[pgpool-general: 2811] Re: pgpool 3.3 and watchdog

[Yugo Nagata]

Hi,

On Tue, 6 May 2014 10:54:08 +0200
Joar Jegleim <joar.jegleim at gmail.com> wrote:

> Hi list !
> 
> So I'm closing in on my 2 node postgres streaming replication setup.
> 
> I've advanced to my 'next step' where I now have 1 pgpool running on
> each of the 2 postgresql servers.
> I'm using watchdog and pgpool's heartbeat .
> 
> When I stop pgpool from pgpool admin gui, the other pgpool instance
> take over and bring up the delegate ip .
> 
> But if I simply pull the power cable from the master server  the other
> server doesn't take over, it never promotes itself to 'master pgpool
> instance' and doesn't bring up the delegate ip.
> The 'non active' pgpool node then doesn't promote the slave postgres
> to master either.
> 
> Is the pgpool 3.3 setup with watchdog and delegate ip  supposed to
> handle such situations, or do I have to setup some other services to
> handle automatic failover when primary node dies ?

This is odd. Could you please show me pgpool.conf and log messages?

> 
> 
> 
> -- 
> ----------------------
> Joar Jegleim
> Homepage: http://cosmicb.no
> Linkedin: http://no.linkedin.com/in/joarjegleim
> fb: http://www.facebook.com/joar.jegleim
> AKA: CosmicB @Freenode
> 
> ----------------------
> _______________________________________________
> pgpool-general mailing list
> pgpool-general at pgpool.net
> http://www.pgpool.net/mailman/listinfo/pgpool-general


-- 
Yugo Nagata <nagata at sraoss.co.jp>