| From: | "Ross J(dot) Reedstrom" <reedstrm(at)rice(dot)edu> | 
|---|---|
| To: | Micah Anderson <micah(at)colltech(dot)com> | 
| Cc: | pgsql-general(at)postgresql(dot)org | 
| Subject: | Re: Memory bloating | 
| Date: | 2000-10-03 17:26:53 | 
| Message-ID: | 20001003122652.A30898@rice.edu | 
| Views: | Whole Thread | Raw Message | Download mbox | Resend email | 
| Thread: | |
| Lists: | pgsql-general | 
On Tue, Oct 03, 2000 at 11:55:45AM -0500, Micah Anderson wrote:
> Hello,
> 
> We are using postgresql exclusively for our database backend for a
> high-traffic site, sustaining 3-4 pages per second, in many cases
> bursting well over that. At least half of those accesses are pgsql
> SELECT, we rarely, if at all, use DELETE. It seems worst on tables with
> more than about 1000 rows or 1000 hits an hour, or both.
> 
Hmm, how about UPDATE? The exploding files and VACUUM stats indicate that
you're creating no longer valid tuples _somehow_. Any recent changes in
SQL scripts? I'd be suspicious of a 'view' page that got cloned from an
'edit' page, and is updating the 'modified' field, as an example.
You could turn on a higher level of logging, and capture the actual
queries, for a time, and find the culprit that way. Use of varchar isn't
the problem: there's got to be either UPDATE or DELETE going on.
Ross
-- 
Open source code is like a natural resource, it's the result of providing
food and sunshine to programmers, and then staying out of their way.
[...] [It] is not going away because it has utility for both the developers 
and users independent of economic motivations.  Jim Flynn, Sunnyvale, Calif.
| From | Date | Subject | |
|---|---|---|---|
| Next Message | chris markiewicz | 2000-10-03 17:27:39 | RE: trouble with trigger/function??? | 
| Previous Message | Adam Lang | 2000-10-03 17:18:13 | Updates |