[Spacewalk-list] [Spacewalk-devel] I think I found the root cause of the PostgreSQL Idle in transaction connection build up.

[James Hogarth]

> After I've upgraded to spacewalk 1.8, I've not seen any of these.  I'd
> typically see 40 - 50 hanging out at any given time.
>
> I'm not running the updated jdbc connectoid thingamajig, just whatever
> comes with doing an install from the spacewalk 1.8 repos.  On Friday, I had
> one process idle in transaction.  After just checking now, I have none.
>
>
I normally have around 10 idle and didn't seem to vary between old and new
jar...

Is anyone else running OSAD as well? Since 1.8 I've had a significant
situation of needing a daily restart but I"m unsure if this is a purely
postgres+spacewalk 1.8 issue (didn't have a problem with 1.7 in this
manner) or whether OSAD causes the additional trouble - I'm running
slightly different configurations each day at the moment to check the
behaviour for sure...

At around 11pm (so config difference check on taskomatic scheduled tasks)
the number of locks and connections increases at a pretty linear rate until
it caps out and the postgres server refuses any new connections...

I've attached a PDF from my monitoring instance to this bug:
https://bugzilla.redhat.com/show_bug.cgi?id=873379

My current environment is only 40 systems in spacewalk - not a significant
number by any real degree - so I'm rather surprised by the behaviour...
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/spacewalk-list/attachments/20121114/79c773ff/attachment.htm>