[
http://jira.jboss.com/jira/browse/JBCACHE-871?page=comments#action_12347564 ]
Brian Stansberry commented on JBCACHE-871:
------------------------------------------
Add edtest testNodeRemove to o.j.c.transaction.IsolationLevelReadComittedTest. Also added
new class IsolationLevelSerializableTest, which shows the issue exists even with
SERIALIZABLE isolation. On 1.4.0 branch. Will port to HEAD later.
Calling remove(FQN) inside a transaction destroys the node before the
commit phase
----------------------------------------------------------------------------------
Key: JBCACHE-871
URL:
http://jira.jboss.com/jira/browse/JBCACHE-871
Project: JBoss Cache
Issue Type: Bug
Security Level: Public(Everyone can see)
Affects Versions: 1.3.0.SP3, 1.4.0.SP1
Reporter: Galder Zamarreno
Assigned To: Manik Surtani
Attachments: Siemens00012761Test.java
Removing a node that has been just been put inside a transaction
(Pessimistic Locking, READ_COMMITTED) destroys the node
without leaving any trace.
Remove should not actually destroy until the commit has been
executed.
This allows for any paralell transactions to get a WL on this node
as there's no any locks any more in this node (it was destroyed)
--
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