[pgpool-general: 1577] Re: pgpool 3.2.3 node gets down after some hours of work.

[Dzmitry]

Yes, I've upgraded to 3.2-stable.
Also I found before pgpool show node as down, it was recovery conflict in
the node for some times(4-5):

2013-04-07 19:38:10 UTC FATAL:  terminating connection due to conflict
with recovery
2013-04-07 19:38:10 UTC DETAIL:  User query might have needed to see row
versions that must be removed.
2013-04-07 19:38:10 UTC HINT:  In a moment you should be able to reconnect
to the database and repeat your command.
2013-04-07 19:38:10 UTC LOG:  could not send data to client: Connection
reset by peer
2013-04-07 19:38:10 UTC ERROR:  canceling statement due to conflict with
recovery
2013-04-07 19:38:10 UTC DETAIL:  User query might have needed to see row
versions that must be removed.
2013-04-07 19:38:10 UTC LOG:  could not send data to client: Broken pipe
2013-04-07 19:38:10 UTC FATAL:  connection to client lost

Can it be the case ?


No, I do not see pool_log("postmaster on DB node %d was shutdown by
administrative command", i);

Thanks,
  Dzmitry





On 4/8/13 1:46 AM, "Tatsuo Ishii" <ishii at postgresql.org> wrote:

>> May be this will help:
>> 
>> 2013-04-07 11:48:52 LOG:   pid 9785: do_child: exits with status 1 due
>>to
>> error
>> 2013-04-07 11:48:52 LOG:   pid 9760: pool_read_kind: error message from
>>2
>> th backend:canceling statement due to conflict with recovery
>> 2013-04-07 11:48:53 LOG:   pid 9767: detect_stop_postmaster_error:
>> detect_error error
>> 2013-04-07 11:48:53 LOG:   pid 9767: detect_postmaster_down_error
>>returns
>> error on backend 2. Going to close this session.
>> 2013-04-07 11:59:28 LOG:   pid 9651: do_child: exits with status 1 due
>>to
>> error
>
>Those themselves do not related to node down or fail over.
>
>Anyway are yo still using 3.2.3? or upgraded to 3.2-stable?
>
>Also do yo see this kind of message?
>
>	 pool_log("postmaster on DB node %d was shutdown by administrative
>command", i);
>
>If you see this, someone used pg_terminate_backend(). This makes
>pgpool confuses because it returned exactly same error code as
>postmaster shutdown. So pgpool will detach the node where the function
>issued.
>--
>Tatsuo Ishii
>SRA OSS, Inc. Japan
>English: http://www.sraoss.co.jp/index_en.php
>Japanese: http://www.sraoss.co.jp