On 07/02/2015 08:41 AM, eudald_v wrote:
> All that was recorded during a spike. From this log I have to point
> something:
> Tables TABLE_X and TABLE_Y have both a TRIGGER that does an INSERT to
> TABLE_Z
> As you can see, TABLE_Z was being VACUUM ANALYZED. I wonder if TRIGGERS and
> VACUUM work well together, just to check another perspective.
Well, it's not triggers in particular, but vacuum does create some
contention and possible sluggishness. Questions:
* what kind of writes to the triggers do?
* can they conflict between sessions? that is, are different writes on X
and/or Y possibly overwriting the same rows on Z?
* is that autovacuum a regular autovacuum, or is it "to prevent wraparound"?
--
Josh Berkus
PostgreSQL Experts Inc.
http://pgexperts.com