Re: core system is getting unresponsive because over 300 cpu load

Andres Freund wrote
> Others mentioned already that that's worth improving.

Yes, we are just setting up pgbouncer

Andres Freund wrote
> Some versions of this kernel have had serious problems with transparent
> hugepages. I'd try turning that off. I think it defaults to off even in
> that version, but also make sure zone_reclaim_mode is disabled.

Yes, I'm aware of that so always set it to never.
but thank you for the zone_reclaim_mode.

Andres Freund wrote
> 9.6 has quite some scalability improvements over 9.5. I don't know
> whether it's feasible for you to update, but if so, It's worth trying.
>
> How about taking perf profile to investigate?

Both are on my to do list :)

Andres Freund wrote
> I'd suggest monitoring /proc/meminfo for the amount of Dirty and
> Writeback memory, and see whether rapid changes therein coincide with
> periodds of slowdown.

yes, I was monitoring it the whole day and that's the reason why I've
changed dirty_background_ratio but both of them were flat - without any
bigger spikes.

--
Sent from: http://www.postgresql-archive.org/PostgreSQL-general-f1843780.html