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

Hi Andres,

The ssl option is on in postgres.conf but we don't have any hostssl lines or cert configured as an authentication method in our pg_hba.conf file.

In our pg_hba.conf we have configured each replication host as trust:
host replication all xx.xxx.xxx.xx/32 trust

I replied in a previous email that we have a job that inserts 30+ million records into a table.
During the loading of this table, the wal_sender cpu spikes to 100% and appears to stay at 100% until the replication is completed on the slave.
I have been able to reproduce this. Other than offloading this job, is there anything we can do to improve the replication lag?

Thanks!

-----Original Message-----
From: Andres Freund [mailto:andres(at)2ndquadrant(dot)com]
Sent: Tuesday, May 06, 2014 3:42 PM
To: Jamie Koceniak
Cc: pgsql-bugs(at)postgresql(dot)org
Subject: Re: [BUGS] BUG #9635: Wal sender process is using 100% CPU

Hi,

On 2014-03-19 17:13:56 +0000, jkoceniak(at)mediamath(dot)com wrote:
> Periodically throughout the day, we keep seeing the wal sender process
> utilize 100% of the CPU. We began noticing this after we added 2 new
> slave servers, going from 2 to 4 slaves. See top results and I also
> included our wal settings. Thanks!

A typical reason for this is that the standbys are connecting over ssl. Is that the case? If so, intentionally?

Greetings,

Andres Freund

--
Andres Freund http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services