Craig James <craig_james@emolecules.com> writes:
> On 6/17/11 11:51 AM, Shianmiin wrote:
>> We have a PostgreSQL 9.0.4 on CentOs for performance testing and we are
>> seeing the similar issue.
>> we have a "crazy" setup it has 1 database with 1000 identical schemas. There
>> are occasional I/O write storm
>> of over 100 MB/sec without any disk reads, and it could last for a couple of
>> minutes when the schemas/data are aggressively populated by pg_restore. All
>> the io writes seem to be on pgstat.tmp.
> Based on the advice I got from my original question, I changed autovacuum_naptime to "5min", and the problem
completelydisappeared. (I know that's a long interval, but this particular server gets maybe 5-10 heavy updates per
weekand is idle the rest of the time.)
> select count(1) from pg_database ;
> count
> -------
> 267
> It seems like there's a problem somewhere. Autovacuum has improved enormously in the last couple of years, but some
changeto its algorithm causes a lot of I/O thrashing when there are more than a few hundred separate databases.
Well, if you have a lot of databases then you definitely need to
increase the naptime to keep autovac's demands for stats within bounds.
I don't find that surprising, though I do wonder if we ought to redefine
the way that GUC works: right now, you get one autovac wakeup every
naptime/databases seconds.
The OP seems to have some other issue, though, since he says he's only
got 1 database.
regards, tom lane