Bug #47325 NDB Cluster Management Node core dump during shutdown
Submitted: 15 Sep 2009 12:43 Modified: 16 Sep 2009 13:23
Reporter: Bogdan Kecman Email Updates:
Status: Closed Impact on me:
None 
Category:MySQL Cluster: Cluster (NDB) storage engine Severity:S3 (Non-critical)
Version:7.0.7 OS:Any
Assigned to: Magnus Blåudd CPU Architecture:Any

[15 Sep 2009 12:43] Bogdan Kecman 
Description:
During shutdown, management node sometimes crash & dump core.

It looks like mgm node tries to answer a request from a client and crashes probaby  because as it is shutting down it has  deleted the ConfigManager instance.

How to repeat:
shutdown cluster while having new clients trying to connect

Suggested fix:
First shut down the socket server port so that no new clients can connect.
[15 Sep 2009 19:24] 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/83350
[16 Sep 2009 8:23] Magnus Blåudd 
Pushed to 7.0 and 7.1
[16 Sep 2009 13:23] Jon Stephens 
Documented bugfix in the NDB-7.0.8 changelog as follows:

        Clients attempting to connect to the cluster during shutdown
        could sometimes cause the management server to crash.

Closed.