Bug #48474 Out of redo-log during node-restart
Submitted: 2 Nov 15:00 Modified: 2 Nov 18:57
Reporter: Jonas Oreland
Status: Closed
Category:Server: Cluster Severity:S3 (Non-critical)
Version:mysql-5.1-telco-6.3 OS:Any
Assigned to: Jonas Oreland Target Version:

[2 Nov 15:00] Jonas Oreland 
Description:
During a node restart, logging is enabled on a per fragment basis when
the copying of a fragment has completed.

But LCP is not enabled until all fragments has been copied.
This means that it's possible to run out of REDO (410)

How to repeat:
run test test prg

Suggested fix:
enabled logging only after all fragments has been copied.
just prior to enabling LCP
[2 Nov 18:08] Bugs System 
A patch for this bug has been committed. After review, it may
be pushed to the relevant source trees for release in the next
version. You can access the patch from:

  http://lists.mysql.com/commits/88970

3150 Jonas Oreland	2009-11-02
      ndb - bug#48474 - don't start logging until all fragments has been copied
[2 Nov 18:09] Bugs System 
A patch for this bug has been committed. After review, it may
be pushed to the relevant source trees for release in the next
version. You can access the patch from:

  http://lists.mysql.com/commits/88971

3150 Jonas Oreland	2009-11-02
      ndb - bug#48474 - don't start logging until all fragments has been copied
[2 Nov 18:24] Jonas Oreland 
pushed to 6.3.29 and 7.0.10
[2 Nov 18:57] Jon Stephens 
Documented in the NDB-6.3.29 and 7.0.10 changelogs as follows:

        During a node restart, logging was enabled on a per-fragment
        basis as the copying of each fragment was completed but local
        checkpoints were not enabled until all fragments were copied,
        making it possible to run out of redo log file space (NDB Error
        410) before the restart was complete. Now logging is enabled
        only after all fragments has been copied, just prior to enabling
        local checkpoints.

Closed.