Postgres Pro
Downloads
Home   >   mailing lists

Re: TODO item: adding VERBOSE option to CLUSTER [with patch] - Mailing list pgsql-hackers

From Jim Cox
Subject Re: TODO item: adding VERBOSE option to CLUSTER [with patch]
Date
Msg-id c2ee6dbd0810101013u644bd6bl7c48442b44d6ec7d@mail.gmail.com
Whole thread Raw
In response to Re: TODO item: adding VERBOSE option to CLUSTER [with patch]  (Heikki Linnakangas <heikki.linnakangas@enterprisedb.com>)
Responses Re: TODO item: adding VERBOSE option to CLUSTER [with patch]
List pgsql-hackers
Tree view
On Fri, Oct 10, 2008 at 10:23 AM, Heikki Linnakangas <heikki.linnakangas@enterprisedb.com> wrote:
Kevin Grittner wrote:
"Jim Cox" <shakahshakah@gmail.com> wrote:
if present an INFO message is generated which displays
the schema.tblname just before actual clustering is kicked off (see
example
below).
 
postgres=# CLUSTER VERBOSE ;
INFO:  clustering "public.my_b"
INFO:  clustering "public.my_c"
INFO:  clustering "public.my_a"
CLUSTER
 Would it make sense to display the pg_total_relation_size before and
after?

Assuming you run CLUSTER as a replacement for VACUUM FULL, yes. More interesting would be a metric of "clusteredness", I think.

Something more like the following?

postgres=# CLUSTER VERBOSE ;
INFO:  clustering "public.my_b"
INFO:  complete, 0 rows scanned, 0 rows now live
DETAIL:  CPU 0.00s/0.00u sec elapsed 0.00 sec.
INFO:  clustering "public.my_c"
INFO:  complete, 20 rows scanned, 10 rows now live
DETAIL:  CPU 0.00s/0.00u sec elapsed 0.00 sec.
INFO:  clustering "public.my_a"
INFO:  complete, 10 rows scanned, 10 rows now live
DETAIL:  CPU 0.00s/0.00u sec elapsed 0.00 sec.
CLUSTER

pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: LWLockAcquire with priority
Next
From: Michael Renner
Date:
Subject: Re: How is random_page_cost=4 ok?