From: Tom Lane <tgl@sss.pgh.pa.us>
To: "Mansky,
    Edmund J. (GSFC-671.0)[ADNET Affiliate]" <edmund.j.mansky@nasa.gov>
cc: "pgsql-general@postgresql.org" <pgsql-general@postgresql.org>
Subject: Re: Why is an error not thrown when the time exceeds the lock timeout for an ExclusiveLock on a transaction ?
In-reply-to: <F55DDD73-286A-4432-941E-CB35AF6FD5B9@nasa.gov>
References: <F55DDD73-286A-4432-941E-CB35AF6FD5B9@nasa.gov>
Comments: In-reply-to "Mansky, Edmund J. (GSFC-671.0)[ADNET Affiliate]" <edmund.j.mansky@nasa.gov>
	message dated "Fri, 18 Apr 2025 17:28:26 -0000"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <2643499.1745000524.1@sss.pgh.pa.us>
Content-Transfer-Encoding: quoted-printable
Date: Fri, 18 Apr 2025 14:22:04 -0400
Message-ID: <2643500.1745000524@sss.pgh.pa.us>
Archived-At: <https://www.postgresql.org/message-id/2643500.1745000524%40sss.pgh.pa.us>
Precedence: bulk

"Mansky, Edmund J. (GSFC-671.0)[ADNET Affiliate]" <edmund.j.mansky@nasa.go=
v> writes:
> I see in the log at that point : process 683860 still waiting for ShareL=
ock on transaction 492777941 after 1000.140 ms
>                                               while updating tuple (1282=
991,25) in relation "sum_partn_alloc"

> The Postgres server (12.22), running on RHEL 8.10 is configured with a d=
efault lock timeout of 1 sec.

> Why is Postgres not throwing an error when the ShareLock time has exceed=
ed 1 sec. ?

I think you have misread the description of deadlock_timeout: it is
the lock wait time after which we check to see if there's a deadlock.
If there's not, we just log the above message (if configured to do so)
and keep waiting.

If you want to fail after X amount of time, lock_timeout or perhaps
statement_timeout is what to set for that.

			regards, tom lane