Re: oldest xmin is far in the past :: BUT xmin is not available in system - Mailing list pgsql-general

From Rob Sargent
Subject Re: oldest xmin is far in the past :: BUT xmin is not available in system
Date
Msg-id 9b2148e7-0925-9516-9dad-228c22163a1d@gmail.com
Whole thread Raw
In response to Re: oldest xmin is far in the past :: BUT xmin is not available in system  ("David G. Johnston" <david.g.johnston@gmail.com>)
Responses Re: oldest xmin is far in the past :: BUT xmin is not available in system  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-general
On 4/19/22 00:06, David G. Johnston wrote:


On Monday, April 18, 2022, bhargav kamineni <kbn98406@gmail.com> wrote:
Hi Team,

It seems vacuum is behaving somewhat weird on postgres database , observing below HINTS on the vacuum logs

WARNING:  oldest xmin is far in the past

HINT:  Close open transactions soon to avoid wraparound problems.

You might also need to commit or roll back old prepared transactions, or drop stale replication slots.


What version?

What other databases are present?

Others can give better (more detailed/nuanced) guidance but if you can just start vacuuming every table in every database manually, you probably should just do that.  Vacuum freeze specifically.

David J.

This site has lots of useful queries for this sort of issue:
https://www.shanelynn.ie/postgresql-find-slow-long-running-and-blocked-queries.  I think you're looking for a very old transaction that is probably not going to finish, must be terminated.

pgsql-general by date:

Previous
From: bhargav kamineni
Date:
Subject: Re: oldest xmin is far in the past :: BUT xmin is not available in system
Next
From: Pól Ua L.
Date:
Subject: Joining with calendar table