[
https://issues.apache.org/jira/browse/JCR-1419?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12572882#action_12572882
]
Alexander Klimetschek commented on JCR-1419:
--------------------------------------------
Agreed, if there is no difference in the exception thrown and the handling of
that exception, the simpler case is to be preferred.
My motivation for harsh tests comes from reading the excellent "Release It!"
book by Michael Nygard which talks about making systems stable in production
(he gives a lot of Java examples and typical pitfalls). One of his examples
clarifies that sometimes it does matter to look at the details of the
connection loss - eg. if a TCP server does not answer the SYN with an ACK
(because of some new firewall setting or whatever), the JDBC connection would
hang forever, which is worse than actually failing. But I think I am way
off-topic now ;-)
> Add test for database auto-reconnection
> ---------------------------------------
>
> Key: JCR-1419
> URL: https://issues.apache.org/jira/browse/JCR-1419
> Project: Jackrabbit
> Issue Type: Test
> Components: jackrabbit-core, test
> Reporter: Alexander Klimetschek
> Assignee: Jukka Zitting
> Fix For: 1.5
>
> Attachments: DatabaseConnectionFailureTest.java,
> jackrabbit-core.test-for-1400.patch
>
>
> Create a unit test for the database auto-reconnect feature (JCR-940).
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
