Bug #45146 MaxNoOfOrderedIndexes not getting updated in reload of 7.0.5 configuration

• Submitted: 27 May 2009 22:49
• Modified: 23 Jun 2009 7:10
• Reporter: Jonathon Coombes
• Status: Can't repeat
• Category: MySQL Cluster: Cluster (NDB) storage engine
• Severity: S2 (Serious)
• Version: mysql-5.1-telco-7.0
• OS: Any
• CPU Architecture: Any
• Tags: 7.0.5, cluster, configuration, RELOAD

[27 May 2009 22:49] Jonathon Coombes

Description:
The --reload feature of Cluster 7.0.5 does not seem to update the MaxNoOfOrderedIndexes value.

Values on 6.3 cluster show:
ndb_config --query=id,type,MaxNoOfOrderedIndexes,IndexMemory,DataMemory -r '\n'| grep ndbd
2,ndbd,5000,104857600,524288000
3,ndbd,5000,104857600,524288000
4,ndbd,5000,104857600,524288000
5,ndbd,5000,104857600,524288000

Values on 7.0.5 after --reload show:
ndb_config --query=id,type,MaxNoOfOrderedIndexes,IndexMemory,DataMemory -r '\n'| grep ndbd
2,ndbd,128,104857600,524288000
3,ndbd,128,104857600,524288000
4,ndbd,128,104857600,524288000
5,ndbd,128,104857600,524288000

How to repeat:
Update the value of MaxNoOfOrderedIndexes and do a  --reload to update internal settings.

Suggested fix:
Code review and change

[1 Jun 2009 13:13] Jonathan Miller

Did you do a rolling restart of the data nodes?

[2 Jun 2009 5:28] MySQL Verification Team

The cluster was started using a full restart as it was only a test cluster.

[10 Jun 2009 22:32] Sveta Smirnova

Thank you for the feedback.

If I do:

ndb_mgm> shutdown
....

Modify configuration file

ndb_mgmd --reload I get correct results - value of MaxNoOfOrderedIndexes reloaded. Please provide your actions step-by-step.

[22 Jun 2009 22:03] MySQL Verification Team

Apparently fixed in 7.0.6 as it cannot be reproduced in this new version

[23 Jun 2009 7:10] Sveta Smirnova

Thank you for the feedback.

Closed as "can't repeat" because last comment.