Differences in file size of .ibds across cluster

[Walter Heck]

We have a galera cluster with pxc 5.6.22:

root@elvispresley:~# dpkg -l | grep perc
|/ Err?=(none)/Reinst-required (Status,Err: uppercase=bad)
ii percona-toolkit 2.2.13 all Advanced MySQL and system command-line tools
ii percona-xtrabackup 2.2.10-1.wheezy amd64 Open source backup tool for InnoDB and XtraDB
ii percona-xtradb-cluster-client-5.6 5.6.22-25.8-978.wheezy amd64 Percona XtraDB Cluster database client binaries
ii percona-xtradb-cluster-common-5.6 5.6.22-25.8-978.wheezy amd64 Percona XtraDB Cluster database common files (e.g. /etc/mysql/my.cnf)
ii percona-xtradb-cluster-galera-3.x 3.9.3494.wheezy amd64 Galera components of Percona XtraDB Cluster
ii percona-xtradb-cluster-server-5.6 5.6.22-25.8-978.wheezy amd64 Percona XtraDB Cluster database server binaries
ii xtrabackup 2.2.10-1.wheezy all Transitional package for percona-xtrabackup

we noticed that on disk we have a particular table with different file size on one node then all the other nodes:

root@elvispresley:/mysql/data/example# ls -la api_facebook_com_temp_csv.ibd
-rw-rw---- 1 mysql mysql 4538236928 Sep 9 01:58 api_facebook_com_temp_csv.ibd

while on all other nodes:

root@johnnycash:/mysql/data/example# ls -la api_facebook_com_temp_csv.ibd
-rw-rw---- 1 mysql mysql 11932794880 Sep 9 01:58 api_facebook_com_temp_csv.ibd

When we check inside mysql:

mysql {elvispresley} >SELECT table_name AS "Table", round(((data_length + index_length) / 1024 / 1024), 2) "Size in MB" FROM information_schema.TABLES WHERE table_schema = "example" AND table_name = "api_facebook_com_temp_csv";
+--------------------------+------------+
| Table | Size in MB |
+--------------------------+------------+
| api_facebook_com_temp_csv | 81.02 |
+--------------------------+------------+
1 row in set (0.00 sec)

on all the other nodes:

mysql {johnnycash} >SELECT table_name AS "Table", round(((data_length + index_length) / 1024 / 1024), 2) "Size in MB" FROM information_schema.TABLES WHERE table_schema = "example" AND table_name = "api_facebook_com_temp_csv";
+--------------------------+------------+
| Table | Size in MB |
+--------------------------+------------+
| api_facebook_com_temp_csv | 26.02 |
+--------------------------+------------+
1 row in set (0.00 sec)

However the data seems to be okay:

mysql {elvispresley} >select count(*) from example.api_facebook_com_temp_csv; +----------+
| count(*) |
+----------+
| 1016 |
+----------+
1 row in set (10.19 sec)

while on all others it returns immediately:

mysql {johnnycash} >select count(*) from example.api_facebook_com_temp_csv; +----------+
| count(*) |
+----------+
| 1016 |
+----------+
1 row in set (0.00 sec)

This shouldn't happen?

[Walter Heck (walterheck)]

No response in close to a month?

[Raghavendra D Prabhu (raghavendra-prabhu)]

Please use pt-table-checksum to checksum all the tables across the cluster.

[Launchpad Janitor (janitor)]

[Expired for Percona XtraDB Cluster because there has been no activity for 60 days.]

[Shahriyar Rzayev (rzayev-sehriyar)]

Percona now uses JIRA for bug reports so this bug report is migrated to: https://jira.percona.com/browse/PXC-1850