Re: Better client reporting for "immediate stop" shutdowns

Tom Lane <tgl@sss.pgh.pa.us>

From: Tom Lane <tgl@sss.pgh.pa.us>
To: Andres Freund <andres@anarazel.de>
Cc: pgsql-hackers@lists.postgresql.org
Date: 2020-12-29T23:05:20Z
Lists: pgsql-hackers
I wrote:
> I had a thought while looking at elog.c: we could further reduce the risk
> of quickdie() crashing if we make it do what elog.c does when it gets into
> error recursion trouble:
>             error_context_stack = NULL;
>             debug_query_string = NULL;

On closer inspection, there's not much need to touch debug_query_string
here, because elog.c only consults that for making log entries, which
we're suppressing.  I pushed it with just the error_context_stack reset.

			regards, tom lane



Commits

  1. Suppress log spam from multiple reports of SIGQUIT shutdown.

  2. Improve client error messages for immediate-stop situations.