From: | Lucas Possamai <drum(dot)lucas(at)gmail(dot)com> |
---|---|
To: | Kevin Grittner <kgrittn(at)gmail(dot)com> |
Cc: | Sergey Konoplev <gray(dot)ru(at)gmail(dot)com>, Postgres General <pgsql-general(at)postgresql(dot)org> |
Subject: | Re: Increased I/O / Writes |
Date: | 2016-05-23 21:44:06 |
Message-ID: | CAE_gQfW3dBiELcOppYN6v=8++pEeywD7iXGw-OT3doB8SXO4_A@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
On 24 May 2016 at 09:40, Kevin Grittner <kgrittn(at)gmail(dot)com> wrote:
> On Mon, May 23, 2016 at 4:32 PM, Lucas Possamai <drum(dot)lucas(at)gmail(dot)com>
> wrote:
>
> > The spikes were happening because, after the reboot, the HUGE PAGES were
> > enabled.
> >
> > After disabling them, all got back to normal.
>
> Since you said earlier you were on PostgreSQL version 9.2, you must
> be talking about transparent huge pages. Yeah, be sure those are
> configured to be disabled in a way that "sticks" on your OS. When
> you get to version 9.4 you will notice that we support huge pages
> directly. That would be expected to work without problems even
> though TRANSPARENT huge pages are debilitating.
>
Yep.. you're right, Kevin.
Transparent huge pages...
We do have a script that disables it when the system is rebooted, but that
did not work.
Anyway, the problem was solved..
Cheers
Lucas
From | Date | Subject | |
---|---|---|---|
Next Message | Lucas Possamai | 2016-05-23 23:23:15 | Re: Unused indexes - PostgreSQL 9.2 |
Previous Message | Kevin Grittner | 2016-05-23 21:40:51 | Re: Increased I/O / Writes |