pg.ddx.io pgsql-sql@postgresql.org mailing list archive
help / color / mirror / Atom feedANALYZE error
3+ messages / 3 participants
[nested] [flat]
* ANALYZE error
@ 2004-03-09 04:22 David Witham <davidw@unidial.com.au>
0 siblings, 2 replies; 3+ messages in thread
From: David Witham @ 2004-03-09 04:22 UTC (permalink / raw)
To: pgsql-sql
Hi all,
I run a cron job at 2am every night to do an ANALYZE on my whole database:
su - postgres -c 'echo analyze | /usr/bin/psql buns | grep -v ANALYZE'
Its a batch oriented database so there is no user activity on it at this time. The only other job that could be running is a vacuum I run at 1am but I don't think it would take an hour to run. Occasionally I get this error message from the analyze job:
ERROR: simple_heap_update: tuple concurrently updated
What does this mean and should I do anything about it?
Thanks,
David Witham
Telephony Platforms Architect
Unidial
Australia
^ permalink raw reply [nested|flat] 3+ messages in thread
* Re: ANALYZE error
@ 2004-03-09 04:55 Tom Lane <tgl@sss.pgh.pa.us>
parent: David Witham <davidw@unidial.com.au>
1 sibling, 0 replies; 3+ messages in thread
From: Tom Lane @ 2004-03-09 04:55 UTC (permalink / raw)
To: David Witham <davidw@unidial.com.au>; +Cc: pgsql-sql
"David Witham" <davidw@unidial.com.au> writes:
> Occasionally I get this error message from the analyze job:
> ERROR: simple_heap_update: tuple concurrently updated
> What does this mean and should I do anything about it?
You can get this if two ANALYZEs run in parallel for the same table;
whichever one tries to update pg_statistic second fails with the above
error. It's moderately annoying, but not dangerous.
regards, tom lane
^ permalink raw reply [nested|flat] 3+ messages in thread
* Re: ANALYZE error
@ 2004-03-09 05:15 Iain <iain@mst.co.jp>
parent: David Witham <davidw@unidial.com.au>
1 sibling, 0 replies; 3+ messages in thread
From: Iain @ 2004-03-09 05:15 UTC (permalink / raw)
To: David Witham <davidw@unidial.com.au>; pgsql-sql
>>The only other job that could be running is a vacuum I run at 1am but I
don't think it would take an hour to run.
Famous last words... :-) It should be easy to check this, but I don't know
if that is a likely source of problems anyway.
If I were you I'd just do VACUUM ANALYSE in one comand. It would certainly
eliminate that as a factor.
At this stage you have no idea which table it occurs on, or if it always
happens on the same table, or at the same time, am I right? Failing any more
useful response from those with more internal knowledge you may have to
resort to collecting information and trying to isolate the problem.
It would be a good idea to give all the relevant version numbers too
regards
iain
^ permalink raw reply [nested|flat] 3+ messages in thread
end of thread, other threads:[~2004-03-09 05:15 UTC | newest]
Thread overview: 3+ messages (download: mbox mbox.gz follow: Atom feed)
-- links below jump to the message on this page --
2004-03-09 04:22 ANALYZE error David Witham <davidw@unidial.com.au>
2004-03-09 04:55 ` Tom Lane <tgl@sss.pgh.pa.us>
2004-03-09 05:15 ` Iain <iain@mst.co.jp>
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox