OSEC

Neohapsis is currently accepting applications for employment. For more information, please visit our website www.neohapsis.com or email hr@neohapsis.com
 
RE: occasionally bd index messages

From: Becker, Holger (holger.beckersap.com)
Date: Tue Aug 30 2005 - 11:00:47 CDT


Dirk Naujoks wrote:

> We have a problem with an index which is marked as bad.
> This happens not all the time our application is acessing the
> Table but sometimes.
>
> Several Times we droped the index and recreated it.
>
> What reasons are possible for this behavior?
> Is it possible that the application ma cause this bad index?
>
> The following information are stored in the Database:
> OPERATINGSYSTEM: Linux 2.6.11.4-21.8-smp #1 SMP Tue Jul
> 19 12:42:37 UTC 2005
> PROCESSORTYPE: x86_64
> ADDRESSINGMODE: 64
> ID: Kernel 7.5.0 Build 030-123-100-791
> MAJORVERSION: 7
> MINORVERSION: 5
> CORRECTIONLEVEL: 0
> BUILD: 30
> KERNELVARIANT: fast
>
> What additional information do you need to tell more about
> this?

Hi,

there are some reasons why indexes will be flagged as bad

- an index was created and than the db stops without regular shutdown;
  at restart time this index wouldn't be recreated but only flagged as
bad
  to speed up the restart; this behaviour could be changed by setting
the
  db parameter AUTO_RECREATE_BAD_INDEXES to YES

- at the migration from version 7.3 to version >= 7.4 indexes
  containing only one column which could become null will be flagged
  as bad because of designe changes within indexes between 7.3 and 7.4

- hardware problems

Kind regards
Holger
      

--
MaxDB Discussion Mailing List
For list archives: http://lists.mysql.com/maxdb