On January 12, 2005 01:30 pm, Christopher Browne wrote:
> Josh Berkus wrote:
> >Jan, Chris,
> >
> >Due to James Black's issue, we figured out that Slony won't purge anything
> >from s_log_1 which is younger than the oldest XID on the server.
> >
> >What's the techincal reason for this?   The result is that orphaned
> >transactions cause a pretty fast buildup of s_log_1 rows.  Not that
> > orphaned transactions aren't a problem on their own, but we're wondering
> > if this isn't fixable in Slony *before* PostgreSQL gets an
> > open_idle_transaction_timeout setting.
>
> As you have probably seen, Darcy has a possible patch for this.

My previous patch was no better, in that it mannaged to leave even more dead 
entries laying around.  Attached is a revised patch that is just a more 
aggressive cleanup or the original (unpatched) version.

>
> My question is whether we can get a good diagnosis of the cause of the
> situation.  It seems plausible that this might be something justifying a
> 1.0.6 release, particularly if it is a condition people could fairly
> readily fall into by accident.
>
> What was the "proximate cause"?  If we can document that, that's a good
> part of the battle all by itself.
> _______________________________________________
> Slony1-general mailing list
> Slony1-general at gborg.postgresql.org
> http://gborg.postgresql.org/mailman/listinfo/slony1-general

-- 
Darcy Buskermolen
Wavefire Technologies Corp.
ph: 250.717.0200
fx:  250.763.1759
http://www.wavefire.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: cleanup_thread.c.patch
Type: text/x-diff
Size: 548 bytes
Desc: not available
Url : http://gborg.postgresql.org/pipermail/slony1-general/attachments/20050112/adc4b4d4/cleanup_thread.c.bin