Description:
The Oracle internal bug 16267120 is described in the changelog for MySQL 5.6.12 as follows:
"
InnoDB: For a compressed table, the “page reorganize” function would ignore the innodb_log_compressed_pages option and always log the entire compressed page, which increased the size of the redo log. The “page reorganize” function now adheres to the innodb_log_compressed_pages option and does not log compressed page images to the redo log when innodb_log_compressed_pages is set to “OFF”. (Bug #16267120)
"
http://dev.mysql.com/doc/relnotes/mysql/5.6/en/news-5-6-12.html
However, this is what I see in the corresponding bzr revision comments:
"
$ bzr log --show-ids -r 5023
------------------------------------------------------------
revno: 5023
revision-id: satya.bodapati@oracle.com-20130418044717-n36fflsodyukpvii
parent: yasufumi.kinoshita@oracle.com-20130418042042-llgzfb1g85vw3pda
committer: Satya Bodapati <satya.bodapati@oracle.com>
branch nick: mysql-5.6
timestamp: Thu 2013-04-18 10:17:17 +0530
message:
Bug#16267120 - PAGE REORGANIZE ASSUMES INNODB_LOG_COMPRESSED_PAGES=OFF
...
The problem is that page reorganize is always being logged as
MLOG_ZIP_PAGE_REORGANIZE (with the compression level as a parameter),
which is correct with innodb_log_compressed_pages=OFF.
If innodb_log_compressed_pages=ON, page reorganize should log the
compressed page image.
Fix:
---
If innodb_log_compressed_pages=ON, log the compressed page and do not
write a MLOG_COMP_PAGE_REORGANIZE redolog record.
"
They seem to contradict each other:
- the changelog entry says that the problem was that compressed page image would always be redo-logged, regardless of the innodb_log_compressed_pages value and that the fix was to not log it when innodb_log_compressed_pages is set to “OFF”.
- however, the revision comment says the problem was that compressed page image would never be redo-logged, regardless of the innodb_log_compressed_pages value, and the the fix was to log it when innodb_log_compressed_pages is set to "ON".
From code analysis it appears that the latter one is correct. So this is a request to fix the changelog entry.
How to repeat:
Compare description of the bug in the 5.6.12 changelog, and the corresponding bzr revision.