[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: monitoring status patch (Oliver's patch from 3.2.2003)
From: |
Oliver Jehle |
Subject: |
Re: monitoring status patch (Oliver's patch from 3.2.2003) |
Date: |
11 Feb 2003 15:45:06 +0100 |
so i dont have to monit start process to re-monitor a always running
process after a monit restart ???
i think that the statefull configuration reload should be backed by a
file where the state of monitoring is written and read from after a
restart......
example:
if monit starts/reload -> check if state-file exists and
if process running, set the state
from file
if monit starts/stop process -> write actual state to file
if monit stops normally -> delete state file
the state file should contain the boot-time of the server. if the boot
time is > the boot time in the state file, then there was a server -
crash.... and start as normal startup....
On Tue, 2003-02-11 at 15:13, Martin Pala wrote:
> Oliver Jehle wrote:
>
> >
> >-1 for the do_validate...
> >
> >i think, monit should always monitor a process when its running...
> >
> >if monit dies and restarts all processes are in status not monitoring
> >already running and thats bad.
> >
> >
> >
> >
> I forgot yet one :)
>
> Your patch is in place so now monit behaves as you requested by default
> (e.g. if the process is running, start its monitoring in the case of
> monit reload).
>
> After we will make statefull configuration reload support, it will work
> for you similar way (all your processes which will have monitoring
> enabled before reload will be monitored after reload too, even if they
> are in manual mode).
>
>
>
>
>
> _______________________________________________
> monit-dev mailing list
> address@hidden
> http://mail.nongnu.org/mailman/listinfo/monit-dev