[MDEV-10701] SET global TOKUDB_CHECKPOINT_LOCK=ON will not stop the tokudb data file change Created: 2016-08-30  Updated: 2016-09-28  Resolved: 2016-08-31

Status: Closed
Project: MariaDB Server
Component/s: Storage Engine - TokuDB
Affects Version/s: 10.0.26
Fix Version/s: N/A

Type: Bug Priority: Major
Reporter: dennis Assignee: Unassigned
Resolution: Not a Bug Votes: 0
Labels: TOKUDB_CHECKPOINT_LOCK, tokudb, upstream
Environment:

centos6.5, mariadb 10.0.26



 Description   

How to reproduce the bug:

Open a new shell "shell1" , and use sysbench to do the modification on the tokudb table, this sysbench
will running through the whole test:

sysbench --mysql-host=127.0.0.1 --mysql-port=3314 --mysql-user=test --mysql-password=test \
--mysql-db=test1 --oltp-tables-count=1 --oltp-table-size=6000000 \
--mysql-table-engine=tokudb --test=tests/db/oltp.lua prepare

open a new shell "shell2" , to check the data file and log file modification of tokudb table in the disk:

cd $datadir
[root@mysql1 data3314]# ls -lh | grep test1
drwx------. 2 mysql mysql 4.0K Aug 30 04:05 test1
-rw-rw----. 1 mysql mysql 9.0M Aug 30 04:07 _test1_sbtest1_key_k_1_8c02_1_1c_B_1.tokudb
-rw-rw----. 1 mysql mysql  85M Aug 30 04:07 _test1_sbtest1_main_8c02_1_1c_B_0.tokudb
-rw-rw----. 1 mysql mysql  64K Aug 30 04:07 _test1_sbtest1_status_8bf7_1_1c.tokudb

we can see the log file and data file is changing.

Then open a new shell "shell3" login to the mariadb and acquire the checkpoint lock:

root@localhost:mysql3314.sock  04:07:33 [(none)]>SET TOKUDB_CHECKPOINT_LOCK=ON;
Query OK, 0 rows affected (0.32 sec)
root@localhost:mysql3314.sock  04:07:36 [(none)]>SET global TOKUDB_CHECKPOINT_LOCK=ON;
Query OK, 0 rows affected (0.04 sec)

And during the test, this connection will not quit or broken.

Back to the shell2, we still can see the data file changing!

It seams that, the "TOKUDB_CHECKPOINT_LOCK=ON" fail to stop the data file modification on tokudb.



 Comments   
Comment by Elena Stepanova [ 2016-08-31 ]

The upsteam bug was invalidated with the following comment:

TokuDB checkpoint lock does not stop data files from changing, it only prevents the checkpoint from occurring. When a dirty node is evicted from memory due to cache pressure, it gets written to disk, The block file map header does not get updated on disk as it does during a checkpoint.

There is one condition where a block file header will get updated even with the checkpointer disabled and that is when an index/dictionary file is closed and evicted from the cache table. This will be addressed in coming versions of Percona Server when the LOCK CHECKPOINT FOR BACKUP is introduced and tokudb_checkpoint_lock removed.

Closing this replica accordingly.
When the change described in the second paragraph is implemented in TokuDB, it will be merged into MariaDB.

Comment by bohutang [ 2016-09-28 ]

We can stop the block header updated when setting TOKUDB_CHECKPOINT_LOCK=ON with this patch:
https://github.com/BohuTANG/ft-index/commit/e48ac39c6644114c674f600dbdc4b34788e47091

With it, now we can backup both InnoDB and TokuDB on PS5.6/PS5.7 and MariaDB.
Patch is:
https://github.com/BohuTANG/percona-xtrabackup/commit/b43ba1b25deff2ba21eb9030b49413e1851f263b

Generated at Thu Feb 08 07:44:16 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.