[MDEV-12555] Error log full of with [Note] InnoDB: Online DD Created: 2017-04-22  Updated: 2017-04-25  Resolved: 2017-04-25

Status: Closed
Project: MariaDB Server
Component/s: Server
Affects Version/s: 10.1.22
Fix Version/s: 10.1.23

Type: Bug Priority: Minor
Reporter: Nicola Assignee: Unassigned
Resolution: Duplicate Votes: 0
Labels: None
Environment:

Centos 6 64 bit


 Description   

Hello, I've a curious problem also explained here:
https://serverfault.com/questions/777444/mariadb-10-1-x-online-ddl-spaming-error-log-files

My log files is very big because it's full of lines like this:

2017-04-22  4:09:09 140466725649152 [Note] InnoDB: Online DDL : Start
 
2017-04-22  4:09:09 140466725649152 [Note] InnoDB: Online DDL : Start reading clustered index of the table and create temporary files
 
2017-04-22  4:09:09 140466725649152 [Note] InnoDB: Online DDL : End of reading clustered index of the table and create temporary files
 
2017-04-22  4:09:09 140466725649152 [Note] InnoDB: Online DDL : Start merge-sorting index `PRIMARY` (1 / 4), estimated cost : 8.0000
 
2017-04-22  4:09:09 140466725649152 [Note] InnoDB: Online DDL : End of  merge-sorting index `PRIMARY` (1 / 4)
 
2017-04-22  4:09:09 140466725649152 [Note] InnoDB: Online DDL : Start building index `PRIMARY` (1 / 4), estimated cost : 12.0000
 
2017-04-22  4:09:09 140466725649152 [Note] InnoDB: Online DDL : End of building index `PRIMARY` (1 / 4)
 
2017-04-22  4:09:09 140466725649152 [Note] InnoDB: Online DDL : Completed
 
2017-04-22  4:09:09 140466725649152 [Note] InnoDB: Online DDL : Start merge-sorting index `id_country` (2 / 4), estimated cost : 8.0000
 
2017-04-22  4:09:09 140466725649152 [Note] InnoDB: Online DDL : End of  merge-sorting index `id_country` (2 / 4)
 
2017-04-22  4:09:09 140466725649152 [Note] InnoDB: Online DDL : Start building index `id_country` (2 / 4), estimated cost : 12.0000
 
2017-04-22  4:09:09 140466725649152 [Note] InnoDB: Online DDL : End of building index `id_country` (2 / 4)
 
2017-04-22  4:09:09 140466725649152 [Note] InnoDB: Online DDL : Completed
 
2017-04-22  4:09:09 140466725649152 [Note] InnoDB: Online DDL : Start merge-sorting index `name` (3 / 4), estimated cost : 8.0000
 
2017-04-22  4:09:09 140466725649152 [Note] InnoDB: Online DDL : End of  merge-sorting index `name` (3 / 4)
 
2017-04-22  4:09:09 140466725649152 [Note] InnoDB: Online DDL : Start building index `name` (3 / 4), estimated cost : 12.0000
 
2017-04-22  4:09:09 140466725649152 [Note] InnoDB: Online DDL : End of building index `name` (3 / 4)
 
2017-04-22  4:09:09 140466725649152 [Note] InnoDB: Online DDL : Completed
 
2017-04-22  4:09:09 140466725649152 [Note] InnoDB: Online DDL : Start merge-sorting index `id_zone` (4 / 4), estimated cost : 8.0000
 
2017-04-22  4:09:09 140466725649152 [Note] InnoDB: Online DDL : End of  merge-sorting index `id_zone` (4 / 4)
 
2017-04-22  4:09:09 140466725649152 [Note] InnoDB: Online DDL : Start building index `id_zone` (4 / 4), estimated cost : 12.0000
 
2017-04-22  4:09:09 140466725649152 [Note] InnoDB: Online DDL : End of building index `id_zone` (4 / 4)
 
2017-04-22  4:09:09 140466725649152 [Note] InnoDB: Online DDL : Completed

This happen every night at 4 AM. Is this a bug or I need to disable this? I can't find any doc about that.
Thank you

 Comments   
Comment by Daniel Black [ 2017-04-23 ]

You are on Windows or Centos?

You want to disable the online ddl (probably caused by EVENTS or cron), or the log messages? Or something to make looking at the log more manageable?

Comment by Nicola [ 2017-04-23 ]

Hello Daniel, I'm on CentOS 6.9.
If I understand correctly, the Online DDL is caused by this cron:
0 4 * * * mysqlcheck -A -o > /dev/null 2>&1
which I run every night. The problem is that error log is growing in size because of this log (about 6 MB every night) and log_warnings = 0 doesn't stop that.

Thank you

Comment by Daniel Black [ 2017-04-25 ]

I'd suggest you need to run mysqlcheck with -A -vvv --auto-repair to obtain output as to what repair is being done. Its a bit unclear as to why some difference occurred or why you need to run it nightly. If the repair of tables doesn't quieten the log.

If this failes consider the --extended option of mysqlcheck.

There shouldn't be this much output regularly so attempting to resolve the reason for the output should be first. However If the size of the log files are bothering you you can implement a logrotate script to truncate/compress/delete the log files.

Comment by Daniel Black [ 2017-04-25 ]

Actually - looks like a duplicate of MDEV-12322 fixed in 10.1.23

Comment by Nicola [ 2017-04-25 ]

Thank you Daniel

Generated at Thu Feb 08 07:58:37 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.