"Timeout while answering request"

[CeeEss]

Having spent the last 2 weeks disabling extraneous objects in my templates, i have a virtually error-free zabbix_server.log (clap clap), except for loads of these:

Timeout while answering request

I suspect it's simply ping unavailability as the string doesn't occur anywhere in the server code. Is there anything I can do/modify that will show which hosts timed out?

thanks

- cal

[walterheck]

Old issue, but there is no answer yet. I'm seeing the same thing agent side, and there's no explanation anywhere.

11923:20110211:135816 Timeout while answering request
11923:20110211:140055 Timeout while answering request
11923:20110211:140349 Active check [system.cpu.util[,iowait,avg1]] is not supported. Disabled.
11923:20110211:140408 Timeout while answering request
11923:20110211:140438 Timeout while answering request
11923:20110211:140611 Timeout while answering request
11923:20110211:140617 Timeout while answering request
11923:20110211:140941 Timeout while answering request

[walterheck]

I noticed this server was running stock 1.4.6 that came with debian lenny, so I upgraded them to lenny-backports' 1.8.2. I'll keep an eye on them and report back if it is fixed or not.

[nelsonab]

Digging through the code that statement is only shown when there is a SIGALARM from somewhere else. What I can't determine right now is if the SIGALARM is caused by a timeout signal handler somewhere.

My hunch, one of your items is taking a long time to return, or there is an error getting it, thus the message. Perhaps increasing the debug level on the agent to 4, to maybe see the request come in.

[walterheck]

My hunch, one of your items is taking a long time to return, or there is an error getting it, thus the message. Perhaps increasing the debug level on the agent to 4, to maybe see the request come in.

it stopped happening after I upgraded the agents to 1.8.2, but here's what surrounds the log messages in debuglevel 4 on the agents (this was 1.4.6):

--
12160:20110211:145834 OK
12160:20110211:145834 For key [system.cpu.load[,avg15]] received value [0.510000]
12160:20110211:145837 XML before sending [<req><host>c2VydmVyMDIub3BsZXhpbmcuY29t</host><key>c3lzdGVtLmNwdS5sb2FkWyxhdmcxNV0=</key><data>MC41MTAwMDA=</data></req>]
12160:20110211:145837 Timeout while answering request
12160:20110211:145837 Send value error: [recv] ZBX_TCP_READ() failed [Interrupted system call]
12160:20110211:145837 In get_min_nextcheck()
12160:20110211:145837 No sleeping
--
12160:20110211:145854 OK
12160:20110211:145854 For key [system.cpu.util[,user,avg1]] received value [3.208435]
12160:20110211:145857 XML before sending [<req><host>c2VydmVyMDIub3BsZXhpbmcuY29t</host><key>c3lzdGVtLmNwdS51dGlsWyx1c2VyLGF2ZzFd</key><data>My4yMDg0MzU=</data></req>]
12160:20110211:145857 Timeout while answering request
12160:20110211:145857 Send value error: [recv] ZBX_TCP_READ() failed [Interrupted system call]
12160:20110211:145857 In get_min_nextcheck()
12160:20110211:145857 No sleeping
--
12160:20110211:145952 OK
12160:20110211:145952 For key [system.swap.size[,pfree]] received value [99.590022]
12160:20110211:145952 XML before sending [<req><host>c2VydmVyMDIub3BsZXhpbmcuY29t</host><key>c3lzdGVtLnN3YXAuc2l6ZVsscGZyZWVd</key><data>OTkuNTkwMDIy</data></req>]
12160:20110211:145955 Timeout while answering request
12160:20110211:145955 Send value error: [recv] ZBX_TCP_READ() failed [Interrupted system call]
12160:20110211:145955 In get_min_nextcheck()
12160:20110211:145955 No sleeping
--
14435:20110211:205315 OK
14435:20110211:205315 For key [system.cpu.util[,nice,avg1]] received value [0.003565]
14435:20110211:205318 XML before sending [<req><host>c2VydmVyMDIub3BsZXhpbmcuY29t</host><key>c3lzdGVtLmNwdS51dGlsWyxuaWNlLGF2ZzFd</key><data>MC4wMDM1NjU=</data></req>]
14435:20110211:205318 Timeout while answering request
14435:20110211:205318 Send value error: [recv] ZBX_TCP_READ() failed [Interrupted system call]
14435:20110211:205318 In get_min_nextcheck()
14435:20110211:205318 No sleeping

[nelsonab]

Hmmm, do a tcpdump if you can, I have a hunch you're sending the SYN, but not receiving the SYN+ACK. It should look like a one way communication in the tcpdump.