Bug #74460 delay in alter table
Submitted: 20 Oct 2014 15:17 Modified: 3 Nov 2015 18:31
Reporter: Claudio Eros Beccalossi Email Updates:
Status: Can't repeat Impact on me:
None 
Category:MySQL Cluster: Cluster (NDB) storage engine Severity:S3 (Non-critical)
Version:mysql-5.1.56 ndb-7.1.15 OS:Linux
Assigned to: MySQL Verification Team CPU Architecture:Any

[20 Oct 2014 15:17] Claudio Eros Beccalossi
Description:
Since few  days we have noticed in logerror file of mysql server the following trace:
141020 16:42:28 [Note] NDB online alter table commit: waiting max 77 sec for distributing ./ag22_repdbinfomodel_v010000/sessionidforsoapstandby.  epochs: (0/0,0/0,408329/9)  injector proc_info: Waiting for event from ndbcluster map: 020000
141020 16:42:29 [Note] NDB online alter table commit: waiting max 76 sec for distributing ./ag22_repdbinfomodel_v010000/sessionidforsoapstandby.  epochs: (0/0,0/0,408330/9)  injector proc_info: Waiting for event from ndbcluster map: 020000
141020 16:42:30 [Note] NDB online alter table commit: waiting max 75 sec for distributing ./ag22_repdbinfomodel_v010000/sessionidforsoapstandby.  epochs: (0/0,0/0,408331/9)  injector proc_info: Waiting for event from ndbcluster map: 020000
141020 16:42:31 [Note] NDB online alter table commit: waiting max 74 sec for distributing ./ag22_repdbinfomodel_v010000/sessionidforsoapstandby.  epochs: (0/0,0/0,408332/8)  injector proc_info: Waiting for event from ndbcluster map: 020000
141020 16:42:32 [Note] NDB online alter table commit: waiting max 73 sec for distributing ./ag22_repdbinfomodel_v010000/sessionidforsoapstandby.  epochs: (0/0,0/0,408333/8)  injector proc_info: Waiting for event from ndbcluster map: 020000
141020 16:42:33 [Note] NDB online alter table commit: waiting max 72 sec for distributing ./ag22_repdbinfomodel_v010000/sessionidforsoapstandby.  epochs: (0/0,0/0,408334/7)  injector proc_info: Waiting for event from ndbcluster map: 020000
141020 16:42:34 [Note] NDB online alter table commit: waiting max 71 sec for distributing ./ag22_repdbinfomodel_v010000/sessionidforsoapstandby.  epochs: (0/0,0/0,408335/7)  injector proc_info: Waiting for event from ndbcluster map: 020000

the problem occurs for all tables in all databases.

This behavior produces a sharp slowdown in performance, since we need to remove and
recreate databases or schemas.

any suggestions ?

How to repeat:

Perform drop all tables of database and re-create with final istructions of alter table
[3 Nov 2015 18:31] MySQL Verification Team
Hi,

I can't reproduce this problem. 
Can you share bit more data about
 - what changed when you started experiencing this problem
 - what is happening at your system when you start experiencing this problem
 - is this intermittent problem or is it present non stop

kind regards
Bogdan Kecman