From: | Tom Darci <tom(at)nuws(dot)com> |
---|---|
To: | pgsql-general(at)postgresql(dot)org |
Cc: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
Subject: | Re: getting our bearings on "out of memory. failed on request of size..." |
Date: | 2008-10-28 16:05:43 |
Message-ID: | 8F56C9DE-49BE-4F95-9810-9F69D58B0326@nuws.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Thanks Tom,
The "n" has always been small, like 20, or 4 or 32 (these numbers
are from memory, not written down), but the one we ran into today was
large: 67108864.
I will head to the EXPLAIN, as well as to getting the memory dump
from the log.
Regards,
-Tom
------------------------------------------------------------
Tom Darci
tom(at)nuws(dot)com
On Oct 28, 2008, at 7:44 AM, Tom Lane wrote:
> Tom Darci <tom(at)nuws(dot)com> writes:
>> Our PG server is serving up 50 databases or so (with identical
>> schemas), and the largest one of those has started giving us "out of
>> memory. failed on request of size n" errors in many places. This is a
>> 9 GB database with 100+ tables, the largest of which have 4 million
>> or
>> so rows. We see the Out of Memory errors when trying to run a "hard"
>> SQL statement (SELECTing from a complex view, updating all rows in a
>> large table, etc...)
>
> Well, it would be important to know if "n" is little or big, and to
> see
> the EXPLAIN output of the problem query. Also, an out-of-memory error
> will cause a memory usage map to be dumped to postmaster stderr ---
> that would be pretty useful to see too, if your logging setup captures
> it.
>
> regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Chris Browne | 2008-10-28 16:06:55 | PostgreSQL talk at CASCON |
Previous Message | Sam Mason | 2008-10-28 15:48:21 | Re: How to know the password for the user 'postgres' |