Bug #19931 Undo*Buffer can only be increased, not decreased on rolling restart
Submitted: 19 May 2006 8:21 Modified: 9 Aug 2006 14:45
Reporter: Hartmut Holzgraefe Email Updates:
Status: Closed Impact on me:
None 
Category:Connector / ODBC Documentation Severity:S3 (Non-critical)
Version:4.1->5.1 OS:Linux (linux)
Assigned to: Jon Stephens CPU Architecture:Any

[19 May 2006 8:21] Hartmut Holzgraefe
Description:
Changing UndoIndexBuffer and UndoDataBuffer to lower values
triggers a 2341 error

Time: Friday 19 May 2006 - 10:17:00
Status: Temporary error, restart node
Message: Internal program error (failed ndbrequire) (Internal error, programming error or missing error message, please report a bug)
Error: 2341
Error data: DbtupUndoLog.cpp
Error object: DBTUP (Line: 275) 0x00000008
Program: libexec/ndbd
Pid: 28813
Trace: /usr/local/mysql-5.0/cluster/ndb_3_trace.log.3
Version: Version 5.0.22
***EOM***

How to repeat:
Set up a cluster with default settings, start it,
change the Undo Buffer settings to their minimum
values of UndoIndexBuffer=4M, UndoDataBuffer=1M,
restart the management node, stop a data node and
try to restart it

Suggested fix:
If this is really a setting that can't be changed at runtime then
a more meaningfull error message and documentation about 
this fact should be added.
[19 May 2006 10:06] Hartmut Holzgraefe
node trace log

Attachment: ndb_3_trace.log.3.gz (application/x-gzip, text), 66.56 KiB.

[9 Aug 2006 13:55] Jon Stephens
Changing category to Docs, reassigning to myself per discussion today with Hartmut and Martin.
[9 Aug 2006 14:45] Jon Stephens
Thank you for your bug report. This issue has been addressed in the documentation. The updated documentation will appear on our website shortly, and will be included in the next release of the relevant products.

Updated documentation as discussed with NDB devs and Support.

NOTE: These parameters are no longer present in 5.1 Cluster.