From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Gary Webster <webster(at)lexmark(dot)com> |
Cc: | pgsql-admin(at)postgresql(dot)org, "Joshua D(dot) Drake" <jd(at)commandprompt(dot)com> |
Subject: | Re: db size growing out of control when using clustered Jackrabbit |
Date: | 2012-07-24 17:08:35 |
Message-ID: | 24028.1343149715@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-admin |
Gary Webster <webster(at)lexmark(dot)com> writes:
> By "routine maintenance", do you mean autovacuum, or something else?
> Autovacuum does appear to usually get 'auto-canceled' by a lock.
That's bad and you should look into the reason why it happens. Ordinary
DML (CRUD) operations should not kick autovac off a table. If it's
happening, it's probably because something is fooling with the table's
schema, which doesn't seem like something you want to have happening
during routine operations; especially not on tables that are large
enough for this to be an issue in the first place. Or it might be
something doing a LOCK TABLE as a substitute for more fine-grained
locking; which again is bad for performance reasons that have nothing
to do with hobbling autovacuum.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Joshua D. Drake | 2012-07-24 17:41:41 | Re: db size growing out of control when using clustered Jackrabbit |
Previous Message | Gary Webster | 2012-07-24 15:58:53 | Re: db size growing out of control when using clustered Jackrabbit |