[pgpool-general: 2938] Re: configurable timeout for 1st stage recovery

[Tatsuo Ishii]

Hi Sean,

>> So this is a dilemma.  statement_timeout can be set on a
>> per-connection basis, but I believe it is not possible to do it for
>> the connection that pgpool uses to run the 1st stage recovery command.
>> If I am wrong about that then please let me know, but otherwise I
>> think it would be extremely useful to have a configuration property to
>> set the statement_timeout on that particular connection.

I have committed a fix following your direction, i.e. disable
statement timeout in the connection to run the online recovery. Thanks
for the hint!

The commit for 3.3-stable is here:
http://git.postgresql.org/gitweb/?p=pgpool2.git;a=commitdiff;h=b9bb86d3bb69873f2e1b441820a02b632f00e3e6

You can get patch agaist pgpool-II 3.3.3 by following "patch" link.

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

> Hi Sean,
> 
> Sounds like a resonal complain. I will look into this issue.
> 
> Best regards,
> --
> Tatsuo Ishii
> SRA OSS, Inc. Japan
> English: http://www.sraoss.co.jp/index_en.php
> Japanese:http://www.sraoss.co.jp
> 
>> Hi Tatsuo,
>> 
>> You may remember some time ago I had a problem with PostgreSQL's
>> statement_timeout interfering with (actually, breaking) the 1st stage
>> of recovery.  It was worked around by increasing the statement_timeout
>> to longer than a recovery might take.
>> 
>> Unfortunately our system has grown to the point that a recovery takes
>> almost two hours, while at the same time we realize the
>> statement_timeout needs to be small (10 minutes) to reduce problems
>> from runaway queries.
>> 
>> So this is a dilemma.  statement_timeout can be set on a
>> per-connection basis, but I believe it is not possible to do it for
>> the connection that pgpool uses to run the 1st stage recovery command.
>> If I am wrong about that then please let me know, but otherwise I
>> think it would be extremely useful to have a configuration property to
>> set the statement_timeout on that particular connection.
>> 
>> I'm sorry that I am not quite comfortable enough hacking the source
>> myself.  Is this a request you could accommodate?
>> 
>> Thanks very much,
>> Sean
> _______________________________________________
> pgpool-general mailing list
> pgpool-general at pgpool.net
> http://www.pgpool.net/mailman/listinfo/pgpool-general