Re: row-attribute in EXPLAIN-output doesn't match count(*)

On Thursday 14 April 2005 22:44, Andrew Sullivan wrote:
> On Thu, Apr 14, 2005 at 09:57:03PM +0200, Andreas Joseph Krogh wrote:
> > Is it safe to issue a "kill -2 <pid-of-idle-process>", will the
> > connection-pool reconnect?
>
> The kill -2 will definitely work to shut the connectiond down
> cleanly. Actually, I should put that more precisely: you should send
> SIGINT. I don't know of any systems where that's not 2, but I
> suppose it's logically possible.
>
> As for the reconnection, it'll depend entirely on what your pool does
> when its connections are closed. You'll have to test it.

Just did, and SIGINT didn't help(or I was too impatient), so I sent a SIGTERM,
then it wen away. But - it didn't help on my "vacuum dead rows" problem,
vacuum still won't reclaim those > 100.000 "dead" rows, not even after a
web-app restart(shut down Tomcat, ran "vacuum full verbose" then started
Tomcat again).

INFO: vacuuming "public.onp_web_index"
INFO: "onp_web_index": found 0 removable, 160057 nonremovable row versions in
14892 pages
DETAIL: 159759 dead row versions cannot be removed yet.
...

Any hints?

--
Andreas Joseph Krogh <andreak(at)officenet(dot)no>
Senior Software Developer / Manager
gpg public_key: http://dev.officenet.no/~andreak/public_key.asc
------------------------+---------------------------------------------+
OfficeNet AS | The most difficult thing in the world is to |
Hoffsveien 17 | know how to do a thing and to watch |
PO. Box 425 Skøyen | somebody else doing it wrong, without |
0213 Oslo | comment. |
NORWAY | |
Phone : +47 22 13 01 00 | |
Direct: +47 22 13 10 03 | |
Mobile: +47 909 56 963 | |
------------------------+---------------------------------------------+