MySQL Bugs: #34445: Incorrectly dimensioned pool, could cause cluster-failure when doing +3 to

Bug #34445	Incorrectly dimensioned pool, could cause cluster-failure when doing +3 to
Submitted:	10 Feb 2008 12:03	Modified:	10 Mar 2008 17:09
Reporter:	Jonas Oreland	Email Updates:
Status:	Closed	Impact on me:	None
Category:	MySQL Cluster: Cluster (NDB) storage engine	Severity:	S3 (Non-critical)
Version:	>6.3.8	OS:	Any
Assigned to:	Jonas Oreland	CPU Architecture:	Any

Description:
Incorrectly dimensioned pool, could cause cluster-failure when doing more than
3 parallel take-overs (node restart or take-over during system restart)

The higher the parallelism, the greater the risk

How to repeat:
run testNodeRestart -n pnr with more than 6 nodes

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/41986

ChangeSet@1.2874, 2008-02-10 13:06:21+01:00, jonas@perch.ndb.mysql.com +3 -0
  ndb - bug#34445
    Incorrectly dimensioned pool, could cause cluster-failure when doing
    more than 3 parallel take-overs (node restart or take-over during system restart)
  
    Move usage of getParam into separate function (called by BLOCK_CONSTRUCTOR) so that 
      it's called on correct block

pushed to telco-6.x (>=3)

Documented bugfix in 5.1.23-ndb-6.3.9 changelog as follows:

        Cluster failures could sometimes occur when performing
        more than three parallel takeovers during node restarts or
        system restarts.

Left in PQ pending further merges.

this bug is only in telco 6.3 and upwards

Updated changelog entry per note from Tomas.