| From: | Jeff Janes <jeff(dot)janes(at)gmail(dot)com> |
|---|---|
| To: | Lonni J Friedman <netllama(at)gmail(dot)com> |
| Cc: | Kevin Grittner <kgrittn(at)ymail(dot)com>, Eduardo Morras <emorrasg(at)yahoo(dot)es>, pgsql-general <pgsql-general(at)postgresql(dot)org> |
| Subject: | Re: upgrade from 9.2.x to 9.3 causes significant performance degradation |
| Date: | 2013-09-18 18:00:59 |
| Message-ID: | CAMkU=1wFyYajUaiW=XD5OtFwZVuvTOXRQyVvpUg1N+ZQ8AsJCQ@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
On Wed, Sep 18, 2013 at 10:30 AM, Lonni J Friedman <netllama(at)gmail(dot)com>wrote:
>
> >
> > This looks similar to cases I've seen of THP defrag going wild.
> > Did the OS version or configuration change? Did the PostgreSQL
> > memory settings (like shared_buffers) change?
>
> I think you're onto something here with respect to THP defrag going
> wild. I set /sys/kernel/mm/transparent_hugepage/defrag to 'never' and
> immediately the load dropped on both slaves from over 5.00 to under
> 1.00.
>
> So this raises the question, is this a kernel bug, or is there some
> other solution to the problem?
> Also, seems weird that the problem didn't happen until I switched from
> 9.2 to 9.3. Is it possible this is somehow related to the change from
> using SysV shared memory to using Posix shared memory and mmap for
> memory management?
>
I would guess that it is probably a kernel bug which gets efficiently
exercised by the SysV to Posix change.
Cheers,
Jeff
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Vladislav Lysov | 2013-09-18 19:05:31 | PostgreSQL with big data |
| Previous Message | Lonni J Friedman | 2013-09-18 17:30:16 | Re: upgrade from 9.2.x to 9.3 causes significant performance degradation |