Inband Health Monitoring

[cprice2k7]

Recently I have been logging the following error messages on my CSM:

Oct 17 14:17:02.277 CST: %CSM_SLB-6-RSERVERSTATE: Module 4 server state changed: server x.x.x.x:443 in serverfarm 'WEB_SEAL2' has reached configured inband health retries

Oct 17 14:17:02.289 CST: %CSM_SLB-6-RSERVERSTATE: Module 4 server state changed: server x.x.x.x:443 in serverfarm 'WEB_SEAL2' is considered operational by Inband health monitoring

I have configured inband health monitoring with "health retries 3 failed 100" When I look at the real server state it is always operational. I have yet to see the server in a failed or down state. Any ideas why this would occur?

Thanks

[ivillegas]

This error message is saying that the Content Switching Module arped out, but did not get a respponce from the IP listed in te error message,

[Gilles Dufour]

it has nothing to do with arp.

Inband health monitoring uses the production traffic to detect server failure [no response or reset] and decide to outofservice it when necessary.

I would say, are you sure you are looking at the same ip ?

With your config, a server should be removed for 100 sec upon failover.

If not, capture a failure with sniffer trace, then send it with the config and log file to the TAC.

Gilles.

[cprice2k7]

I did just what you suggested. Tac said that the threshold was being hit but not exceeded. In this instance three connections were terminated but the fourth was accepted and kept the server in service.

Thanks for the input.