[pgpool-hackers: 4211] Re: Deal with idle_session_timeout

[Tatsuo Ishii]

> Hi,
> 
> I have found an issue with idle_session_timeout which was introduced
> in PostgreSQL 14.0.  If this parameter is enabled in postgresql.conf
> and the timeout fires, Pgpool-II triggers failover if
> failover_on_backend_error is on (the default). This is not good.
> 
> This is a similar issue fixed in:
> https://git.postgresql.org/gitweb/?p=pgpool2.git;a=commit;h=3f5986eee360f12e6a0bb77aa46f95abf5f6bc10
> In the commit we dealt with idle-in-transaction timeout.
> 
> Attached patch fixes the issue in the same direction as the commit
> above: add detect_idle_session_timeout_error() to detect the error
> code for idle_session_timeout. If the error is detected, Pgpool-II
> returns the error code to frondend as a fatal error and disconnects
> the session.

Fix pushed through 4.0. In 3.7 issue with
idle_in_transaction_session_timeout was not fixed and I don't think we
need to fix idle_session_timeout problem in 3.7 either. Also 3.7 will
be EOL in coming November. So the extra work in 3.7 is not worth the
problem.

Best reagards,
--
Tatsuo Ishii
SRA OSS LLC
English: http://www.sraoss.co.jp/index_en/
Japanese:http://www.sraoss.co.jp