Re: FW: problema en un nodo de postgres en sun solaris cluster failover

Seria mejor si te envía el log de postgresql para ver cual es el error.

2009/7/8 Yolanda Sanchez <fysanchez(at)hotmail(dot)com>:
> Hola Alvaro,
>
>
>
> Yo soy nueva en Postgres, pero el operador  me reporto esta informacion que
> proviene del Orion  syslog.
>
> El ip del nodo del postgres es 192.168.28.33.
>
>
>
> El operador me indica lo sgte. que me podria aconsejar.
>
>
>
>
>
> The message is something like this:
>
> ========================================================
>
> 04:33 PM  192.168.28.33  Notice  SC[SUNW.gds:6,PgeGroup,RS-PGS2,gds_probe]:
> [ID 646865 daemon.notice] Monitor probe time of 18.00 seconds is 59.98
> percent of Probe timeout.
>
> ========================================================
>
>
>
> We have been researching the error, and so far we have pointers to believe
> something on the Postgres configuration might be slowing down the Probe
> Interfaces on the Cluster (these Probes are not related to the LOC-AID
> Probes. They are 'Cluster Probes')
>
> If these probes slow down to a certain point, the Cluster could potentially
> start initiating failovers, that will result in downtime. This problem must
> be fixed.
>
>
>
> Since we cannot play much with the production cluster, we are going to
> schedule a Maintenance Window to failover and restart both nodes, hoping for
> a fresh state after restart.
>
> We plan to take as much measurement data as we can, before and after the
> restart.
>
> Please advise what data you would need in order to contribute to the
> analysis/resolution of the problem.
>
>
>
> Saludos,
>
>
>
> Yolanda
>
>
>
> ===
>
>
>
>> Date: Wed, 8 Jul 2009 12:39:19 -0400
>> From: alvherre(at)alvh(dot)no-ip(dot)org
>> To: fysanchez(at)hotmail(dot)com
>> CC: pgsql-es-ayuda(at)postgresql(dot)org
>> Subject: Re: [pgsql-es-ayuda] problema en un nodo de postgres en sun
>> solaris cluster failover
>>
>> Yolanda Sanchez escribió:
>> >
>> > Hola,
>> >
>> > Tengo un error en un nodo de Postgres que esta configurado en Cluster
>> > failover.
>> >
>> > 04:33 PM 192.168.28.33 Notice SC[SUNW.gds:6,PgeGroup,RS-PGS2,gds_probe]:
>> > [ID 646865 daemon.notice] Monitor probe time of 18.00 seconds is 59.98
>> > percent of Probe timeout.
>> > 04:32 PM 192.168.28.33 Notice SC[SUNW.gds:6,PgeGroup,RS-PGS2,gds_probe]:
>> > [ID 646865 daemon.notice] Monitor probe time of 15.51 seconds is 51.69
>> > percent of Probe timeout.
>> > 04:29 PM 192.168.28.33 Notice SC[SUNW.gds:6,PgeGroup,RS-PGS2,gds_probe]:
>> > [ID 646865 daemon.notice] Monitor probe time of 22.52 seconds is 75.07
>> > percent of Probe timeout.
>> > 04:22 PM 192.168.28.33 Notice SC[SUNW.gds:6,PgeGroup,RS-PGS2,gds_probe]:
>> > [ID 646865 daemon.notice] Monitor probe time of 18.00 seconds is 59.98
>> > percent of Probe timeout.
>>
>> Estos no son mensaje de Postgres.
>>
>> --
>> Alvaro Herrera http://www.flickr.com/photos/alvherre/
>> Jude: I wish humans laid eggs
>> Ringlord: Why would you want humans to lay eggs?
>> Jude: So I can eat them
>
> ________________________________
> Explore the seven wonders of the world Learn more!
> ________________________________
> Discover the new Windows Vista Learn more!

--
Atentamente,
Moisés Alberto Lindo Gutarra
Asesor - Desarrollador Java / Open Source
Linux Registered User #431131 - http://counter.li.org/
Cel: (511) 995081720
MSN: mlindo(at)tumisolutions(dot)com