Re: Database system was interrupted. Possible reasons for a database to suddenly stop accepting connections?

From: Laurenz Albe <laurenz(dot)albe(at)cybertec(dot)at>
To: "Buzenets, Yuriy (GE Renewable Energy, consultant)" <Yuriy(dot)Buzenets(at)ge(dot)com>, "pgsql-general(at)postgresql(dot)org" <pgsql-general(at)postgresql(dot)org>
Subject: Re: Database system was interrupted. Possible reasons for a database to suddenly stop accepting connections?
Date: 2020-11-09 16:58:35
Message-ID: 8ee6aa4afa4e068efe9a0153804a0baf8514174a.camel@cybertec.at
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

On Mon, 2020-11-09 at 13:53 +0000, Buzenets, Yuriy (GE Renewable Energy, consultant) wrote:

If I delete all the noise from the log, file, this remains:

> < 2020-10-29 11:51:59.345 PDT >STATEMENT: SELECT NULL AS TABLE_CAT, n.nspname AS TABLE_SCHEM, [...]

> < 2020-10-29 12:04:09.700 PDT >LOG: database system was interrupted; last known up at 2020-10-29 12:03:16 PDT
> < 2020-10-29 12:04:09.812 PDT >LOG: database system was not properly shut down; automatic recovery in progress
> < 2020-10-29 12:04:09.817 PDT >LOG: redo starts at 43C/E887E1C8
> < 2020-10-29 12:04:09.916 PDT >LOG: record with zero length at 43C/E8ED94D8
> < 2020-10-29 12:04:09.917 PDT >LOG: redo done at 43C/E8ED94A8
> < 2020-10-29 12:04:09.917 PDT >LOG: last completed transaction was at log time 2020-10-29 12:03:46.422937-07
> < 2020-10-29 12:04:09.938 PDT >LOG: MultiXact member wraparound protections are now enabled
> < 2020-10-29 12:04:09.941 PDT >LOG: autovacuum launcher started
> < 2020-10-29 12:04:09.941 PDT >LOG: database system is ready to accept connections

> < 2020-10-29 12:26:44.453 PDT >LOG: received fast shutdown request
> < 2020-10-29 12:26:44.453 PDT >LOG: aborting any active transactions
> < 2020-10-29 12:26:44.544 PDT >LOG: shutting down
> < 2020-10-29 12:26:44.587 PDT >LOG: database system is shut down

> < 2020-10-29 12:27:13.156 PDT >LOG: database system was shut down at 2020-10-29 12:26:44 PDT
> < 2020-10-29 12:27:13.178 PDT >LOG: MultiXact member wraparound protections are now enabled
> < 2020-10-29 12:27:13.179 PDT >LOG: autovacuum launcher started
> < 2020-10-29 12:27:13.179 PDT >LOG: database system is ready to accept connections
> < 2020-10-29 15:02:42.588 PDT >LOG: received fast shutdown request
> < 2020-10-29 15:02:42.588 PDT >LOG: aborting any active transactions
> < 2020-10-29 15:02:42.712 PDT >LOG: shutting down
> < 2020-10-29 15:02:42.732 PDT >LOG: database system is shut down

> < 2020-10-29 15:02:43.711 PDT >LOG: database system was shut down at 2020-10-29 15:02:42 PDT
> < 2020-10-29 15:02:43.714 PDT >LOG: MultiXact member wraparound protections are now enabled
> < 2020-10-29 15:02:43.715 PDT >LOG: autovacuum launcher started
> < 2020-10-29 15:02:43.716 PDT >LOG: database system is ready to accept connections

There was a operating system crash or a "kill -9" somewhere between 11:51 and 12:04 on Oct. 29.

PostgreSQL got restarted and recovered at 12:04.

Then there were two more clean shutdowns and restarts at 12:26 and 15:02.

Your problem is probably the first crash. If you don't have any indication that the machine crashed,
look into the kernel log - perkaps the out-of-memory killer struck (assuming this is Linux).

Yours,
Laurenz Albe
--
Cybertec | https://www.cybertec-postgresql.com

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Matt Zagrabelny 2020-11-09 17:00:23 initdb --data-checksums
Previous Message David G. Johnston 2020-11-09 16:14:08 Re: How to set up a schema default date to '2020-01-01'?