Re: Failure to create GiST on ltree column

From: Justin Pryzby <pryzby(at)telsasoft(dot)com>
To: Victor Yegorov <vyegorov(at)gmail(dot)com>
Cc: pgsql-hackers(at)lists(dot)postgresql(dot)org
Subject: Re: Failure to create GiST on ltree column
Date: 2020-05-24 22:52:48
Message-ID: 20200524225248.GS4472@telsasoft.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Sun, May 24, 2020 at 09:30:15PM +0300, Victor Yegorov wrote:
> Greetings.
>
> I am getting random failures in `CREATE INDEX USING gist` over ltree column
> while performing pg_restore.
> I get either
> ERROR: stack depth limit exceeded
> or
> ERROR: failed to add item to index page
>
> Thing is — if I retry index creation manually, I get it successfully built
> in ~50% of the cases.
>
> I would like to find out what's the real cause here, but I am not sure how
> to do it.
> If anybody could provide some guidance, I am open to investigate this case.
>
> I'm on PostgreSQL 11.8 (Debian 11.8-1.pgdg90+1), debugging symbols
> installed.

I think you'd want to attach a debugger to the backend's PID and set breakpoint
on errfinish() or pg_re_throw() and reproduce the problem to get a stacktrace
(separate stacks for both errors).
https://wiki.postgresql.org/wiki/Getting_a_stack_trace_of_a_running_PostgreSQL_backend_on_Linux/BSD

What is value of maintenance_work_mem ?

What's the definition of the index and relevant table columns ?

Do you know if that was that an issue under 11.7 as well ?

Are you running on any interesting hardware ?

--
Justin

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Atsushi Torikoshi 2020-05-25 01:54:05 Re: Is it useful to record whether plans are generic or custom?
Previous Message Michael Paquier 2020-05-24 22:44:09 WAL reader APIs and WAL segment open/close callbacks