[Pgcluster-general] Problem with two replicators

[John Gardner]

Rick Vernam wrote:
> can you run both replicators with -nv, and send the logs just up until it 
> fails?
> I usually run like this
> pgreplicate -D $PGDATA -W $PGDATA -nv > pgr.log &
> then, to watch the logs real-time, I run tail -f pgr.log
> 
> anyway, if you could attach something like pgr.log from the above example, 
> that would help...

Rick

I have attached the two logs for the upper (primary) and
lower(secondary) replicators.

I made sure that everything was running as normal and then stopped the
upper replicator manually, shown in the log as:

2007-05-01 11:22:59 [19674] DEBUG:quick_exit:signo = 15
2007-05-01 11:22:59 [19553] DEBUG:quick_exit:signo = 15
2007-05-01 11:22:59 [19552] DEBUG:quick_exit:signo = 15
2007-05-01 11:22:59 [19551] DEBUG:quick_exit:signo = 15

Unfortunately, it appears that nothing about the upper replicator being
stopped was recorded in the lower log, and the system fell over again.

Hopefully you will be able to spot some obvious problems in the log.

Thanks for you help

John
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: pgr_upper.log
Url: http://pgfoundry.org/pipermail/pgcluster-general/attachments/20070501/2d593d4b/attachment-0002.ksh 
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: pgr_lower.log
Url: http://pgfoundry.org/pipermail/pgcluster-general/attachments/20070501/2d593d4b/attachment-0003.ksh