Bug #28235 GCP stop after cluster restart

• Submitted: 4 May 2007 11:57
• Modified: 22 Jul 2007 9:18
• Reporter: Kris Buytaert (Candidate Quality Contributor)
• Status: No Feedback
• Category: MySQL Cluster: Cluster (NDB) storage engine
• Severity: S2 (Serious)
• Version: 5.1.16
• OS: Linux
• CPU Architecture: Any

[4 May 2007 11:57] Kris Buytaert

Description:
Ndbd crashes with fairly large dataset in memory with following error.

There is no reason why the arbitrator should decide to shut down the node,  no communication loss between the nodes 

Time: Thursday 3 May 2007 - 16:02:45
Status: Temporary error, restart node
Message: Node lost connection to other nodes and can not form a unpartitioned cl
uster, please investigate if there are error(s) on other node(s) (Arbitration er
ror)
Error: 2305
Error data: Arbitrator decided to shutdown this node
Error object: QMGR (Line: 4897) 0x0000000e
Program: ndbd
Pid: 15531
Trace: /var/lib/mysql/mysql-cluster//ndb_3_trace.log.4
Version: Version 5.1.16 (beta)
***EOM***

How to repeat:
Happens periodiacally

[4 May 2007 11:58] Kris Buytaert

ndb eror report

Attachment: ndb_error_report_20070504100751.tar.bz2 (application/x-bzip2, text), 150.95 KiB.

[7 May 2007 13:25] Jonas Oreland

Possibly a duplicate of http://bugs.mysql.com/bug.php?id=27942

Could you test if patch applied to that bug resolves this problem?

/Jonas

[22 Jun 2007 9:18] Sveta Smirnova

Status changed, because feedback is needed.

[22 Jul 2007 23:00] Bugs System

No feedback was provided for this bug for over a month, so it is
being suspended automatically. If you are able to provide the
information that was originally requested, please do so and change
the status of the bug back to "Open".