Bug #62206 Complex series of restart can lead to "Error while reading the REDO log"
Submitted: 19 Aug 2011 7:51 Modified: 19 Aug 2011 16:11
Reporter: Jonas Oreland Email Updates:
Status: Closed Impact on me:
None 
Category:MySQL Cluster: Cluster (NDB) storage engine Severity:S3 (Non-critical)
Version: OS:Any
Assigned to: Jonas Oreland CPU Architecture:Any

[19 Aug 2011 7:51] Jonas Oreland
Description:
A series of restart, and crashes during restarts can in rare circumstances lead
to "Error while reading the REDO log" as code tries to read too much.

This happens during node-restart

How to repeat:
Seen rarely in internal testing

Suggested fix:
Make sure that startGci is not bigger than stopGci
[19 Aug 2011 8:02] Jonas Oreland
pushed to 6.3.46, 7.0.27 and 7.1.16
[19 Aug 2011 16:11] Jon Stephens
Documented bugfix in the NDB 6.3.46, 7.0.27, and 7.1.17 changelogs as follows:

        In rare cases, a series of node restarts and crashes during
        restarts could lead to errors while reading the REDO log.

Closed.