Icon View Thread

The following is the text of the current message along with any replies.
Messages 1 to 5 of 5 total
Thread Index Page Buffers Corrupt revisited
Tue, May 9 2006 5:06 AMPermanent Link

"Al VAs"
Hi,

We have now converted almost all of our clients over to DBISAM and things
have settled down nicely.  One thing is that we are still getting
spasmodically reports of 'Index Page Buffers Corrupt' issues -
client/server.  These are spasmodic, one client received 2 of them last
week, another client one, and Im sure a couple are not being reported.

Any suggestions on what could be causing this or the best way to decipher
what might be causing it or what potentially can cause this sort of issue?
We would dearly like to track down the source of the issues.

V3.30 with the updated file

Thanks

Alex

Tue, May 9 2006 1:26 PMPermanent Link

Rich Walker
"Al VAs" <deleteprefix_alex@favour.com.au> wrote:

>Any suggestions on what could be causing this or the best way to decipher
>what might be causing it or what potentially can cause this sort of issue?
>We would dearly like to track down the source of the issues.
>
>V3.30 with the updated file

Although we're not running in client/server mode, we have occasional
Index Page Buffers Corrupt errors with 3.30.  We recently recompiled
DBISAM with the updated file and haven't received any new reports of
8965 errors.  However, it is probably too soon to say whether the
updated file has fixed the problem.

One thing I've noticed is that Index Page Buffers Corrupt errors tend
to occur on tables with a lot of secondary indexes.  For instance, the
problem has occurred several times on one of our tables that has 21
secondary indexes.

Another thing I've noticed is that Index Page Buffers Corrupt errors
tend to occur on tables with full text indexes.  As an experiment, we
recently removed the full text index from one of our tables, and that
seems to have reduced but not entirely eliminated the 8965 errors.

One client that has had more than their share of 8965 errors (all on
one particular table with a lot of secondary indexes and a full text
index) is running in a Terminal Server environment.

Do any of these things apply in your case?  I, too, would like track
down the source of these errors.  Luckily for us, the problems have
been fairly rare, but it would be nice to get rid of the 8965 errors
once and for all.

-Rich
Tue, May 9 2006 1:59 PMPermanent Link

Roy Lambert

NLH Associates

Team Elevate Team Elevate

Rich

Does your or Al's app have any threads? That's where I was getting 8965's

Roy Lambert
Tue, May 9 2006 4:17 PMPermanent Link

Tim Young [Elevate Software]

Elevate Software, Inc.

Avatar

Email timyoung@elevatesoft.com

Alex,

<< We have now converted almost all of our clients over to DBISAM and things
have settled down nicely.  One thing is that we are still getting
spasmodically reports of 'Index Page Buffers Corrupt' issues -
client/server.  These are spasmodic, one client received 2 of them last
week, another client one, and Im sure a couple are not being reported.

Any suggestions on what could be causing this or the best way to decipher
what might be causing it or what potentially can cause this sort of issue?
We would dearly like to track down the source of the issues. >>

Is the table that is reporting this problem actually corrupted when you
check it ?

--
Tim Young
Elevate Software
www.elevatesoft.com

Thu, May 11 2006 1:26 AMPermanent Link

"Clive"
I get this on occasion too. generally a bad shutdown of the app,
I just trap the error and repair all the tables automatically in the app
then let the app continue.. Seems to work well when it does occur.

"Al VAs" <deleteprefix_alex@favour.com.au> wrote in message
news:958305FD-53FA-472A-92B2-14D327F5D659@news.elevatesoft.com...
> Hi,
>
> We have now converted almost all of our clients over to DBISAM and things
> have settled down nicely.  One thing is that we are still getting
> spasmodically reports of 'Index Page Buffers Corrupt' issues -
> client/server.  These are spasmodic, one client received 2 of them last
> week, another client one, and Im sure a couple are not being reported.
>
> Any suggestions on what could be causing this or the best way to decipher
> what might be causing it or what potentially can cause this sort of issue?
> We would dearly like to track down the source of the issues.
>
> V3.30 with the updated file
>
> Thanks
>
> Alex
>

Image