[ https://forge.continuent.org/jira/browse/SEQUOIA-662?page=all ]
     
Pascal Bresson closed SEQUOIA-662:
----------------------------------


Considered as fixed. WIll be reopened if needed.

> Cannot restore a mysql backend to consistent state after failure: "Error 
> thrown in operation copyLogFromCheckpoint"
> -------------------------------------------------------------------------------------------------------------------
>
>          Key: SEQUOIA-662
>          URL: https://forge.continuent.org/jira/browse/SEQUOIA-662
>      Project: Sequoia
>         Type: Bug
>   Components: Recovery Log
>     Versions: Sequoia 2.9
>  Environment: Windows Server 2003 x64
> JDK 1.5.0u7 32-bit
> Mysql v5.0.22
>     Reporter: Patrick Eger
>      Fix For: Sequoia 2.9

>
>
> Ran into another problem doing failure recovery as outlined in the Admin 
> Guide Section 7.1 "Recover from a controller failure". I was not able to 
> restore the backend on the failed Controller, as the Console (from Controller 
> 2, the non-failed one) would report "Error thrown in operation 
> copyLogFromCheckpoint". The dump i am trying to restore is a mysql native 
> backup with just the initial schema (no data). I think the error is on log 
> recovery though.
> Perhaps this is a mysql specific bug? Should i be useing hsqldb or postgresql 
> for the local logs, ie are they better tested/supported?
> Configuration is based off the demo distributed controller, with 3 primary 
> differences:
> 1) All databases (backends and recovery log) are Mysql v5.0.22, connected via 
> latest mysql 3.1.13 JDBC driver.
> 2) Each Controller only has 1 backend, not 2 as in the demo
> 3) The group communication is using Appia/Hedera latest for CVS as well.
> Please let me know if you need any more info.
> --------------- Controller 2 Admin Console ---------------------
> Launching the Sequoia controller console
> Initializing Controller module...
> Initializing VirtualDatabase Administration module...
> Initializing SQL Console module...
> Sequoia driver (Sequoia core v2.9) successfully loaded.
> localhost:1092 > admin testdb
> Virtual database Administrator Login > admin
> Virtual database Administrator Password >
> Ready to administrate virtual database testdb
> testdb(admin) > show dumps
> Exception thrown in the getter for the attribute AvailableDumps
> testdb(admin) > show backends
> testdb2
> testdb(admin) > show dumps
> +-----------+----------------------------+----------------+-------------+-----------------+-----------+--------+
> | Name      | Checkpoint                 | Format         | Path        | 
> Date            | Backend   | Tables |
> +-----------+----------------------------+----------------+-------------+-----------------+-----------+--------+
> | init_dump | Initial_empty_recovery_log | MySQL raw dump | /tmp/backup | 
> 6/27/06 6:41 PM | testdb1   | *      |
> +-----------+----------------------------+----------------+-------------+-----------------+-----------+--------+
> testdb(admin) > expert on
> Expert mode on
> testdb(admin) > transfer dump init_dump localhost:1091 nocopy
> Transferring dump init_dump to controller localhost:1091...
> Remote dump 'init_dump' already exists.
> testdb(admin) > expert on
> Expert mode on
> testdb(admin) > restore log init_dump localhost:1091
> Error thrown in operation copyLogFromCheckpoint
> testdb(admin) >
> --------------- Controller 1 (localhost:1091) Logs  ---------------------
> Starting Controller 1 in distributed Configuration
> 18:46:50,643 INFO  controller.core.Controller Sequoia controller (2.9)
> 18:46:50,815 INFO  controller.core.Controller Loading configuration file: 
> ..\config\controller\testdb-controller-1.xml
> 18:46:50,940 INFO  controller.core.Controller JMX is enabled
> 18:46:50,971 INFO  controller.core.Controller Starting JMX server on host: 
> localhost
> 18:46:51,487 INFO  controller.backup.BackupManager Registering backuper 
> MySQLBackuper to handle format MySQL raw dump
> 18:46:51,534 INFO  backend.DatabaseBackend.testdb1 Adding connection manager 
> for virtual user "testuser"
> 18:46:52,049 ERROR controller.backend.BackendStateListener Could not store 
> information for backend:testdb1
> java.sql.SQLException: Unable to get checkpoint  from the recovery log 
> (Checkpoint  does not exist in recovery log)
>         at 
> org.continuent.sequoia.controller.recoverylog.events.GetCheckpointLogIdEvent.execute(GetCheckpointLogIdEvent.java:96)
>         at 
> org.continuent.sequoia.controller.recoverylog.LoggerThread.run(LoggerThread.java:857)
> 18:46:52,049 INFO  controller.RequestManager.testdb Request manager will 
> parse requests with the following granularity: TABLE
> 18:46:52,065 INFO  controller.virtualdatabase.testdb Using Hedera properties 
> file: /hedera_appia.properties
> 18:46:53,299 INFO  continuent.hedera.appia AppiaThread: Starting Appia.
> 18:46:59,893 INFO  controller.virtualdatabase.testdb Group testdb connected 
> to Member(address=localhost/10.0.0.7:61618:61618, uid=localhost:61618)
> 18:46:59,893 INFO  controller.virtualdatabase.testdb Group now contains 2 
> controllers.
> 18:46:59,893 INFO  controller.virtualdatabase.testdb Checking virtual 
> database configuration with remote controllers.
> 18:46:59,956 INFO  controller.virtualdatabase.testdb Updating backend list 
> for controller Member(address=localhost/10.0.0.7:61524:61524, 
> uid=localhost:61524)
> 18:46:59,971 WARN  controller.virtualdatabase.testdb Some virtual database 
> users are missing from this configuration, trying to create them 
> transparently...
> 18:46:59,971 INFO  controller.virtualdatabase.testdb Virtual database 
> configuration is compatible with other controllers.
> 18:46:59,971 INFO  controller.virtualdatabase.testdb Controller identifier is 
> set to: 4
> 18:46:59,987 INFO  controller.virtualdatabase.testdb Resyncing recovery log 
> ...
> 18:46:59,987 INFO  controller.virtualdatabase.testdb No shutdown checkpoint 
> found in recovery log. Clearing recovery log (dirty).
> 18:46:59,987 INFO  controller.virtualdatabase.testdb Please resync manually 
> using 'recover log'.
> 18:47:00,159 INFO  controller.virtualdatabase.testdb Resyncing recovery log 
> done
> 18:47:00,174 INFO  controller.virtualdatabase.testdb Unexpected enabled state 
> (1) for backend testdb1. Forcing backend to disabled state.
> 18:47:00,174 WARN  controller.core.Controller Failed to automatically enable 
> backends, manual resynchronization is probably needed
> 18:47:00,174 INFO  controller.core.Controller Adding VirtualDatabase testdb 
> 18:47:00,190 INFO  controller.core.Controller Waiting for connections on 
> 10.0.0.7:25325
> 18:47:00,206 INFO  controller.core.Controller Controller started on 
> 2006.06.27 26 at 06:47:00 PM PDT
> 18:47:00,206 INFO  controller.core.Controller Controller localhost:25325 
> ready, listening to requests ...
> 13:39:21,984 INFO  controller.virtualdatabase.testdb Checkpoint 
> now-localhost: 25326-Wed_Jun_28_13:39:21_PDT_2006 was stored
> 13:39:22,452 ERROR sequoia.controller.recoverylog Failed to log begin for 
> transaction 1
> java.sql.SQLException: No operations allowed after statement closed.
>         at com.mysql.jdbc.Statement.checkClosed(Statement.java:274)
>         at 
> com.mysql.jdbc.ServerPreparedStatement.checkClosed(ServerPreparedStatement.java:411)
>         at 
> com.mysql.jdbc.ServerPreparedStatement.setLong(ServerPreparedStatement.java:1723)
>         at 
> org.continuent.sequoia.controller.recoverylog.events.LogRequestEvent.execute(LogRequestEvent.java:75)
>         at 
> org.continuent.sequoia.controller.recoverylog.LoggerThread.run(LoggerThread.java:857)

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
   https://forge.continuent.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see:
   http://www.atlassian.com/software/jira


_______________________________________________
Sequoia mailing list
[email protected]
https://forge.continuent.org/mailman/listinfo/sequoia

Reply via email to