[ 
https://issues.apache.org/jira/browse/DERBY-5237?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13035665#comment-13035665
 ] 

Jim Newsham commented on DERBY-5237:
------------------------------------

Thanks for the comments Dag.

Unfortunately JIRA says that I can't vote for this issue since I created it (I 
guess the logic is that there's an implicit vote of 1 from the creator).  Not 
sure if that matters since this issue has been closed as "duplicate".  I have 
voted for DERBY-5233.

Thanks for the workaround suggestion.  It seems like it should work, with some 
limitations, therefore a proper fix of DERBY-5233 would be beneficial when it 
happens.  For example, with code 08000 I know it's an interrupt, so I can keep 
retrying forever until I succeed (with the assumption that eventually the 
operation will complete before the next interrupt).  With code XSDF1 I don't 
know what the true cause is (could be a filesystem permission, for example), so 
I can retry a few times but I'll need to eventually give up, making the error 
handling less robust.

> interrupting thread performing "create table" doesn't cause SQLException with 
> sql state "08000"
> -----------------------------------------------------------------------------------------------
>
>                 Key: DERBY-5237
>                 URL: https://issues.apache.org/jira/browse/DERBY-5237
>             Project: Derby
>          Issue Type: Bug
>    Affects Versions: 10.8.1.2
>         Environment: Windows 7, JDK 1.6.0_25
>            Reporter: Jim Newsham
>         Attachments: DerbyInterruptTest.java
>
>
> We would like to take advantage of the recent Derby enhancement for handling 
> thread interruption.  
> http://db.apache.org/derby/docs/dev/devguide/devguide-single.html, section 
> "Working with database threads in an embedded environment" states that in the 
> case of exception, the calling code should receive a SQLException with code 
> 08000.  In my testing, this appears to work for sql "insert" statements, but 
> not for sql "create table" statements.  This is an issue for us since 
> database tables are created dynamically over the course of the application's 
> normal operation.
> Junit test repro to follow.

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira

Reply via email to