|
> - ICMP_PORT_UNREACH: when there is no real services defined or when
> the real service does not work (the last depends on the used method
> for deliveryng TCP/UDP requests locally)
>
Just start tcpdump -n proto ICMP on your LVS director.
>
tcpdump showed a large number of destination port unreachable messages.
15:48:04.693986 eth0 > 64.77.118.84 > 209.178.165.214: icmp: 64.77.118.84
tcp port www unreachable [tos 0xc0]
15:48:04.913280 eth0 > 64.77.118.84 > 199.48.25.11: icmp: 64.77.118.84 tcp
port www unreachable [tos 0xd0]
15:48:05.556277 eth0 < 62.252.209.155 > 64.77.118.84: icmp: echo request
(DF)
15:48:05.556322 eth0 > 64.77.118.84 > 62.252.209.155: icmp: echo reply
15:48:05.696958 eth0 > 64.77.118.84 > 195.23.81.111: icmp: 64.77.118.84
tcp port www unreachable [tos 0xc0]
15:48:05.794981 eth0 > 64.77.118.84 > 212.17.108.101: icmp: 64.77.118.84
tcp port www unreachable [tos 0xc0]
IP Virtual Server version 1.0.5 (size=8192)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port Forward Weight ActiveConn InActConn
TCP 64.77.118.84:80 rr
-> 192.168.12.3:80 Masq 1 171 30672
TCP 64.77.118.84:5555 rr
-> 192.168.12.3:5555 Masq 1 0 0
This doesn't seem to make sense as we have a virtual service with a real
server on port 80. I thought these messages would be sent out if there
were no real servers on the virtual service. Why would these messages be
sent out?
Hayden Myers
Skyline Network Technologies
hayden@xxxxxxxxxxx
(888)917-1600x120
| 