Comments - Migrating from InfiniDB 4.x to MariaDB ColumnStore

7 years, 3 months ago Kaifeng Chen

It seems that it doesn't work for migration from Infinidb 4.6.6 to MCS 1.0.6。

[root@localhost columnstore]# mcsadmin startSystem
startsystem Thu Jan 5 11:21:55 2017
startSystem command, 'columnstore' service is down, sending command to start the 'columnstore' service on all modules

System being started, please wait..................
startSystem Failed : check log files

crit.log

Jan 5 11:22:17 localhost controllernode[6815]: 17.557831 |0|0|0| C 29 CAL0000: ExtentMap::load(): open: No such file or directory
Jan 5 11:22:42 localhost DMLProc[7120]: 42.781900 |0|0|0| C 20 CAL0002: DMLProc failed to start due to : Rollback will be deferred due to DBRM is in read only state.
Jan 5 11:27:38 localhost controllernode[8209]: 38.129917 |0|0|0| C 29 CAL0000: ExtentMap::save(): got request to save an empty BRM

err.log

Jan 5 11:22:17 localhost ProcessMonitor[6368]: 17.561537 |0|0|0| E 18 CAL0000: Error return DBRM load_brm
Jan 5 11:22:42 localhost controllernode[6453]: 42.193287 |0|0|0| E 29 CAL0000: DBRM: error: SessionManager::setSystemState() failed (network)

warning.log

Jan 5 11:22:45 localhost ProcessMonitor[6368]: 45.804749 |0|0|0| W 18 CAL0000: DMLProc failed initialization

mcsadmin> getProcessStatus
getprocessstatus Thu Jan 5 11:34:56 2017

MariaDB Columnstore Process statuses

Process Module Status Last Status Change Process ID
------------------ ------ --------------- ------------------------ ----------
ProcessMonitor pm1 ACTIVE Thu Jan 5 11:22:05 2017 6368
ProcessManager pm1 ACTIVE Thu Jan 5 11:22:11 2017 6453
DBRMControllerNode pm1 FAILED Thu Jan 5 11:22:17 2017
ServerMonitor pm1 ACTIVE Thu Jan 5 11:22:20 2017 6829
DBRMWorkerNode pm1 ACTIVE Thu Jan 5 11:22:20 2017 6846
DecomSvr pm1 ACTIVE Thu Jan 5 11:22:24 2017 6903
PrimProc pm1 ACTIVE Thu Jan 5 11:22:26 2017 6923
ExeMgr pm1 ACTIVE Thu Jan 5 11:22:31 2017 6967
WriteEngineServer pm1 ACTIVE Thu Jan 5 11:22:36 2017 7050
DDLProc pm1 ACTIVE Thu Jan 5 11:22:38 2017 7089
DMLProc pm1 FAILED Thu Jan 5 11:23:01 2017 7120
mysqld pm1 ACTIVE Thu Jan 5 11:22:23 2017 6713

How can we sovle this issue? Thanks.

 
7 years, 3 months ago David Thompson

Are you sure you completely removed the data1 etc directory in the 1.0.6 columnstore file before restoring the infinidb data1 directory to avoid any overlap of files?

Is there any difference in the directory structure between the old and new otherwise?

If you have an infinidb support contract with mariadb it would be better to file this through support channels.

 
7 years, 3 months ago Kaifeng Chen

We have moved the data1 directory of columnstore 1.0.6 to /tmp instead of removing it.

The data directories(data, data1) of the old and new are the same(after migration).

Actually, we setup two new instance: one is InfiniDB 4.6.6(single mode), the other is columnstore 1.0.6. For the purpose of testing, we just create one table(say, mydb. t192_168_3_215_if011_generic_s_app_host_flow_1hour) in these two instance.
It's OK to create database &table, and can enter mcsmysql console. But can not fetch data from this table after migration( use the old data1).

[root@localhost columnstore]# mcsmysql
Welcome to the MariaDB monitor. Commands end with ; or \g.
Your MariaDB connection id is 4
Server version: 10.1.19-MariaDB Columnstore 1.0.6-1

Copyright © 2000, 2016, Oracle, MariaDB Corporation Ab and others.

Type 'help;' or '\h' for help. Type '\c' to clear the current input statement.

MariaDB [(none)]> use mydb;
Database changed
MariaDB [mydb]> show tables;
+----------------------------------------------------+

Tables_in_mydb

+----------------------------------------------------+

t192_168_3_215_if011_generic_s_app_host_flow_1hour

+----------------------------------------------------+
1 row in set (0.00 sec)

MariaDB [mydb]> show create table t192_168_3_215_if011_generic_s_app_host_flow_1hour;

t192_168_3_215_if011_generic_s_app_host_flow_1hourCREATE TABLE `t192_168_3_215_if011_generic_s_app_host_flow_1hour` (

`start_time` int(10) unsigned DEFAULT NULL,
`end_time` int(10) unsigned DEFAULT NULL,
`host_value` int(10) unsigned DEFAULT NULL,
`input_octets` int(10) unsigned DEFAULT NULL,
`output_octets` int(10) unsigned DEFAULT NULL,
`total_octets` int(10) unsigned DEFAULT NULL
) ENGINE=Columnstore DEFAULT CHARSET=utf8 COMMENT='schema sync only' |

1 row in set (0.00 sec)

MariaDB [mydb]> select * from t192_168_3_215_if011_generic_s_app_host_flow_1hour;
ERROR 1815 (HY000): Internal error: IDB-2033: Error occurred when calling system catalog.
MariaDB [myb]>

 
7 years, 3 months ago David Hill

Bugs can be reported in MariaDB Jira: https://jira.mariadb.org (see Reporting Bugs). Please file bugs under the MCOL project.

Please open a BUG in JIRA and you will be able to get support from there. This area is for comments on the document only.

Thanks, David Hill

 
7 years, 3 months ago David Thompson

I retested and could reproduce, there was some missing steps fixing a dbrm file so this was a doc bug. kaifengchen - can you verify?

 
7 years, 3 months ago Kaifeng Chen

It's OK now. Thanks.

 
Content reproduced on this site is the property of its respective owners, and this content is not reviewed in advance by MariaDB. The views, information and opinions expressed by this content do not necessarily represent those of MariaDB or any other party.