Bug #74858 Replication hangs but is reported as running after aborted index creation.
Submitted: 14 Nov 2014 9:49 Modified: 30 Apr 2015 21:31
Reporter: Lars Torstensson Email Updates:
Status: Can't repeat Impact on me:
None 
Category:MySQL Server: Replication Severity:S3 (Non-critical)
Version:5.6.19-ndb-7.3.6-cluster-gpl-log OS:Linux (2.6)
Assigned to: CPU Architecture:Any
Tags: INDEX, replication

[14 Nov 2014 9:49] Lars Torstensson
Description:
After aborting a create index on a slave
the "show slave status" reported the replication as running but the Exec_Master_Log_Pos did not change and the cpu load was low. So the mysqld appear to hang.

How to repeat:
1 Have master replicate to slave with high load.

2 Create index on the slave.

3 When the cpu load for (ndbmtd) drop drastically and the "create index" statement doesn't finish after about a minute use Ctrl-C.

4 issue "show slave status" and it will report replication threads as running but the Exec_Master_Log_Pos will not change.

Slave_IO_Running: Yes
Slave_SQL_Running: Yes

5 To get the Exec_Master_Log_Pos to move we did "stop slave" and "start slave"
[30 Apr 2015 21:31] MySQL Verification Team
Hello Lars,

Thank you for the report.
I couldn't reproduce this issue at my end with latest builds(master/slave with same version). Could you please provide conf files from both master/slave, results of SHOW PROCESSLIST from slave, and is this NDB->Innodb replication?

If you can provide more information, feel free to add it to this bug and change the status back to 'Open'.

Thank you for your interest in MySQL.

Thanks,
Umesh