[MDEV-14172] mariabackup doesn't store innodb_fast_checksum (removed variable) in backup-my.cnf when taking backup of 5.5 Created: 2017-10-27  Updated: 2017-10-28  Resolved: 2017-10-27

Status: Closed
Project: MariaDB Server
Component/s: Backup
Affects Version/s: 10.1, 10.2
Fix Version/s: N/A

Type: Bug Priority: Minor
Reporter: Andrii Nikitin (Inactive) Assignee: Vladislav Vaintroub
Resolution: Won't Fix Votes: 0
Labels: None

Issue Links:
Relates
relates to MDEV-14936 Test MariaDB Backup 10.1 against newe... Closed

 Description   

see test bug733651.sh for ways to reproduce



 Comments   
Comment by Vladislav Vaintroub [ 2017-10-27 ]

What would be the purpose of this variable in backup-my.cnf?
it is read by mariabackup that does not understand it because it was removed.

Comment by Andrii Nikitin (Inactive) [ 2017-10-27 ]

I am really not sure what these questions try to address.
I will just consider that they assume that mariabackup will not be used to backup 5.5 ; thus leaving them without answer.

Comment by Vladislav Vaintroub [ 2017-10-27 ]

Maybe you should explain why do you think it is a bug, and what is the expected behavior of --prepare if backup-my.cnf has this parameter

Comment by Andrii Nikitin (Inactive) [ 2017-10-27 ]

I don't know expected behavior, but it quite obvious that problem will happen if backup is attempted on instance with innodb_fast_checksum enabled.
I assigned this to an engineer to prioritize and investigate possible outcome.

Comment by Vladislav Vaintroub [ 2017-10-27 ]

The real problem is not the content of backup-my.cnf, but the fact that fast checksum code was removed. This apparently did not bother anyone during the upgrade, there is no mentioning of the upgrade procedure in our documentation, and I cannot find a single bug report re MariaDB upgrade and fast checksums. So overall, this feature was probably not used much and it does not make sense to implement any workarounds, but rather document the limitation .

Comment by Andrii Nikitin (Inactive) [ 2017-10-28 ]

The aim is to have expected behavior for all input configuration. I agree that you mention valid points, but not sure why you mentioned them.

Generated at Thu Feb 08 08:11:29 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.