<div dir="ltr">Ok, had a chance to try out the test branch, and also have some more clarity on how the issue has occurred for me.<div><br></div><div>Turns out that the issue happens for me when talking between two vagrant machines on my development laptop, which do indeed use prflx (or local) candidates. I tested with the latest master branch and confirmed that the prflx candidates were failing, then with the branch at <a href="https://git.collabora.com/cgit/user/tester/libnice.git/log/?h=bellet" rel="noreferrer" target="_blank" style="font-size:12.800000190734863px">https://git.collabora.com/cgit<wbr>/user/tester/libnice.git/log/?<wbr>h=bellet</a> and confirmed that all connections were now succeeding (although in this success case, it appears that local candidates were being finally selected as the used pair in some cases).</div><div><br></div><div>I will rescind my earlier comment that srvflx candidates might also be failing, as I probably never actually tested with servers on a public IP after they were failing in my dev environment.</div><div><br></div><div>I’m attaching Chrome debug logs of both the failed connections on the master libnice branch, and the successful connections on the bellet branch.</div><div><br></div><div>Hopeful these fixes get into mainline!</div><div><br></div><div>Chad</div></div><div class="gmail_extra"><br><div class="gmail_quote">On Wed, Apr 12, 2017 at 11:07 AM, Chad Phillips <span dir="ltr"><<a href="mailto:chad@apartmentlines.com" target="_blank">chad@apartmentlines.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr">Olivier,<div><br></div><div>I’m on vacation this week, I’ve added testing this patch to my todo list for when I get back, thanks for looking into it!</div><span class="HOEnZb"><font color="#888888"><div><br></div><div>Chad</div></font></span></div><div class="HOEnZb"><div class="h5"><div class="gmail_extra"><br><div class="gmail_quote">On Wed, Apr 12, 2017 at 5:03 AM, Lorenzo Miniero <span dir="ltr"><<a href="mailto:lminiero@gmail.com" target="_blank">lminiero@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr"><div class="gmail_extra"><div class="gmail_quote"><span>2017-04-12 0:00 GMT+02:00 Olivier Crête <span dir="ltr"><<a href="mailto:olivier.crete@collabora.com" target="_blank">olivier.crete@collabora.com</a>></span>:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Hi,<br>
<br>
Can you try testing this branch? and let me know how it works for you?<br>
<br>
<a href="https://git.collabora.com/cgit/user/tester/libnice.git/log/?h=bellet" rel="noreferrer" target="_blank">https://git.collabora.com/cgit<wbr>/user/tester/libnice.git/log/?<wbr>h=bellet</a><br>
<br></blockquote><div><br></div><div><br></div></span><div>Hi Olivier,</div><div><br></div><div>thanks for the quick patch!</div><div><br></div><div>I asked my colleagues to test this, and it looks like it fixes the issue in one scenario (Janus on AWS), but it doesn't in another (Janus in a docker container with a bridge network, which behaves like a symmetric NAT), where both were working fine with 0.1.13. For the sake of completeness, though, the scenario where it still doesn't seem to work, doesn't work with D735 either, so not sure what this means there.</div><div><br></div><div>Thanks,</div><div>Lorenzo</div><div><div class="m_1154220229532820401h5"><div><br></div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
Olivier<br>
<div class="m_1154220229532820401m_4014152085393166891gmail-HOEnZb"><div class="m_1154220229532820401m_4014152085393166891gmail-h5"><br>
<br>
On Mon, 2017-04-10 at 16:28 -0400, Olivier Crête wrote:<br>
> Hi,<br>
><br>
> That is really interesting,<br>
><br>
> On Mon, 2017-04-10 at 18:42 +0200, Lorenzo Miniero wrote:<br>
> > 2017-02-27 22:30 GMT+01:00 Chad Phillips <<a href="mailto:chad@apartmentlines.com" target="_blank">chad@apartmentlines.com</a>>:<br>
> > > Some git bisect work got me to the problem commit:<br>
> > > 1ab9d7c104978ea1904aaaad708c1c<wbr>8c23c77592 is the first bad commit<br>
> > > commit 1ab9d7c104978ea1904aaaad708c1c<wbr>8c23c77592<br>
> > > conncheck: Separate valid and succeded states<br>
> > He then tried a patch another user referenced on our issue page,<br>
> > and it worked instead: <a href="https://phabricator.freedesktop.org/D735" rel="noreferrer" target="_blank">https://phabricator.f<wbr>reedesktop.org/D735</a><br>
><br>
> I've been looking at this patch a couple times and the reason I<br>
> didn't merge it is that I don't understand what it tries to<br>
> accomplish.<br>
><br>
> But now I think I figured it out. It seems that libnice sets the<br>
> "nominated" flag on all the pairs when it creates them as a<br>
> controlling agent, which seems wrong. I think it should set the flag<br>
> only when it gets a reply... And this is why propagating the flag<br>
> fixes it in some cases, but I think we need to go over the whole<br>
> conncheck code and make it set the nominated flag at the right time<br>
> (after the reply is received!).<br>
><br>
> -- <br>
> Olivier Crête<br>
> <a href="mailto:olivier.crete@collabora.com" target="_blank">olivier.crete@collabora.com</a><br>
</div></div><div class="m_1154220229532820401m_4014152085393166891gmail-HOEnZb"><div class="m_1154220229532820401m_4014152085393166891gmail-h5">> ______________________________<wbr>_________________<br>
> nice mailing list<br>
> <a href="mailto:nice@lists.freedesktop.org" target="_blank">nice@lists.freedesktop.org</a><br>
> <a href="https://lists.freedesktop.org/mailman/listinfo/nice" rel="noreferrer" target="_blank">https://lists.freedesktop.org/<wbr>mailman/listinfo/nice</a><br>
</div></div><span class="m_1154220229532820401m_4014152085393166891gmail-HOEnZb"><font color="#888888">--<br>
Olivier Crête<br>
<a href="mailto:olivier.crete@collabora.com" target="_blank">olivier.crete@collabora.com</a><br>
</font></span></blockquote></div></div></div><br></div></div>
</blockquote></div><br></div>
</div></div></blockquote></div><br></div>