From: | "Jeffrey W(dot) Baker" <jwbaker(at)acm(dot)org> |
---|---|
To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
Cc: | Postgres general mailing list <pgsql-general(at)postgresql(dot)org> |
Subject: | Re: Meaning of vacuum output |
Date: | 2002-01-16 18:10:30 |
Message-ID: | 1011204630.16847.20.camel@heat |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
On Wed, 2002-01-16 at 10:08, Tom Lane wrote:
> "Jeffrey W. Baker" <jwbaker(at)acm(dot)org> writes:
> > rupert=# vacuum verbose resp_header;
> > NOTICE: --Relation resp_header--
> > NOTICE: Pages 28276: Changed 4, Empty 0; Tup 2437108: Vac 0, Keep
> > 2295773, UnUsed 96275.
> > Total CPU 1.79s/0.38u sec elapsed 2.72 sec.
> > VACUUM
> > rupert=# select count(*) from resp_header;
> > count
> > --------
> > 141357
> > (1 row)
>
> > There's nowhere near 2.437e6 tuples in the table, but the difference
> > between Tup and Keep seems to be the number of tuples in the table. But
> > what does this mean? Are there 2.295e6 deleted tuples hanging around in
> > the table?
>
> Kinda looks that way. Do you have open transactions hanging around that
> might still be able to see those tuples?
I would be very surprised but it isn't impossible. Is there a way to
query the transaction counter for an open connection?
-jwb
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2002-01-16 18:17:18 | Re: Meaning of vacuum output |
Previous Message | Tom Lane | 2002-01-16 18:08:37 | Re: Meaning of vacuum output |