Re: Why will vacuum not end? - Mailing list pgsql-performance

From Shea,Dan [CIS]
Subject Re: Why will vacuum not end?
Date
Msg-id F2D63B916C88C14D9B59F93C2A5DD33F0B924E@cisxa.cmc.int.ec.gc.ca
Whole thread Raw
In response to Why will vacuum not end?  ("Shea,Dan [CIS]" <Dan.Shea@ec.gc.ca>)
Responses Re: Why will vacuum not end?
List pgsql-performance
PWFPM_DEV=# select * from pg_locks;
 relation | database | transaction |  pid  |           mode           |
granted
----------+----------+-------------+-------+--------------------------+-----
----
    17472 |    17347 |             |  2618 | ShareUpdateExclusiveLock | t
          |          |    10858533 | 28778 | ExclusiveLock            | t
    17472 |    17347 |             |  2618 | ShareUpdateExclusiveLock | t
          |          |    10803814 |  2618 | ExclusiveLock            | t
    16759 |    17347 |             | 28778 | AccessShareLock          | t
(5 rows)

PWFPM_DEV=#

17347 is the database PWFPM_DEV iod, The pids are below

[root@murphy root]# ps -ef |grep 28778|grep -v "grep"
postgres 28778   504  0 18:06 ?        00:00:00 postgres: scores PWFPM_DEV
[local] idle
[root@murphy root]# ps -ef |grep 2618|grep -v "grep"
postgres  2618   504  8 Apr22 ?        02:31:00 postgres: postgres PWFPM_DEV
[local] VACUUM
[root@murphy root]#
A vacuum is running now.  I restarted the database, set vacuum_mem =
'196608'; and started a new vacuum.  I also stopped inserting into the
database.
I hoping I will get some results.

PWFPM_DEV=# select now();vacuum verbose analyze forecastelement;select
now();
              now
-------------------------------
 2004-04-22 13:38:02.083592+00
(1 row)

INFO:  vacuuming "public.forecastelement"
INFO:  index "forecastelement_rwv_idx" now contains 391385895 row versions
in 5051132 pages
DETAIL:  27962015 index row versions were removed.
771899 index pages have been deleted, 496872 are currently reusable.
CPU 4499.54s/385.76u sec elapsed 55780.91 sec.
INFO:  "forecastelement": removed 33554117 row versions in 737471 pages
DETAIL:  CPU 135.61s/83.99u sec elapsed 1101.26 sec.
-----Original Message-----
From: Christopher Kings-Lynne [mailto:chriskl@familyhealth.com.au]
Sent: Tuesday, April 20, 2004 9:26 PM
To: Shea,Dan [CIS]
Cc: pgsql-performance@postgresql.org
Subject: Re: [PERFORM] Why will vacuum not end?


> No, but data is constantly being inserted by userid scores.  It is
postgres
> runnimg the vacuum.
> Dan.

Well, inserts create some locks - perhaps that's the problem...

Otherwise, check the pg_locks view to see if you can figure it out.

Chris


---------------------------(end of broadcast)---------------------------
TIP 2: you can get off all lists at once with the unregister command
    (send "unregister YourEmailAddressHere" to majordomo@postgresql.org)

pgsql-performance by date:

Previous
From: Josh Berkus
Date:
Subject: Re: Help with performance problems
Next
From: Josh Berkus
Date:
Subject: Re: Why will vacuum not end?