[
https://issues.apache.org/jira/browse/DERBY-2141?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Mike Matrigali updated DERBY-2141:
----------------------------------
I don't know if this is the bug, but I think you are right about
OverflowInputStream. Most of store doesn't require synchronization as it
assumes single thread access to it's objects as controlled by the derby's jdbc
level which only allows one thread into a given
xact at the same time.
But, OverflowInputStream is handed back to the caller and then can be accessed
outside of jdbc. Does jdbc say anything about
callers responsibility regarding multi-thread access to a stream handed back
through jdbc? Seems safer if possible to just
correctly synchronize OverflowInputStream, but this will add some overhead to
the streams coming out of the server.
> BlobClob4BlobTest.testPositionBlob() fails with NullPointerException
> --------------------------------------------------------------------
>
> Key: DERBY-2141
> URL: https://issues.apache.org/jira/browse/DERBY-2141
> Project: Derby
> Issue Type: Bug
> Components: Regression Test Failure
> Affects Versions: 10.3.0.0
> Environment: Java SE 6, Java SE 5.0 (less frequently). Linux and
> Solaris.
> Reporter: Knut Anders Hatlen
> Priority: Minor
> Attachments: stacktrace.diff
>
>
> Seen in the nightly regression tests:
> http://dbtg.thresher.com/derby/test/Daily/jvm1.6/testing/testlog/lin/481811-org.apache.derbyTesting.functionTests.suites.All_diff.txt
> http://dbtg.thresher.com/derby/test/Daily/jvm1.6/testing/testlog/lin/480626-org.apache.derbyTesting.functionTests.suites.All_diff.txt
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
