Re: BUG #9635: Wal sender process is using 100% CPU

From: Heikki Linnakangas <hlinnakangas(at)vmware(dot)com>
To: Jamie Koceniak <jkoceniak(at)mediamath(dot)com>
Cc: "pgsql-bugs(at)postgresql(dot)org" <pgsql-bugs(at)postgresql(dot)org>
Subject: Re: BUG #9635: Wal sender process is using 100% CPU
Date: 2014-05-06 11:00:45
Message-ID: 5368C0DD.3030205@vmware.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs

On 05/05/2014 10:37 PM, Jamie Koceniak wrote:
> Hi Heikki,
>
> I performed an strace on one of the pids that was consuming 100% of the cpu.
> This is just a portion of the log.
> I also trimmed out some data provided in the quotes in the read and sends, not sure if you need it.
> A majority of the calls in the trace log were recvfrom() and sendto().
> EAGAIN (Resource temporarily unavailable) came up time and time again.
>
> # strace log during 100% cpu utilization on the master

Hmm, that trace looks normal to me. It doesn't look like it's
busy-looping, as I would expect if it's using 100% of the CPU.

Can you post the full log, or at least a larger chunk? I don't need the
data in the quotes from read() and sends() that you removed, but I would
like to see the trace from a longer period of time to see what the
pattern is.

- Heikki

In response to

Responses

Browse pgsql-bugs by date

  From Date Subject
Next Message sathishp87 2014-05-06 11:24:39 BUG #10235: Postgres is opening
Previous Message Alvaro Herrera 2014-05-06 03:46:01 Re: Re: BUG #10189: Limit in 9.3.4 no longer works when ordering using a composite multi-type index