Bug #47984 Job buffer congestion during startup with huge DataMemory

• Submitted: 12 Oct 2009 5:48
• Modified: 12 Oct 2009 11:48
• Reporter: Jonas Oreland
• Status: Closed
• Category: MySQL Cluster: Cluster (NDB) storage engine
• Severity: S3 (Non-critical)
• Version: mysql-5.1-telco-6.3
• OS: Any
• Assigned to: Jonas Oreland
• CPU Architecture: Any

[12 Oct 2009 5:48] Jonas Oreland

Description:
When starting a node with DataMemory at ~90G
it could after long allocation period, crash with job-buffer-congestion.

This was cause by the fact that ndb_mgmd send registration signals during this
and these was left unanswered, and after a while they got to many and job-buffer
got full.

How to repeat:
.

Suggested fix:
.

[12 Oct 2009 6:22] 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/86515

3101 Jonas Oreland	2009-10-12
      ndb - bug#47984 - move mem allocation to thread

[12 Oct 2009 7:30] Jonas Oreland

pushed to 6.3.28 and 7.0.9

[12 Oct 2009 11:48] Jon Stephens

Documented bugfix in the NDB-6.3.28 and 7.0.9 changelogs as follows:

        Starting a data node with a very large amount of DataMemory
        (approximately 90G or more) could lead to crash of the node due
        to job buffer congestion.

Closed.