Re: Error msgs from PostgreSQL server : specified item offset is too large, failed to add item to the page

Hi Tom,

If you remembered I had posted the reindexing query on the forum last
week.
http://archives.postgresql.org/pgsql-novice/2004-05/msg00102.php

I have external schedule reindexing which DROPs and reCREATEs all the
user-defined indexes on the tables periodically.
Probably I was too early to think the problem have been fixed by
reindexing but seems the problem is still there.
So I have following few queries and I hope I shall get the answers from
all you expert guys here.

1) Is external reindexing required with Postgres 7.4? If yes then why?
2) Whats the overhead of reindexing (DROP and CREATE INDEX) on the live
database?
3) IS there any thing I can add along with reindexing (VACCUM ANALYSE
also already being performed) or
4) Can I avoid reindexing all together and try some thing else?
5) What additional logs should be monitored/enabled to get more help to
resolve this problem?

There is also a PR probably still in open state? Is anybody working on
this issue of PostgreSQL?
BUG #1104: postmaster crashes

Please post your valuable replies.

Thanks,
Rajan Bhide

-----Original Message-----
From: Tom Lane [mailto:tgl@sss.pgh.pa.us]
Sent: Friday, May 28, 2004 10:32 PM
To: Rajan Bhide
Cc: Bruce Momjian; pgsql-novice@postgresql.org; Bruno Wolff III; Oliver
Fromme
Subject: Re: [NOVICE] Error msgs from PostgreSQL server : specified item
offset is too large, failed to add item to the page


"Rajan Bhide" <rbhide@nulinkinc.com> writes:
> WARNING:  specified item offset is too large
> PANIC:  failed to add item to the page for "transid_req_idx"

I'm suspicious that that index is corrupted.  Does reindexing it make
the problem go away?

            regards, tom lane

---------------------------(end of broadcast)---------------------------
TIP 4: Don't 'kill -9' the postmaster