Bug #22698 Master node fails to start backup

• Submitted: 26 Sep 2006 11:18
• Modified: 4 Jan 2007 14:16
• Reporter: Lars Torstensson
• Status: Can't repeat
• Category: MySQL Cluster: Cluster (NDB) storage engine
• Severity: S2 (Serious)
• Version: mysqlcluster 5.0.24
• OS: Linux (Linux 2.6.9-34.ELsmp #1)
• CPU Architecture: Any

[26 Sep 2006 11:18] Lars Torstensson

Description:
We have a 6 node 3 replicas cluster and when I try to start a backup from the Mgmt-client I get this message:

ndb_mgm> start backup
Waiting for completed, this may take several minutes
Node 1: Backup request from 7 failed to start. Error: 1304
Start of backup failed
*  3001: Could not start backup
*        Sequence failure: Permanent error: Internal error
ndb_mgm> quit

How to repeat:
The cluster was restarted after a router crach but not the mgmt server.

[26 Sep 2006 11:19] Lars Torstensson

c

[26 Sep 2006 17:08] Jonas Oreland

can you try "ndb_select_all -d sys SYSTAB_0" and upload result here ?

also please upload cluster log and config.ini ?

Is this repeatable ?

/Jonas

[27 Sep 2006 12:28] Lars Torstensson

Cluster log:

Sep 27 14:21:56 nl2-db4 NDB[27126]: [MgmSrvr] Node 1: Backup request from 7 failed to start. Error: 1304

[27 Sep 2006 14:09] Jonas Oreland

thx for ndb_select all....
but i need full cluster log
  (too see which kind of restarts/failure has occured)

[27 Sep 2006 15:19] Lars Torstensson

Have uploaded the cluster log (bug-data-22698.tar.gz)

If you need additional info about what has happened you can try asking Stewart Smith, he was doing debug stuff there last week. He might have left something in debug mode.

We have not managed to reproduce it.

[4 Jan 2007 13:45] Geert Vanderkelen

We would need a repeatable test case for bug #22698. It can't be repeated currently.

[4 Jan 2007 14:16] Lars Torstensson

Unfortunately I have not been able to repeat this sins we now are using a two replicas cluster. Or is it fortunately!

[26 Jan 2009 19:46] Hartmut Holzgraefe

See also bug #42354