On 2013-11-20 06:21:13 -0800, Kevin Grittner wrote:
> Andres Freund <and...@2ndquadrant.com> wrote:
> > On 2013-11-20 05:59:58 -0800, Kevin Grittner wrote:
> >> I don't understand where that would make sense; especially since
> >> I thought that a database FREEZE followed by a checkpoint
> >> releases old clog space anyway.
> >
> > It only releases them up to the (cluster wide) xmin horizon. So
> > if there are older snapshots or prepared xacts around...
> So as long as there are no open transactions or prepared
> transactions on the master which started before the release with
> the fix is applied, VACUUM FREEZE would be guaranteed to work? 
> Since I don't see how a non-prepared transaction would be running
> from before a minor release upgrade, that just means we have to
> make sure there are no prepared transactions from before the
> upgrade?

That's not a bad point. So the way to fix it would be:

1) Restart the standby to the new minor release, wait for catchup
2) Restart the primary (fast or smart) to the new minor release
3) Acquire enough new xids to make sure we cross a clog page (?)
4) Jot down a new xid:  SELECT txid_current()::bigint % (1::bigint<<33-1)
5) vacuumdb -z -a
6) Ensure that there are no prepared xacts older than 3) around
FROM pg_prepared_xacts
ORDER BY age(transaction) DESC LIMIT 1;
7) Ensure the xmin horizon is above the one from: 3:
SELECT datname, datfrozenxid
FROM pg_database
WHERE datname != 'template0'
ORDER BY age(datfrozenxid) DESC LIMIT 1;
8) Get the current lsn: SELECT pg_current_xlog_location();
9) verify on each standby that SELECT pg_last_xlog_receive_location() is
   past 7)
10) be happy

I am not sure how we can easily compute that 6) and 7) are past 3) in
the presence of xid wraparounds.


Andres Freund

 Andres Freund                     http://www.2ndQuadrant.com/
 PostgreSQL Development, 24x7 Support, Training & Services

Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:

Reply via email to