Freeze break request: move staging improvements to production for askbot

[Toshio Kuratomi]

On Fri, May 24, 2013 at 12:27:16PM +1000, Ankur Sinha wrote:
> On Thu, 2013-05-23 at 11:59 -0700, Toshio Kuratomi wrote:
> > Given that askbot isn't critical to getting a release out the only
> > part that
> > worries me is the database update.
> > 
> > if it's  tested in stg, sounds okay to me, though.
> 
> Yep. I was worried that the database update might explode, but it seemed
> to work just right on the staging server so I wouldn't expect it to be
> any different over on the production one. I won't know until I try it
> out though. 
> 
> I could backup the db, just to be on the safe side. Is there a fixed
> method infra follows for db backups (django db in this case)?

Probably just jump into #fedora-admin and see if there's someone about who
can dump the db.  Since it's on the postgres server, it needs a
sysadmin-main that can log onto db01 and run something like this:

  sudo -u postgres pg_dump -C $DBNAME |xz -2 > $DBNAME-$(date --rfc-3339=date).dump.xz

Note that the db will be consistent with a dump like this but we can't
guarantee that the app will be consistent (if the app doesn't use
transactions at appropriate places it could perform half of an action).  For
that you'd also have to shut down ask before the db is dumped.

-Toshio
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://lists.fedoraproject.org/pipermail/infrastructure/attachments/20130523/fe6efaa6/attachment.sig>