Re: oldest xmin is far in the past: PG 11 - Mailing list pgsql-admin

From Laurenz Albe
Subject Re: oldest xmin is far in the past: PG 11
Date
Msg-id 64f723e621f1f199441a1b9ee6d0c331926c1115.camel@cybertec.at
Whole thread Raw
In response to oldest xmin is far in the past: PG 11  (Ameen Abbas <abbas.dba@gmail.com>)
List pgsql-admin
On Wed, 2020-01-22 at 13:31 +0530, Ameen Abbas wrote:
> We are facing "oldest xmin is far in the past" in PostgreSQL 11 in logs, and we don't have
> any halted/aborted open transactions, and even no replication slots and no prepared transactions,
> we had ran vacuum verbose on the db multiple times still the below warning message not gone.

Ok, there are no prepared transactions and no replication slots.

Check for old transactions:

SELECT pid, age(current_timestamp, xact_start)
FROM pg_stat_activity
WHERE state <> 'idle';

Also, are there standby servers with "hot_standby_feedback = on"?
They might have a long running transaction as well.

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




pgsql-admin by date:

Previous
From: Ameen Abbas
Date:
Subject: oldest xmin is far in the past: PG 11
Next
From: "srikkanth"
Date:
Subject: SQL Query Syntax help