On Tue, Nov 20, 2012 at 8:03 AM, Merlin Moncure <mmoncure@xxxxxxxxx> wrote: > On Tue, Nov 20, 2012 at 9:02 AM, Shaun Thomas <sthomas@xxxxxxxxxxxxxxxx> wrote: >> On 11/16/2012 02:31 PM, Merlin Moncure wrote: >> >>> no single thing really stands out -- contention is all over the place. >>> lwlock, pinbuffer, dynahash (especially). I am again suspicious of >>> bad scheduler interaction. any chance we can fire up pgbouncer? >> >> >> Just want to throw it out there, but we've been having really bad luck with >> the scheduler recently. But only when we use 8GB (on our 72GB system) for >> shared_buffers. Cut that down to 4GB, and everything is fine and dandy. >> >> I think the kernel devs have added in some overzealous scheduler code on us. > > Shared buffer manipulation changing contention is suggesting you're > running into free list lock issues. I wouldn't expect so. Increasing shared_buffers should either fix free list lock contention, or leave it unchanged, not make it worse. Sounds more like that NUMA/interleave/zone reclaim stuff I could never wrap my head around. Cheers, Jeff -- Sent via pgsql-general mailing list (pgsql-general@xxxxxxxxxxxxxx) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-general
