Package: vrrpd Version: 1.0-1 I am using two Linux 2.6.11.7 kernel servers with VRRP implemented. The shut down of the Master server goes without any problem, in fact the Backup takes over like it should.
The problem appears when the Master server comes back on and sends VRRP HELO messages with higher priority, the Backup server doesnt stop sending his own messages with lower priority like it should conforming to RFC2338. The problem isnt resolved before killing one of the vrrpd Daemons. The Backup server also doesnt stop answering to ARP requests to the virtual IP which causes multiple answers to clients requests and changes in the network topology. Here's a tcpdump printout that shows the problem after Master is back on: routeur1:~# tcpdump -vv proto 112 tcpdump: listening on eth0, link-type EN10MB (Ethernet), capture size 96 bytes 02:03:19.077826 IP (tos 0x0, ttl 255, id 51717, offset 0, flags [none], length: 40) 10.0.0.1 > 224.0.0.18: VRRPv2, Advertisement, vrid 20, prio 25, authtype none, intvl 1s, length 20, addrs: 10.0.0.1 02:03:19.389520 IP (tos 0x0, ttl 255, id 24580, offset 0, flags [none], length: 40) 10.0.0.3 > 224.0.0.18: VRRPv2, Advertisement, vrid 20, prio 24, authtype none, intvl 1s, length 20, addrs: 10.0.0.1 02:03:20.129262 IP (tos 0x0, ttl 255, id 51973, offset 0, flags [none], length: 40) 10.0.0.1 > 224.0.0.18: VRRPv2, Advertisement, vrid 20, prio 25, authtype none, intvl 1s, length 20, addrs: 10.0.0.1 02:03:20.433584 IP (tos 0x0, ttl 255, id 24836, offset 0, flags [none], length: 40) 10.0.0.3 > 224.0.0.18: VRRPv2, Advertisement, vrid 20, prio 24, authtype none, intvl 1s, length 20, addrs: 10.0.0.1 Patrick Chammas Telecom Paris
