On Wed, Jan 5, 2011 at 07:55, Greg Smith <greg(at)2ndquadrant(dot)com> wrote:

<snip>

> a bit of work in userland, I don't see this even being justified as an INFO
> or LOG level message.  Anytime I can script a SQL-level monitor for
> something that's easy to tie into Nagios or something, I greatly prefer that
> to log file scraping for it anyway.

+<INF-1>

Log files can be very useful for details, but they suck for noticing
the pattern in the first place :-)

<snip>

> verbosity of similar failure mode tests that follow it.  Right now failure
> to acquire a lock is just not considered a log-worthy issue, and I agree
> that it's worth considering whether it should be.

Or should it perhaps be a per-table counter in pg_stat_user_tables,
given your statement above?

-- 
 Magnus Hagander
 Me: http://www.hagander.net/
 Work: http://www.redpill-linpro.com/

In response to

• Re: We need to log aborted autovacuums at 2011-01-05 06:55:02 from Greg Smith

pgsql-hackers by date

Next:	From: Magnus Hagander	Date: 2011-01-05 13:24:45
	Subject: Re: Streaming replication as a separate permissions
Previous:	From: Pavel Stehule	Date: 2011-01-05 12:08:32
	Subject: Re: Support for negative index values in array fetching