Postgres Pro
Downloads
Home   >   mailing lists

Re: brin index vacuum versus transaction snapshots - Mailing list pgsql-hackers

From Simon Riggs
Subject Re: brin index vacuum versus transaction snapshots
Date
Msg-id CANP8+jK8uJSHe4j7tjuQRQsthUawUx+-JMnoFnzXzCovRiA9fQ@mail.gmail.com
Whole thread Raw
In response to Re: brin index vacuum versus transaction snapshots  (Kevin Grittner <kgrittn@ymail.com>)
List pgsql-hackers
Tree view
On 31 July 2015 at 15:48, Kevin Grittner <kgrittn@ymail.com> wrote:

Reread the thread. 

I stand corrected.
 
The error is being hit on a VACUUM command
(which is not, of course in any explicit transaction).

Seems like VACUUM only ever needs a read-committed snapshot. Whether it takes a read-committed or repeatable read makes no difference.

Does brin_summarize_new_values() need to track what it reads for serializable transactions?

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

pgsql-hackers by date:

Previous
From: Kevin Grittner
Date:
Subject: Re: brin index vacuum versus transaction snapshots
Next
From: Jesper Pedersen
Date:
Subject: RequestAddinLWLocks(int n)