Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.tools -> Re: runaway db processes after index creation?

Re: runaway db processes after index creation?

From: Frank Langelage <frank_at_lafr.de>
Date: Thu, 30 Nov 2000 22:45:14 +0100
Message-ID: <3A26CA6A.7658B5FF@lafr.de>

Eliot Shepard schrieb:
>
> I added a non-unique descending index to a table in our DB, and
> immediately started noticing problems with the server. Oracle processes
> would lock up at high CPU usage. Our server monitor then started new ones,
> and the whole thing would just go haywire.
>
> Here's an index I created. PUBLICATION_DATE is of type DATE. The table
> already has a unique primary key index.
>
> CREATE INDEX THE451.ENTITIES_PUBDATE_IDX
> ON THE451.T451_ENTITIES(PUBLICATION_DATE DESC)
>
> I noticed this in the alert log:
>
> alert_the451.log:ORA-00600: internal error code, arguments: [kcbgtcr_5],
> [1], [6], [25184429], [3427], [3426], [0], []
>
> Environment: SunOS 5.6, Oracle 8.1.6
>
> Any ideas? Thanks.
>

You shuold post the structure of the table and all indexes of this table.
How many rows in this table ?
Rule based or cost based optimizer ?

We had some trouble when our customers create additional indexes on tables. Access with rule based optimizer oracle chosed a bad access path.

Regards
Frank Received on Thu Nov 30 2000 - 15:45:14 CST

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US