[Info-ingres] RES: [Info-ingres] Re: Blobs and a server crash

#1

Hi Martin,

To workaround the problem you can try use 'set trace point QE70'.
I have an open issue with CA where a similar problem happens in R3 and
2.6 too. CA is working to fix it in R3.

"The problem seems to be related to parallel index creation and setting
this trace point serializes it. This can also cause the modify to take a
little bit longer, but should not impact the database maintenance."

HTH.

Leandro Fava
DBA-SINF-UNISC http://www.unisc.br
Presidente IUG-BR http://www.iug-br.org
Fone: +55 51 3717 7636

Quote:

-----Mensagem original-----
De: info-ingres-admin (AT) cariboulake (DOT) com
[mailto:info-ingres-admin (AT) cariboulake (DOT) com] Em nome de James Keane
Enviada em: Tuesday, September 27, 2005 8:57 AM
Para: info-ingres (AT) cariboulake (DOT) com
Assunto: [Info-ingres] Re: Blobs and a server crash

Martin,

I got a similar problem. The CA call reference number is 13032314-1.
The problem was fixed by patch 10200 for Ingres 2.6 on Solaris. There
was no problem with the table.

A modify of a table, por_templatelookup, which has a long varch column
and 2
persistent indices, results in the following error;
E_DMF009_DM0M_FREE_OVERLAP_PREV

Segmentation violations also occurred.

regards,
James Keane
Irish Revenue Commissioners

martin.bowes (AT) ctsu (DOT) ox.ac.uk wrote:
Hi dudes,

I'm running IngresII2.6/0305 patch 9892 on a linux box.

A weekend process to modify tables in a database caused some
segmentation violations shown in the errlog. A subsequent attempt to
repeat the modify crashed the server.

A check with verifydb on the database has shown that a large number
of the tables holding BLOBS are in trouble with:
W_DM505E_NO_ETAB_TABLE WARNING: The extension table with
index 163850, refferenced by coupon in attribute monitor_comment of
record with tuple ID 1024 does not exist.

W_DM5060_NO_ETAB_SEGMENT WARNING: Segment number 1
in extension table with index 319 was not found when looking for
extension segments for peripheral data-type in field monitor_comment
for record with Tuple ID 1536.
...
W_DM505F_TOO_MANY_BAD_SEGS WARNING: Verifydb was
unable to find a peripheral segment that matches the current coupon
after more than 319 tries, checking on this coupon will be
terminated,
and verifydb will skip to the next coupon.

The database in question is a test database and can be nuked if
necessary. Its so unimportant its not journalled and the
prior checkpoint
was Sep 14.

But before I nuke it, I'd like to get some idea what has
happened and
how to fix it. I've backed up the database in its current
crap state and
have restored the prior checkpoint

Has anyone run across anything like this before?

Martin Bowes
--
Random Duckman Quote.
Geek: This 'Duckman' you speak of, does he have any super powers?
Ajax: Whens he's scared he can scream like a little girl.

_______________________________________________
Info-ingres mailing list
Info-ingres (AT) cariboulake (DOT) com
http://mailman.cariboulake.com/mailm...py/info-ingres

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

[Info-ingres] RES: [Info-ingres] Re: Blobs and a server crash

comp.databases.ingres comp.databases.ingres

[Info-ingres] RES: [Info-ingres] Re: Blobs and a server crash - 09-27-2005 , 07:34 AM