On 9/4/15 6:59 AM, Tomas Vondra wrote:
>> Finally, whatever causes the database to wait also causes the
>> Windows instance to slow down. But restarting Postgresql fixes
>> this as well.
>
> That's a bit strange, I guess. If you're not observing light CPU and I/O
> usage, then the instance should not be slow, unless there's something
> else going on - possibly at the virtualization level (e.g. another busy
> instance on the same hardware, some sort of accounting that limits the
> resources after a time, etc.)
I've experienced something similar on linux before. Database is slow,
but neither CPU or IO is maxed. IIRC there was nothing disturbing in
pg_locks either. I don't recall the server actually slowing down, but
this would have been on something with at least 24 cores, so...
My suspicion has always been that there's some form of locking that
isn't being caught by the tools.
--
Jim Nasby, Data Architect, Blue Treble Consulting, Austin TX
Experts in Analytics, Data Architecture and PostgreSQL
Data in Trouble? Get it in Treble! http://BlueTreble.com