On Mon, Apr 12, 2010 at 12:41 PM, Alain.Moulle <[email protected]> wrote:
> Because I need that at the very first time I do //etc/init.d/corosync
> start/ the time to be available
> on only one node of the HA cluster_ is reduced to minimum_, and to get
> this, I need
> to have values of 5s for both (I've tested after modification via CRM
> API and stop and start again, it starts
> fine with 5s) , so then I can configure all the things via CRM API and
> end the configuration
> by re-set the dc-deadtime and cluster-delay to their defaults values
> (60s) so that Pacemaker will
> then work correctly as usual. I mean the setting of 5s is of course not
> to be used once
> the configuration is ended (I know that HA will not work with these
> values),
really?
> it's just to
> set the configuration on 1 node of the HA cluster the more quickly
> possible.
You mean when there is only one node?
It shouldn't have any effect when a new node joins an active cluster.
Anyway, if you must, then look for crmd_opts in crmd/control.c
Around line 714 or so.
> Regards
> Alain
>
>> > Thanks for your response Andrew.
>> >
>> > Further questions :
>> > I would like to change /*cluster-delay*/ and /*dc-deadtime*/ default
>> > values in sources,
>
>>
>> why?
>> > I found cluster-delay default value in common.c in pacemaker pengine
>> part :
>> > pacemaker/lib/pengine/common.c :
>> > ? { "cluster-delay", "transition_idle_timeout", "time", NULL, "60s",
>> > &check_time,
>> >
>> > but can't find in pacemaker or heartbeat or ... etc. where I could
>> > change the
>> > dc-deadtime default value ?
>> >
>> > Thanks a lot
>> > Regards
>> > Alain
>
> _______________________________________________
> 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
