[pgpool-general: 7578] Re: jboss - jdbc - pgpool

[Tatsuo Ishii]

> Hello Tatsuo,
> Thank you for your answer. I checked pgpool log and i found such errors:
> 2021-05-18T09:13:15.974748+02:00 pgs00000205 pgpool[18000]: [18-1] 2021-05-18 09:13:15: pid 18000: FATAL:  connection was terminated due to conflict with recovery
> 2021-05-18T09:13:15.975235+02:00 pgs00000205 pgpool[18000]: [18-2] 2021-05-18 09:13:15: pid 18000: DETAIL:  User was holding a relation lock for too long.
> 2021-05-18T09:13:15.975436+02:00 pgs00000205 pgpool[18000]: [18-3] 2021-05-18 09:13:15: pid 18000: HINT:  In a moment you should be able to reconnect to the database and repeat your command.
> 2021-05-18T09:13:15.978130+02:00 pgs00000205 pgpool[17201]: [19-1] 2021-05-18 09:13:15: pid 17201: LOG:  child process with pid: 18000 exits with status 256
> 2021-05-18T09:13:15.978502+02:00 pgs00000205 pgpool[17201]: [20-1] 2021-05-18 09:13:15: pid 17201: LOG:  fork a new child process with pid: 21211
> 2021-05-18T09:13:15.979342+02:00 pgs00000205 pgpool[17534]: [40-1] 2021-05-18 09:13:15: pid 17534: FATAL:  connection was terminated due to conflict with recovery
> 2021-05-18T09:13:15.979515+02:00 pgs00000205 pgpool[17534]: [40-2] 2021-05-18 09:13:15: pid 17534: DETAIL:  User was holding a relation lock for too long.
> 2021-05-18T09:13:15.979642+02:00 pgs00000205 pgpool[17534]: [40-3] 2021-05-18 09:13:15: pid 17534: HINT:  In a moment you should be able to reconnect to the database and repeat your command.
> 2021-05-18T09:13:15.981441+02:00 pgs00000205 pgpool[17201]: [21-1] 2021-05-18 09:13:15: pid 17201: LOG:  child process with pid: 17534 exits with status 256
> 2021-05-18T09:13:15.982657+02:00 pgs00000205 pgpool[17201]: [22-1] 2021-05-18 09:13:15: pid 17201: LOG:  fork a new child process with pid: 21212
> 2021-05-18T09:13:15.984406+02:00 pgs00000205 pgpool[18310]: [18-1] 2021-05-18 09:13:15: pid 18310: FATAL:  connection was terminated due to conflict with recovery
> database and repeat your command.
> I have synchronous replication set on my two-node postgresql cluster (synchronous_commit parameter has a value of "remote_apply") - may it be a matter of synchronization problems between postgresql nodes? I haven't noticed any delays.

I don't think you have synchronization problems. Maybe you have a long
running transaction on standby? Anyway this is a problem of usage for
PostgreSQL, not Pgpool-II. I suggest to check the PostgreSQL log when
the error above happened to find out what transaction/query is
involved.

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