Re: BUG #17755: database queries get stuck for certain IDs

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: pavol(dot)sekeres8(at)gmail(dot)com
Cc: pgsql-bugs(at)lists(dot)postgresql(dot)org
Subject: Re: BUG #17755: database queries get stuck for certain IDs
Date: 2023-01-20 17:19:15
Message-ID: 532118.1674235155@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs

PG Bug reporting form <noreply(at)postgresql(dot)org> writes:
> we are using PostgreSQL version 9.6.8 on our production environment which we
> operate ourselves.

You realize, I hope, that you are running a very old minor release
of a branch that's been out of support entirely for more than a year.
You might consider consulting our release note archives

https://www.postgresql.org/docs/release/

to see if your problem matches up with any of the large number of
post-9.6.8 fixes.

> We noticed that the insert/select queries which get stuck and run infinitely
> have specific request_id - starting with 'a6bda' for example
> 'a6bda748-0e31-4da6-8fed-331c6c4a7cdb'.

Smells a little bit like index corruption, perhaps. I'd try reindexing
whatever index(es) exist on that column. But if that doesn't fix it ...

> Do you know what could be the cause for our problems and possibly suggest
> any solution?

Running ancient minor releases is fundamentally unsound database
administration practice. At minimum you should be running the last
9.6.x minor release (9.6.24); but really now would be a good time
to be thinking about upgrading to some still-in-support branch.
If there is an actual Postgres problem here, and it wasn't fixed
by 9.6.24, we aren't ever going to fix it in that branch.

regards, tom lane

In response to

Browse pgsql-bugs by date

  From Date Subject
Next Message Sergey Belyashov 2023-01-20 17:45:57 Re: BUG #17756: Invalid replica indentity set order in a dump
Previous Message Tom Lane 2023-01-20 15:42:10 Re: BUG #17756: Invalid replica indentity set order in a dump