Re: Unexpectedly high disk space usage RESOLVED (Manual reindex/vacuum) - Mailing list pgsql-general

From Greg Williamson
Subject Re: Unexpectedly high disk space usage RESOLVED (Manual reindex/vacuum)
Date
Msg-id 1352515545.72495.YahooMailNeo@web125901.mail.ne1.yahoo.com
Whole thread Raw
In response to Re: Unexpectedly high disk space usage RESOLVED (Manual reindex/vacuum)  (Scott Marlowe <scott.marlowe@gmail.com>)
List pgsql-general
Scott -- 

<...>

>To tune autovacuum with 50 databases, start by dropping nap time to
>something much lower, like 10s.  Then if you need to, drop cost delay
>until you get to 0.  If you get to 0 and it's still not hitting your
>IO too hard, but not keeping up, then increase cost limit.  If you get
>to something in the 5000 to 10000 range, and its still not keeping up
>then start bumping the thread count
>


Thanks for outlining a strategy on this -- useful advice.

Greg Williamson



pgsql-general by date:

Previous
From: Chris Angelico
Date:
Subject: PG defaults and performance (was Re: Unexpectedly high disk space usage RESOLVED (Manual reindex/vacuum))
Next
From: "Tefft, Michael J"
Date:
Subject: "could not write to output file: Permission denied" during pg_dump