Re: Log rotation - Mailing list pgsql-hackers

From Robert Treat
Subject Re: Log rotation
Date
Msg-id 200403141246.17674.xzilla@users.sourceforge.net
Whole thread Raw
In response to Re: Log rotation  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
On Sunday 14 March 2004 12:19, Tom Lane wrote:
> Fernando Nasser <fnasser@redhat.com> writes:
> > Lamar Owen wrote:
> >> Third, it seems that you don't have enough profiling data to support
> >> a 'syslog is bad' position.
> >
> > That is true.  It is from hearsay, from people who run production
> > environments.  It may be a belief based on old experiences though.
>
> I think it's pretty well established that syslog sucks for high log
> volume if you run it in the mode where it fsyncs its log after every
> message.  But I don't believe we have any data that says it's a problem
> even if you avoid that pitfall.
>

Yep. The other basic thing I recomend is putting syslog/log output on a 
different disk than where your data goes.  Maybe not proactical for everyone, 
but not a big deal for "enterprise" level users (which was the op's target 
audience)

Robert Treat
-- 
Build A Brighter Lamp :: Linux Apache {middleware} PostgreSQL


pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: listening addresses
Next
From: Bruce Momjian
Date:
Subject: Re: Log rotation