tomcat-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Brian Burch <>
Subject Re: [VOTE] Release Apache Tomcat 7.0.40
Date Mon, 06 May 2013 17:10:05 GMT
On 06/05/13 15:27, Konstantin Kolinko wrote:
> 2013/5/6 Brian Burch <>:
>> On 05.05.2013 12:44, Mark Thomas wrote:
>>> The proposed Apache Tomcat 7.0.40 release is now available for voting.
>>> It can be obtained from:
>>> The Maven staging repo is:
>>> The svn tag is:
>>> The proposed 7.0.40 release is:
>>> [ ] Broken - do not release
>>> [X] Stable - go ahead and release as 7.0.40 Stable
>> I would like to use this release as early as possible, so I picked up the
>> source, built and tested it. I noticed some errors which do not matter in my
>> environment, but because I haven't tested any earlier releases, I don't know
>> whether they are generally acceptable. I couldn't find a reference to them
>> being known problems, and (of course) it could be a "user error".
>> environment:
>> Linux version 3.5.0-28-lowlatency (buildd@roseapple) (gcc version 4.7.2
>> (Ubuntu/Linaro 4.7.2-2ubuntu1) ) #32-Ubuntu SMP PREEMPT Fri Apr 26 11:05:36
>> UTC 2013 (Ubuntu 3.5.0-28.32-lowlatency
>> brian@bacchus:~$ /usr/lib/jvm/java-6-openjdk-i386/bin/java -version
>> java version "1.6.0_27"
>> OpenJDK Runtime Environment (IcedTea6 1.12.3) (6b27-1.12.3-0ubuntu1~12.10.1)
>> OpenJDK Server VM (build 20.0-b12, mixed mode)
>> I have /real/ test failures for both BIO and NIO in:
>> for example, one of them is:
>> Testcase: testMemberArrival took 7.191 sec FAILED
>> Checking member arrival length (Listener-10) expected:<9> but was:<0>
>> junit.framework.AssertionFailedError: Checking member arrival length
>> (Listener-10) expected:<9> but was:<0>
>>          at
>> I don't want to waste anybody's time, but perhaps you could let me know
>> whether the errors are significant? If you think it is because I've made a
>> simple error in my configuration, perhaps you point me in the right
>> direction to resolve it.
> Tribes is communication layer that is used to implement clustering
> (aka "ha", high availability)
> If you do not use clustering, then those do not matter for you.
> You can compare your test log files with ones produced by buildbot:
> If you want to (re)run a single test from the testsuite, BUILDING.txt
> says how to do so.
> Maybe some network or security configuration at your box prevents
> those tests from working? IPv6? Any error messages in the logs?
> The tests respect the "conf/" file, so it is
> possible to enable fine logging if you need so.

Thank you for your suggestions, Konstantin. Thanks also to Rainer for 
his suggestions.

My primary objective was to raise the issue at this opportune moment, 
just in case it turned out to be a general problem.

I am satisfied with your theory that the failures I see are related to 
multicast support on my particular development system. As I said at the 
start, I do not need tribes support on my production system.

When I deploy the new release on my production system, hopefully later 
this week, I will run all the tests there too. It (now) seems quite 
likely that I will observe different behaviour under the different 
execution environment.

If anything strikes me as strange, then I will raise the issue, with 
more diagnostics, on the users list.

In the meantime, please discount my report when considering the 
stability of 7.0.40.



> Best regards,
> Konstantin Kolinko
> ---------------------------------------------------------------------
> To unsubscribe, e-mail:
> For additional commands, e-mail:

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message