In article <200207162123.34323.mstagarescu@telogy.com> you wrote: > The Linux router performs ok till about few seconds and then fails very > badly. Besides the normal cache Timeout there are multiple strategies to detect dead neighbours, for example for fall back gateways and quicker usermode notification. It might help to set gc_stale_time higher, set locktime higher, or set ucast_solicit and mcast_solicit to 0. 2ecki@calista:/proc/sys/net/ipv4/neigh/eth0> egrep . * anycast_delay:100 app_solicit:0 base_reachable_time:30 delay_first_probe_time:5 gc_stale_time:60 locktime:100 mcast_solicit:3 proxy_delay:80 proxy_qlen:64 retrans_time:100 ucast_solicit:3 unres_qlen:3 See for example http://www.bb-zone.com/zope/bbzone/docs/slgfg/part2/cha04/sec08 Greetings Bernd - : send the line "unsubscribe linux-net" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
