|
From: | Jan-Henrik Haukeland |
Subject: | Re: Monit reliability in monitoring Apache |
Date: | Mon, 17 Jul 2006 19:13:45 +0200 |
On 17. jul. 2006, at 18.22, Peter M. Abraham wrote:
As shared in the original post, monit reports the failure. This is not a manual check, but that when monit states there is a failure, everything on the server level is fine including the PID and process tree.
Have you checked the monit and apache logs and can confirm that apache was not restarted by monit? From your configuration, both of these tests can restart apache if they should fail. I believe this is the reason monit fires of that alert mail or even more likely that apache in fact did not run and monit started it again.
> if failed port 80 protocol http then restart > if loadavg (1min) > 5 then restart
Monit does not randomly fire of alert mails unless there is a reason. However if can save the pid number in apache's pid file and if the pid file has the same pid number _after_ you get the alert mail from monit then I believe you have found a bug, although improbable :-)
-- Jan-Henrik Haukeland Mobil +47 97141255
[Prev in Thread] | Current Thread | [Next in Thread] |