[jboss-jira] [JBoss JIRA] Resolved: (JBCACHE-1028) getTables() length > 1 on Derby even when cache table DNE

Mircea Markus (JIRA) jira-events at lists.jboss.org
Sat May 5 16:19:02 EDT 2007


     [ http://jira.jboss.com/jira/browse/JBCACHE-1028?page=all ]

Mircea Markus resolved JBCACHE-1028.
------------------------------------

    Resolution: Done

Couldn't reproduce the issue.
There is an unit test in place relying on the EmbeddedXADataSource for obtaining the database connection; it works fine.
Possible explanation of the problem: 
- even when using derby in embedded mode, the database is created and stored on the file system. "Embedded " refers to the fact that the engine runs in the same JVM, but managed data is still persisted. So if the application runs for the first time, the database will be created, and so is the jboss cache table. All subsequent times when the application runs the 'tableExist()' correctly returns true - that's because the in-process engine works with the same database files.  The place where those DB files are created is specified by derby.system.home sys property. If missing then default is the running dir.

> getTables() length > 1 on Derby even when cache table DNE
> ---------------------------------------------------------
>
>                 Key: JBCACHE-1028
>                 URL: http://jira.jboss.com/jira/browse/JBCACHE-1028
>             Project: JBoss Cache
>          Issue Type: Bug
>      Security Level: Public(Everyone can see) 
>          Components: Cache loaders
>    Affects Versions: 2.0.0.BETA2
>         Environment: Derby 10.2, Managed transactions
>            Reporter: Bill Middleton
>         Assigned To: Mircea Markus
>             Fix For: 2.0.0.CR2
>
>
> The EmbeddedXADataSource which is provided by Apache derby 10.2  always returns a resultset which has length > 1 when calling getTables(), even if the table in question does not exist.  This causes the return value from AdjListJDBCCacheLoader.tableExists() to always be true, and the jbosscache table is never created. 
>  I'm not sure, but perhaps there's a better way to conclude that a table does not exist in the result set other than than its length.

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

        



More information about the jboss-jira mailing list