[Linux-HA] Reasonable values for timeouts

[Andrew Beekhof]

On 7/13/07, Dejan Muhamedagic <dejanmm at fastmail.fm> wrote:
>
> On Fri, Jul 13, 2007 at 12:23:16PM +0200, Andrew Beekhof wrote:
> > On 7/13/07, Max Hofer <max.hofer at apus.co.at> wrote:
> > >
> > >On Friday 13 July 2007, Andrew Beekhof wrote:
> > >> On 7/13/07, Max Hofer <max.hofer at apus.co.at> wrote:
> > >> >
> > >> > I agree with the experience dscribed by Eddie.
> > >> >
> > >> > For the 'monitor' keep in mind that the timeout should be lower
> than
> > >the
> > >> > interval. It does not make sense to start a 2nd monitor cycle when
> the
> > >> > first one did not finish.
> > >>
> > >>
> > >>
> > >> i _think_ that the interval is the time between one action ending and
> > >the
> > >> next one starting (rather than between both starting)
> > >>
> > >> at least i hope that
> > >Hmm. I always thought the cluster environment kills the processes which
> do
> > >not
> > >return withing the timeout period.
> >
> >
> > older versions were _supposed_ to
> > more recent versions do
> >
> >
> > >So what happens if i make:
> > >- timeout 30 seconds
> > >- monitor 20 seconds
> > >- the RA blocks on the monitor indefinitely
>
> I'm not sure what does "20 seconds" above stand for.



I believe he means an op such as:
   <op name=monitor interval=20s timeout=30s/>

and the question is, will the second invocation begin 20s after the first
began or 20s after the first was timed out.

At any rate,
> after the timeout has occured, lrmd sends the process (RA) the
> TERM signal. If the process is still around after five seconds,
> then it's sent the KILL signal. After another five seconds, if the
> process did not exit, lrmd complains about sending signals in vain
> and gives up on this process.
>
> > /me defers to dejan for a definitve answer
>
> Actually, this part was done by Alan.
>
> > _______________________________________________
> > Linux-HA mailing list
> > Linux-HA at lists.linux-ha.org
> > http://lists.linux-ha.org/mailman/listinfo/linux-ha
> > See also: http://linux-ha.org/ReportingProblems
>
> --
> Dejan
> _______________________________________________
> Linux-HA mailing list
> Linux-HA at lists.linux-ha.org
> http://lists.linux-ha.org/mailman/listinfo/linux-ha
> See also: http://linux-ha.org/ReportingProblems
>