Bug #37039 backup causes crash for certain table definitions
Submitted: 28 May 2008 12:35 Modified: 30 May 2008 14:48
Reporter: Jonas Oreland Email Updates:
Status: Closed Impact on me:
None 
Category:MySQL Cluster: Cluster (NDB) storage engine Severity:S3 (Non-critical)
Version:telco-6.3..+ OS:Any
Assigned to: Jonas Oreland CPU Architecture:Any

[28 May 2008 12:35] Jonas Oreland
Description:
Some tables with a specific amount of null values and data
can cause crash when running backup

How to repeat:
.

Suggested fix:
fix computation of maxRecordSize
(patch already written)
[29 May 2008 11:29] 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/47191

ChangeSet@1.2614, 2008-05-29 13:30:21+02:00, jonas@perch.ndb.mysql.com +4 -0
  ndb - bug#37039
    certain table/data could cause backup to crash node
    include nullbits in length of mask (READ_PACKED header)
[29 May 2008 11:32] Bugs System
Pushed into 5.1.24-ndb-6.3.15
[29 May 2008 12:05] Geert Vanderkelen
Just to complete the 'How To Repeat', from test case of Jonas:

mysql> CREATE TABLE t11_c (a int, b00 int null, b01 int null,b02 int null,b03 int null,b04 int null,b05 int null,b06 int null,b07 int null,b08 int null,b90 int null,b10 int null,b11 int null,b12 int null,b13 int null,b14 int null,b15 int null,b16 int null,b17 int null,b18 int null,b19 int null,b20 int null,b21 int null,b22 int null,b23 int null,b24 int null,b25 int null,b26 int null,b27 int null,b28 int null,b29 int null,b30 int null,b31 int null,b32 int null) engine = ndb;
mysql> insert into t11_c values (1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1);

ndb_mgm> START BACKUP;
[29 May 2008 16:18] Bugs System
Pushed into 5.1.23-ndb-6.4.0
[30 May 2008 14:48] Jon Stephens
Documented bugfix in the 5.1.24-ndb-6.3.15 changelog as follows:

      Attempting to create a native backup of NDB tables having a large number
      of NULL columns and data could lead to node failures.