Re: NULL pointer dereference in syslogger with load_libraries() and -DEXEC_BACKEND at startup

From: Michael Paquier <michael(at)paquier(dot)xyz>
To: Álvaro Herrera <alvherre(at)kurilemu(dot)de>
Cc: Anton Voloshin <a(dot)voloshin(at)postgrespro(dot)ru>, Euler Taveira <euler(at)eulerto(dot)com>, Kyotaro Horiguchi <horikyota(dot)ntt(at)gmail(dot)com>, pgsql-hackers(at)lists(dot)postgresql(dot)org
Subject: Re: NULL pointer dereference in syslogger with load_libraries() and -DEXEC_BACKEND at startup
Date: 2026-07-28 02:49:58
Message-ID: amgY1jGjkr5Rmk3V@paquier.xyz
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Sun, Jul 26, 2026 at 05:33:39PM +0900, Michael Paquier wrote:
> +1. I'll check the state of my module later.. I suspect that we have
> other things out there that will be broken. Or perhaps not.

There was a total of three bugs in jsonlog. For reference, perhaps
that will help others that depend on the elog hook:
- Two code paths had the idea to push data to the syslogger too early.
- Extra one I did not suspect: MyBgworkerEntry may be consulted
earlier now.

Using syslogger_setup_done for the first bullet point, combined with a
safety check based on MyBgworkerEntry for the second bullet point is
making the module able to work with v19 onwards.
--
Michael

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Michael Paquier 2026-07-28 02:58:32 Re: Add pg_stat_kind_info system view
Previous Message Tender Wang 2026-07-28 02:36:10 Re: Partition pruning can incorrectly exclude a RANGE default partition