timeout problem

We have a f5 cluster to service for my web application.

recently we meet a question. some of my customer feeded back to us they can’t login application using the f5 virtual servers,but no all user like this,part of them can use it.

I check my f5 log, I found the following information

----------------------------------------------------

Mar 17 15:55:06 yndl httpd[2843]: [warn] [client 10.180.9.140] AUTHCACHE Error processing cookie 6E87CB448B7D7F137D4937878C15

372BB23D2B3F - Cookie Expired (cache_timeout=86400)

Mar 17 15:55:06 yndl httpd[2843]: [error] channelSocket.receive(): Error receiving message body -1 2

Mar 17 15:55:06 yndl httpd[2843]: [error] workerEnv.processCallbacks() Error reading reply

Mar 17 15:55:06 yndl httpd[2843]: [error] ajp13.service() ajpGetReply recoverable error 120000

Mar 17 15:55:10 yndl httpd[679]: [error] channelSocket.receive(): Error receiving message body -1 0

Mar 17 15:55:10 yndl httpd[679]: [error] workerEnv.processCallbacks() Error reading reply

Mar 17 15:55:10 yndl httpd[679]: [error] ajp13.service() ajpGetReply recoverable error 120000

Mar 17 15:55:11 yndl httpd[1672]: [error] channelSocket.receive(): Error receiving message body -1 0

Mar 17 15:55:11 yndl httpd[1672]: [error] workerEnv.processCallbacks() Error reading reply

Mar 17 15:55:11 yndl httpd[1672]: [error] ajp13.service() ajpGetReply recoverable error 120000

Mar 17 15:55:11 yndl httpd[2477]: [error] channelSocket.receive(): Error receiving message body -1 0

Mar 17 15:55:11 yndl httpd[2477]: [error] workerEnv.processCallbacks() Error reading reply

Mar 17 15:55:11 yndl httpd[2477]: [error] ajp13.service() ajpGetReply recoverable error 120000

Mar 17 15:55:23 yndl httpd[1677]: [error] channelSocket.receive(): Error receiving message body -1 0

Mar 17 15:55:23 yndl httpd[1677]: [error] workerEnv.processCallbacks() Error reading reply

Mar 17 15:55:23 yndl httpd[1677]: [error] ajp13.service() ajpGetReply recoverable error 120000

------------------------------------------------

the picture is that I captured it through a website monitoring tool that I used it to monitor my application website.

I want to ask if some of my error configuration may lead the problem

Is that the httpd error log from LTM? If so, that’s for the admin GUI and isn’t related to the issue.

I’d suggest trying to capture a tcpdump of a failure to see if the request is making it to the virtual server and if so where it’s failing. Try searching this forum or the AskF5.com site for ‘tcpdump’ for details on using tcpdump.

Aaron

3ks,It’s not a httpd error of LTM,it’s a secure log located in /var/log

how can I capture a tcpdump?