Re: Query cancellation on hot standby because of buffer pins - Mailing list pgsql-admin

On 23 February 2015 at 08:19, Drazen Kacar <drazen.kacar@oradian.com> wrote:

> At the time they happened on the standby there was vacuuming of one
> table participating in the select query on the primary.

The VACUUM will have generated a WAL record that needs super exclusive
access to the block. Since feedback was enabled that record would not
have removed data visible by the query, but still needs to edit the
block.

The query was pinning that block, so this situation led to a delay on
the standby, which then led to cancellation of the query.

--
 Simon Riggs                   http://www.2ndQuadrant.com/
 PostgreSQL Development, 24x7 Support, RemoteDBA, Training & Services