_decide.c | 20 |
1 files changed, 15 insertions, 5 deletions
diff --git a/usr.sbin/bgpd/rde_decide.c b/usr.sbin/bgpd/rde_decide.c index 6f3a7116df0..98117ae4ae0 100644 --- a/usr.sbin/bgpd/rde_decide.c +++ b/usr.sbin/bgpd/rde_decide.c @@ -1,4 +1,4 @@ -/* $OpenBSD: rde_decide.c,v 1.40 2004/11/11 10:35:15 claudio Exp $ */ +/* $OpenBSD: rde_decide.c,v 1.41 2005/04/12 14:32:01 claudio Exp $ */ /* * Copyright (c) 2003, 2004 Claudio Jeker <claudio@openbsd.org> @@ -154,21 +154,31 @@ prefix_cmp(struct prefix *p1, struct prefix *p2) return -1; } - /* 7. nexthop costs. NOT YET -> IGNORE */ + /* + * 7. local tie-breaker, this weight is here to tip equal long AS + * pathes in one or the other direction. It happens more and more + * that AS pathes are equally long and so traffic engineering needs + * a metric that weights a prefix at a very late stage in the + * decision process. + */ + if ((asp1->weight - asp2->weight) != 0) + return (asp1->weight - asp2->weight); + + /* 8. nexthop costs. NOT YET -> IGNORE */ /* - * 8. older route (more stable) wins but only if route-age + * 9. older route (more stable) wins but only if route-age * evaluation is enabled. */ if (rde_decisionflags() & BGPD_FLAG_DECISION_ROUTEAGE) if ((p2->lastchange - p1->lastchange) != 0) return (p2->lastchange - p1->lastchange); - /* 9. lowest BGP Id wins */ + /* 10. lowest BGP Id wins */ if ((p2->peer->remote_bgpid - p1->peer->remote_bgpid) != 0) return (p2->peer->remote_bgpid - p1->peer->remote_bgpid); - /* 10. lowest peer address wins (IPv4 is better than IPv6) */ + /* 11. lowest peer address wins (IPv4 is better than IPv6) */ if (memcmp(&p1->peer->remote_addr, &p2->peer->remote_addr, sizeof(p1->peer->remote_addr)) != 0) return (-memcmp(&p1->peer->remote_addr, &p2->peer->remote_addr, |