1077641 - Calls not connecting when I initiate (but can connect to someone else's calls fine)

Status: RESOLVED INCOMPLETE

(Hello (Loop) :: Client, defect)

Description

[[:Cww]]

On Nightly: 35.0a1 (2014-10-01)

When I initiate a call via the Loop interface, I get a URL:

https://call.mozilla.com/#call/cYOXpjE7owI (for example)

Sending that room to someone else, it looks like it's should be working (I get the ready to start conversation window and if I click the green button, I get the connecting with a red button.) But after a minute or two, it just goes back to the Ready to Start screen with no errors.

The same happens on the other end: no video screen just Ready to Start and then Connecting and then back to Ready to Start.  I've tried this with numerous other people at the other end.

However if the other person starts the room and I click on their link, I get the Ready to Start and then it joins fine.

Comment 1

[Mark Banner (:standard8)]

:cww, it might be useful if you could come into the #loop channel and chat with some of the developers there to aid debugging this.

In the meantime, turning on websocket debugging might help you with understanding why the connection isn't working:

https://wiki.mozilla.org/Loop/Debugging

Comment 2

[[:Cww]]

Attachment: browser console log

Comment 3

[[:Cww]]

I just tried making a new Loop link: https://call.mozilla.com/#call/Pgm66ax9rzk It basically doesn't work for anyone.

Comment 4

[Mark Banner (:standard8)]

You wouldn't happen to be in "Do not disturb" mode, by any chance?

At the left-hand bottom corner of the panel, there's a drop-down for available or do not disturb. Make sure that's set to available (this is also reflected in the loop.do_not_disturb pref).

Comment 5

[u279076]

Any update here? Are you still able to reproduce this?

Comment 6

[[:Cww]]

Yeah, still doesn't work.

Comment 7

[u279076]

Mark, is there anything in Cheng's log that helps debug this?

Comment 8

[Mark Banner (:standard8)]

(In reply to Anthony Hughes, QA Mentor (:ashughes) from comment #7)
> Mark, is there anything in Cheng's log that helps debug this?

Unfortunately no. I think the best thing to do would be to try and find someone in #loop who can help you debug.

My suspicion is this is on the backend side, although I could be wrong. I'll most likely be around tomorrow morning PDT, so if Cheng's around then that could be a good time.

Comment 9

[u279076]

Cheng, can you please join #loop and ask for help debugging this issue further? If standard8 or mreavy are around they should be able to assist.

Comment 10

[[:Cww]]

I've been hanging out in #loop, asking occasionally. No help yet.

Comment 11

[u279076]

(In reply to [:Cww] from comment #10)
> I've been hanging out in #loop, asking occasionally. No help yet.

Try pinging Standard8.

Comment 12

[Maire Reavy [:mreavy]]

Cww -- Is there a preferred time when Standard8 and I can reach out to you in irc?  Perhaps scheduling a time to meet in irc and debug (for say 30 mins) is the best way to go?

Comment 13

[:shell escalante]

not able  to reproduce - if Chad is still seeing, reply back to bugmail and will set up time to talk to standard8 (mark).

Comment 14

[Mark Banner (:standard8)]

We've changed a lot since the bug was filed, and we're not using the previous call mechanism.

Cww, if this is still affecting you, please lets us know, I suggest a new bug or talking to us on irc. Having the room url of a room where you saw the issue would give us extra information now as we're tracking call setup in more detail.