On Fri, Aug 1, 2008 at 03:35, jijun gao <[EMAIL PROTECTED]> wrote:
> hi, Andreas
>>very short interval and timeout
>>> *Jul 31 16:24:37 node2 last message repeated 9 times
>>>Jul 31 16:24:37 node2 setroubleshoot: SELinux is preventing ifconfig
>>>(ifconfig_t) "read write" to socket:[136168] (initrc_t). For complete
>>> SELinux messages. run sealert -l 0db84664-2bd3-4f8f-a10e-1e0641417484
>
>>hmmm ... I'm not familiar with SELinux, but that looks suspicious to
>>me. I assume on node1 SELinux is disabled?
>
> actually, on node1 SELinux is enabled, but I don't find similar log
> iinformation on node1,
> anyway, the two nodes don't have completely the same software environment,
> and I
> disable SELinux on node2.
>
>>> Jul 31 16:24:37 node2 lrmd: [29544]: WARN: asterisk_2:monitor process
> (PID
>>> 23374) timed out (try 1). Killing with signal SIGTERM (15).
>
>>... and because of the monitoring timeout the resource is declared
>>dead and restarted.
>
> you got it. when I set timeout=10, resources don't restart as used to.
> but I am still not quite sure what timeout mean.
it means that the operation has 10s to complete before we assume it failed
and the operation is performed every {interval} seconds.
> here is my understanding:
> so the moniter action, actually, it's a process that run again and again,
> and the process takes some time to execute, and every interval time,
> a new process runs. Is that true?
>
> still, there is something else I don't understand.
> why the 'restarting' only happens on the standby node?
> (as far as I know, it has nothing to do with SELinux)
>
> Thanks a bounch
> _______________________________________________
> Linux-HA mailing list
> [email protected]
> http://lists.linux-ha.org/mailman/listinfo/linux-ha
> See also: http://linux-ha.org/ReportingProblems
>
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
