Re: [External]Re: Postgres Crash Issue - Mailing list pgsql-general

From Tom Lane
Subject Re: [External]Re: Postgres Crash Issue
Date
Msg-id 316110.1647353866@sss.pgh.pa.us
Whole thread Raw
In response to Re: [External]Re: Postgres Crash Issue  (hubert depesz lubaczewski <depesz@depesz.com>)
List pgsql-general
hubert depesz lubaczewski <depesz@depesz.com> writes:
> On Tue, Mar 15, 2022 at 01:30:25PM +0000, Menon, Deepak (Deepak) wrote:
>> Then what does this message mean ?
>>
>> 2022-03-11 08:58:42.956 UTC [17115] DETAIL:  The postmaster has commanded this server process to roll back the
currenttransaction and exit, because another server process exited abnormally and possibly corrupted shared memory.  

> When someone issues pg_ctl -m immediate stop (which is the immediate
> shutdown). pg basically shutd backends in their head. This, in turn,
> means that all other have to exit to avoid shared memory corruption.

FWIW, v14 and later produce more on-point error reporting in this
case.  The code path used to be the same as for a backend crash,
but now clients see something like

WARNING:  terminating connection due to immediate shutdown command
server closed the connection unexpectedly
        This probably means the server terminated abnormally
        before or while processing the request.

and the server log isn't spammed with copies of that, it just
looks like

2022-03-15 10:07:48.241 EDT [315456] LOG:  received immediate shutdown request
2022-03-15 10:07:48.244 EDT [315456] LOG:  database system is shut down

None of this gets us any closer to understanding where the signal
came from, unfortunately.

            regards, tom lane



pgsql-general by date:

Previous
From: hubert depesz lubaczewski
Date:
Subject: Re: [External]Re: Postgres Crash Issue
Next
From: Tom Lane
Date:
Subject: Re: Re[2]: LISTEN/NOTIFY ultra slow speed