[
https://issues.apache.org/jira/browse/SHINDIG-257?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12595994#action_12595994
]
Santiago Gala commented on SHINDIG-257:
---------------------------------------
When I create a branch in the previous changeset (which works) and then cherry
pick all subsequent commits except for the one in this patch, all tests pass
with head, which looks quite suspicious again for me.
Using a clean working copy on the changeset r654993:
Results of mvn clean; mvn
(...)
Failed tests:
testCacheableForFutureExpires(org.apache.shindig.gadgets.BasicContentCacheTest)
testCacheableForExpiresWithWait(org.apache.shindig.gadgets.BasicContentCacheTest)
Tests run: 252, Failures: 2, Errors: 0, Skipped: 0
I attach the file relevant to the failed tests.
> I don't trust mvn clean very much.
I really need to wonder why are we using a tool that can't even be trusted to
a) give reports that people can understand
b) clean the workspace
(but then I'm really queer, as all the java world seems to be happy using it)
> shindig handling of HTTP dates is not thread safe
> -------------------------------------------------
>
> Key: SHINDIG-257
> URL: https://issues.apache.org/jira/browse/SHINDIG-257
> Project: Shindig
> Issue Type: Bug
> Reporter: Brian Eaton
> Attachments: dateformat.patch,
> org.apache.shindig.gadgets.BasicContentCacheTest.txt
>
>
> I noticed this during a bit of load testing, spurious exceptions from
> SimpleDateFormat because it isn't thread safe.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
