[
https://issues.apache.org/jira/browse/JCR-2749?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12910047#action_12910047
]
Jukka Zitting commented on JCR-2749:
------------------------------------
a) It's not illegal, but doing so is practically always a sign of some resource
management error. Thus I think a warning is in place, just like when we log a
warning when a client forgets to close a session (see JCR-1216).
b) Since a duplicate logout() is not strictly speaking illegal (see a), an
assertion is IMHO the wrong thing to use. Generally speaking nothing that a JCR
client can produce by invoking standard JCR methods should ever be able to
trigger an assertion. StackOverflow summarizes this nicely: "Use assertions for
internal logic checks within your code, and normal exceptions for error
conditions outside your immediate code's control."
(http://stackoverflow.com/questions/1276308/exception-vs-assertion)
Also, I'd rather avoid extra behaviour that's only activated during test runs
(I know we have some of them in the codebase), since such code erodes the
purpose of testing by making the test system behave differently than a
production system. If we want a way for the test suite to check against these
situations, we should either make it scan the log file for logged warnings or
add a more generic mechanism by which a client application can specify custom
handling of problems like this.
> Closing a session twice shouldn't write a warning in the log
> ------------------------------------------------------------
>
> Key: JCR-2749
> URL: https://issues.apache.org/jira/browse/JCR-2749
> Project: Jackrabbit Content Repository
> Issue Type: Improvement
> Components: jackrabbit-core
> Affects Versions: 2.2.0
> Reporter: Thomas Mueller
> Assignee: Thomas Mueller
> Priority: Minor
> Fix For: 2.2.0
>
> Attachments: sessionClose.txt
>
>
> When closing a session twice the following warning is written to the log file
> as of JCR-2741:
> "This session has already been closed. See the chained exception for a trace
> of where the session was closed."
> I think the second "close()" should simply be ignored, without warning.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
