Postgres Pro
Downloads
Home   >   mailing lists

Thread: is there a psql equivilent of fsck?

is there a psql equivilent of fsck?

From
Ben
Date:
We recently installed and populated a new postgres 7.3 server, which was
quickly abused with a good 12 hours of 115-degree heat. Now, we see ~1000
rows missing from a single table, and given our application, a delete of
those rows seems a very remote possibility. Is there some database analogy
to fsck I can run?

FWIW the hardware raid claims everything is just fine.

Re: is there a psql equivilent of fsck?

From
Michael Glaesemann
Date:
On Aug 13, 2007, at 12:50 , Ben wrote:

> We recently installed and populated a new postgres 7.3 server,

Why would you deploy a new server with 7.3? Current release is 8.2.
The 7.3 branch is no longer even updated.

Michael Glaesemann
grzm seespotcode net

Re: is there a psql equivilent of fsck?

From
"Joshua D. Drake"
Date:
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Ben wrote:
> We recently installed and populated a new postgres 7.3 server, which was
> quickly abused with a good 12 hours of 115-degree heat. Now, we see
> ~1000 rows missing from a single table, and given our application, a
> delete of those rows seems a very remote possibility. Is there some
> database analogy to fsck I can run?
>
> FWIW the hardware raid claims everything is just fine.

If you are missing 1000 rows, you deleted your data , you had
transaction wrap or your hardware is toast. If your hardware as toast
there would be a lot of other issues cropping up.

Why are you installing a "new" postgres 7.3 server. It isn't even
officially supported anymore.

Joshua D. Drake

>
> ---------------------------(end of broadcast)---------------------------
> TIP 9: In versions below 8.0, the planner will ignore your desire to
>       choose an index scan if your joining column's datatypes do not
>       match
>


- --

      === The PostgreSQL Company: Command Prompt, Inc. ===
Sales/Support: +1.503.667.4564   24x7/Emergency: +1.800.492.2240
PostgreSQL solutions since 1997  http://www.commandprompt.com/
            UNIQUE NOT NULL
Donate to the PostgreSQL Project: http://www.postgresql.org/about/donate
PostgreSQL Replication: http://www.commandprompt.com/products/

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFGwJ/2ATb/zqfZUUQRArcUAKCPdiqmdLpPHoOvCiT3jI7MKjDQuACfRTVc
FIqCsdmwlos9DqMpMHgnNmA=
=1GPV
-----END PGP SIGNATURE-----

Re: is there a psql equivilent of fsck?

From
Ben
Date:
On Mon, 13 Aug 2007, Michael Glaesemann wrote:

>
> On Aug 13, 2007, at 12:50 , Ben wrote:
>
>> We recently installed and populated a new postgres 7.3 server,
>
> Why would you deploy a new server with 7.3? Current release is 8.2. The 7.3
> branch is no longer even updated.

Because our product uses a decrepit RH7.3 image as a base (no, really) and
we're focusing our efforts on upgrading the whole base, instead of
spending almost as much effort on just moving to PG7.4.

But back to my question.....?

Re: is there a psql equivilent of fsck?

From
Chris Browne
Date:
bench@silentmedia.com (Ben) writes:
> We recently installed and populated a new postgres 7.3 server, which
> was quickly abused with a good 12 hours of 115-degree heat. Now, we
> see ~1000 rows missing from a single table, and given our application,
> a delete of those rows seems a very remote possibility. Is there some
> database analogy to fsck I can run?
>
> FWIW the hardware raid claims everything is just fine.

What we tend to use when we run into such situations is:
  "VACUUM VERBOSE ANALYZE;"

This walks through all tables and indices in the database, and seeks
to clean them up.  If the disk has been mussed up, this will tend to
terminate with suitably scary looking error messages.

FYI, is there some particular reason why you went with PostgreSQL 7.3?
That's almost five years old, which is like (hmm... (* 5 6)) thirty
years old in "Internet Years."  That's, like, way, way, way obsolete.

We haven't been quick about jumping onto fresh new releases - we only
got to 8.1 this year, when 8.2 is now *last* year's flavour.
Nonetheless, we got rid of our last 7.3 instance several years ago...
--
output = ("cbbrowne" "@" "linuxdatabases.info")
http://cbbrowne.com/info/x.html
Rules of the Evil Overlord  #105. "I will design all doomsday machines
myself. If I must hire a mad  scientist to assist me, I will make sure
that he is sufficiently twisted to never regret his evil ways and seek
to undo the damage he's caused." <http://www.eviloverlord.com/>