From: Kouber Saparev <kouber@gmail.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Subject: VACUUM unable to accomplish because of a non-existent MultiXactId
Message-Id: <62FE8F79-7278-4AE0-9FDE-0C0277A7B609@gmail.com>
Date: Fri, 27 Nov 2015 20:25:49 +0200
To: pgsql-admin@postgresql.org
Mime-Version: 1.0 (Mac OS X Mail 9.1 \(3096.5\))
Precedence: bulk
Sender: pgsql-admin-owner@postgresql.org

Hello,

Recently I spotted an auto vacuum that is constantly being run over a pg_to=
ast table in the database. Interestingly the underlying table did not have =
that amount of writes that would trigger the auto vacuum every minute.

When I tried to run the VACUUM manually it died with a non-existent MultiXa=
ctId error:

db=3D# vacuum analyze verbose pg_toast.pg_toast_376621;
INFO:  vacuuming "pg_toast.pg_toast_376621"
ERROR:  MultiXactId 2915905228 does no longer exist -- apparent wraparound

db=3D# select txid_current();
 txid_current
--------------
   2583853583
(1 row)

db=3D# select datfrozenxid from pg_database where datname=3D=E2=80=98db';
datfrozenxid
--------------
   2161848861
(1 row)


We did a pg_upgrade about a month ago to upgrade from 9.2 to 9.4, so I gues=
s there might be some relation, as I do not remember having that issue befo=
re the upgrade.

The question is, how can I clean the table in question and finally be able =
to run a vacuum over it?

Regards,
=E2=80=94
Kouber Saparev

--=20
Sent via pgsql-admin mailing list (pgsql-admin@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin