PostgreSQL Discussion - Corruption on production system

Author

Corruption on production system

Eric B. Ridge

2005-04-25, 8:24 pm

We've got a PG 7.3.6 installation that just started receiving this
error:
2005-04-25 19:28:54 ERROR: Invalid page header in block 1110 of
the_table_name

We're also unable to dump that table using pg_dump. Postgres isn't
actually crashing, so I'm not sure how to get a backtrace out of it.

SELECT version():
PostgreSQL 7.3.6 on i686-pc-linux-gnu, compiled by GCC gcc
(GCC) 3.3.2 20031218 (Gentoo Linux 3.3.2-r5, propolice-3.3-7)

(Don't get upset that it says Gentoo -- I built it myself from
official Postgres sources)

Is page header corruption a known issue with 7.3.6 and should I
upgrade to 7.3.9 at the same time that I'm restoring this table from
backup?

Any quick advice will be greatly appreciated!

thanks

eric

---------------------------(end of broadcast)---------------------------
TIP 3: if posting/reading through Usenet, please send an appropriate
subscribe-nomail command to majordomo@postgresql
.org so that your
message can get through to the mailing list cleanly

Alvaro Herrera

2005-04-26, 3:23 am

On Mon, Apr 25, 2005 at 08:24:03PM -0400, Eric B. Ridge wrote:

> Is page header corruption a known issue with 7.3.6 and should I
> upgrade to 7.3.9 at the same time that I'm restoring this table from
> backup?

Well, you can get a page header corruption on any release as long as you
have faulty hardware ... RAM randomly dropping bits is not unheard of.
Have you run memtest?

If you don't mind losing that page of data, you could turn on
zero_damaged_pages on postgresql.conf. If you do, you could try
pg_filedump (to be found somewhere on sources.redhat.com/rhdb IIRC) to
examine the damaged page more closely.

--
Alvaro Herrera (<alvherre[@]dcc.uchile.cl> )
"Vivir y dejar de vivir son soluciones imaginarias.
La existencia está en otra parte" (Andre Breton)

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

Eric B. Ridge

2005-04-26, 3:23 am

On Apr 25, 2005, at 9:42 PM, Alvaro Herrera wrote:

> Well, you can get a page header corruption on any release as long
> as you
> have faulty hardware ... RAM randomly dropping bits is not unheard of.
> Have you run memtest?

No doubt. :) What I was asking was if 7.3.9 (or .8 or .7) had a
known issue that caused page corruption and if upgrading would fix
that bug. Regardless, I went ahead and upgraded.

eric

---------------------------(end of broadcast)---------------------------
TIP 1: subscribe and unsubscribe commands go to majordomo@postgresql
.org

Sponsored Links