[Nice] RELEASE: libnice 0.1.6

Youness Alaoui kakaroto at kakaroto.homelinux.net
Tue Apr 29 17:20:16 PDT 2014


try with :
export NICE_DEBUG=all
I'm guessing it will stop failing. It's a race condition in the unit test I
think because it never fails if I enable debugging. The test-send-recv, I
never got it to work myself with or without debug.



On Mon, Apr 28, 2014 at 10:49 PM, Bruce Dubbs <bruce.dubbs at gmail.com> wrote:

> Olivier Crête wrote:
>
>> On Mon, 2014-04-28 at 21:17 -0500, Bruce Dubbs wrote:
>>
>>> Olivier Crête wrote:
>>>
>>>> This is mostly a bugfix release from 0.1.5.
>>>>
>>>> But there are a couple new APIs to make it possible to change the relay
>>>> server used by a connection without breaking it.
>>>>
>>>
>>> Grabbed the tarball and it built fine, but one test failed:
>>>
>>> $ cat test-io-stream-pollable.trs
>>> :test-result: FAIL
>>> :global-test-result: FAIL
>>> :recheck: yes
>>> :copy-in-global-log: yes
>>>
>>> That is about the only thing I can find.  No details.
>>>
>>> What should I check?
>>>
>>> Linux kernel: 3.13.3.
>>>
>>
>> Can you just run tests/test-io-stream-pollable directly and see what
>> happens?
>>
>
> Seemed to hang.
>
> Killed it and ran make clean;./configure;make;make check
>
> Now test-send-recv has been running at 100% cpu for 5 minutes.
>
> Killed that.  Re-extracted the tarball.
> ./configure;make;make check
>
> FAIL: test-io-stream-pollable
>
> Running tests/test-io-stream-pollable just gives an immediate return.
> I can try running it with gdb if you want, bu to need to install that on
> this system (x86_64, gcc-4.8.2)
>
>   -- Bruce
>
>
>
>
>
>
>
> _______________________________________________
> nice mailing list
> nice at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/nice
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/nice/attachments/20140429/0cda0738/attachment.html>


More information about the nice mailing list