Hi Gabor! Gabor Gombas [2005-07-19 11:44 +0200]: > I've just upgraded postgresql-8.0 and postgresql-common and found that > postgresql-common is doing a "VACUUM FULL" on upgrade. This was a > _rather_ unpleasant surprise since "VACUUM FULL" takes more than 2 hours > on this machine.
Uh, that must be a big one. How long does "VACUUM" run?
> I think that causing such excessive database downtime for a package
> upgrade is unacceptable. And it is not even mentioned in
> changelog.Debian...
It is mentioned:
postgresql-common (20) unstable; urgency=low
[...]
* Add upgrade-scripts/all_vacuum_t1.sql: Upgrade script that will just
vacuum all available databases. This is pretty useful and harmless, and is
a good thing to test upgrade scripts at a larger scale.
[...]
At the time when the script runs all clusters are already started
again, so it's not really a downtime. This was mainly added by me to
test upgrading scripts, and it will be removed again soon. I don't
want the first security update for a stable release to be the first
time the upgrading scripts mechanism is tested.
> Btw. pg_ctlcluster also prints a warning message:
> Use of uninitialized value in -r at /usr/bin/pg_ctlcluster line 112, <F> line
> 56.
Thanks, I'll look into this.
Sorry for the inconvenience and thanks for using unstable for testing
out the new packages!
Martin
--
Martin Pitt http://www.piware.de
Ubuntu Developer http://www.ubuntu.com
Debian Developer http://www.debian.org
signature.asc
Description: Digital signature
