[pgpool-general: 3088] Re: Fwd: question about insert_lock

[Tatsuo Ishii]

> Hi,
> 
> Does anyone have a comment on this?
> 
> Thanks,
> Sean
> 
> 
> -------- Original Message --------
> Subject: 	question about insert_lock
> Date: 	Thu, 17 Jul 2014 13:36:13 -0230
> From: 	Sean Hogan <sean at compusult.net>
> To: 	pgpool-general at pgpool.net <pgpool-general at pgpool.net>
> 
> 
> 
> Hi,
> 
> I'm a little confused by the documentation of the insert_lock
> configuration property at
> http://www.pgpool.net/docs/latest/pgpool-en.html#replication_mode
> <http://www.pgpool.net/docs/latest/pgpool-en.html>. Specifically, I'm
> trying to determine if the default value (false) is also the
> recommended value.  I have created the insert_lock table and currently
> have insert_lock set to 'on'.  Is that the recommended thing for
> pgpool-II 3.3.3 in native replication mode against PostgreSQL 9.2.7?

Yes, it's the recommended way.
Best regards,

> (The reason: I'm still seeing sporadic, unexplained node degeneration,
> where pgpool-II detects a mismatch in the number of affected tuples.
> The applications run fine for a while, sometimes days, then suddenly
> it happens.  Most often it's an update or delete on a Quartz table.
> I'd like to eliminate sequence number mismatch as a reason for the
> failure.)

If you have a self contained test case, I could invest it.

Best regards,
--
Tatsuo Ishii
SRA OSS, Inc. Japan
English: http://www.sraoss.co.jp/index_en.php
Japanese:http://www.sraoss.co.jp