Re: Crash recovery

On 12 May 2014, at 16:52, Sergey Maslyakov <address@hidden> wrote:

Is there any noticeable performance impact as the number of processes grows? Let's say, if I have 200+ processes running on a box, how long does it take to scan them? My primary OS of concern is Solaris on both, SPARC and X86.

Are there any performance numbers published around this?

Regards,
/Sergey

On Sat, May 10, 2014 at 2:10 AM, Martin Pala <address@hidden> wrote:

yes

On 09 May 2014, at 22:39, Sergey Maslyakov <address@hidden> wrote:

Thank you, Martin! From what I found in the source code afters seeing your response, Monit periodically scans the process list and performs pattern matching to determine whether a certain monitored service is running or not. It also checks whether the PID still exists.

Am I on the right track?

Regards,
/Sergey

On Fri, May 9, 2014 at 11:02 AM, Martin Pala <address@hidden> wrote:

Hello,

if Monit is restarted it checks the current state of monitored services before it does any action - if the processes are running already, test is successful and no restart action is done (i.e. no second set of services).

Regards,
Martin

On 09 May 2014, at 14:12, Sergey Maslyakov <address@hidden> wrote:

> Hello,
>
> What is the expected behavior of Monit daemon upon restart after a crash or other kind of abnormal termination? Is it capable of harvesting the process list for the processes started by the predecessor and taking ownership of monitoring them? Or does it just try to spawn a set of [duplicate] services?
>
>
> Thank you,
> /Sergey

> --
> To unsubscribe:
> https://lists.nongnu.org/mailman/listinfo/monit-general

--
To unsubscribe:
https://lists.nongnu.org/mailman/listinfo/monit-general

--
To unsubscribe:
https://lists.nongnu.org/mailman/listinfo/monit-general

--
To unsubscribe:
https://lists.nongnu.org/mailman/listinfo/monit-general

--
To unsubscribe:
https://lists.nongnu.org/mailman/listinfo/monit-general

From:	Martin Pala
Subject:	Re: Crash recovery
Date:	Tue, 13 May 2014 18:24:51 +0200