[jbossseam-issues] [JBoss JIRA] Commented: (JBSEAM-3188) ConcurrentModificationException on flush of ServerConversationContext
[
https://jira.jboss.org/jira/browse/JBSEAM-3188?page=com.atlassian.jira.pl...
]
Sanne Grinovero commented on JBSEAM-3188:
-----------------------------------------
Hi Pete, nice to see you commenting on this.
You're absolutely right about an example, but this is hard to isolate from the complex
code, as it doesn't reveal itself in simple situations.
When using the "patch" we made we don't get the exception, so this should be
a proof of the fact that
somehow the flush method is doing some potential modification of the "additions"
HashMap.
This didn't happen in version 2.0.2 and previous.
I'll see if we manage to isolate the situation, but that seems unlikely.
Either Seam shouldn't make changes of the map in this block, or make a copy of it as
in the attached code.
What are your thoughts?
ConcurrentModificationException on flush of
ServerConversationContext
---------------------------------------------------------------------
Key: JBSEAM-3188
URL: https://jira.jboss.org/jira/browse/JBSEAM-3188
Project: Seam
Issue Type: Bug
Components: Core
Affects Versions: 2.0.3.CR1
Environment: 2.0.3.CR1, JBoss 4.2.2, JDK 5
Reporter: Sanne Grinovero
Attachments: JBSEAM-3188.patch
As confirmed by other people on the forum at
http://seamframework.org/Community/ConcurrentModificationExceptionAfterPr...
It is possible for ServerConversationContext to iterate on an EntrySet which is modified
in the iteration block.
I couldn't understand exactly in which scenarios it happens, but it's easy to
patch the code
to iterate on a copy instead (I'll attach the patch).
I can't say if it's correct to make a copy, or if you should actually avoid the
changes.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
https://jira.jboss.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira