1111618 - No video frames flowing with OpenH264

Status: RESOLVED DUPLICATE of bug 1110976

(Core :: WebRTC: Signaling, defect)

Description

[Andreas Pehrson [:pehrsons]]

With bug 879717, we're changing the media element to not progress the ready state to HAVE_METADATA or HAVE_CURRENT_DATA until we display a video frame.

It now breaks the basicH264 test: https://treeherder.mozilla.org/ui/#/jobs?repo=try&revision=61090a7a7e25. I'm assuming because a video frame is never set on the VideoFrameContainer. It works for B2G which should mean this only applies to OpenH264. I can also reproduce locally - seeing that one of the media elements never receives the "canplaythrough" event, it matches my suspicion that a video frame is never delivered.

I've bisected it down to being one of the commits from bug 1091242 (hence signaling), but I can't tell which of the commits break it as they seem to be dependent on each other.

Comment 1

[Andreas Pehrson [:pehrsons]]

I can do some debugging here to try to fix it, but if you have any hints to where I should look Byron, I'm all ears.

Comment 2

[Andreas Pehrson [:pehrsons]]

I see this in the log when running it locally btw. What's 12606?

265 INFO Analyzing element: pcLocal_remote_video
266 INFO TEST-FAIL | /tests/dom/media/tests/mochitest/test_peerConnection_basicH264Video.html | The author of the test has indicated that flaky timeouts are expected.  Reason: untriaged 
267 INFO Analyzing element: pcLocal_remote_video
268 INFO TEST-FAIL | /tests/dom/media/tests/mochitest/test_peerConnection_basicH264Video.html | The author of the test has indicated that flaky timeouts are expected.  Reason: untriaged 
WARNING: no real random source present!
180883456[1003428c0]: [Socket Thread|WebrtcVideoSessionConduit] VideoConduit.cpp:1120: ReceivedRTPPacket RTP Processing Failed 12606
269 INFO Analyzing element: pcLocal_remote_video
270 INFO TEST-FAIL | /tests/dom/media/tests/mochitest/test_peerConnection_basicH264Video.html | The author of the test has indicated that flaky timeouts are expected.  Reason: untriaged 
271 INFO canplaythrough fired for media element pcLocal_remote_video
180883456[1003428c0]: [Socket Thread|WebrtcVideoSessionConduit] VideoConduit.cpp:1120: ReceivedRTPPacket RTP Processing Failed 0
272 INFO Analyzing element: pcLocal_remote_video
273 INFO TEST-PASS | /tests/dom/media/tests/mochitest/test_peerConnection_basicH264Video.html | Media flowing for pcLocal_remote_video 
180883456[1003428c0]: [Socket Thread|WebrtcVideoSessionConduit] VideoConduit.cpp:1120: ReceivedRTPPacket RTP Processing Failed 0
274 INFO Run step: PC_REMOTE_CHECK_MEDIA_FLOW_PRESENT
275 INFO Analyzing element: pcRemote_local_video
276 INFO TEST-PASS | /tests/dom/media/tests/mochitest/test_peerConnection_basicH264Video.html | Media flowing for pcRemote_local_video 
277 INFO Analyzing element: pcRemote_remote_video
278 INFO TEST-FAIL | /tests/dom/media/tests/mochitest/test_peerConnection_basicH264Video.html | The author of the test has indicated that flaky timeouts are expected.  Reason: untriaged 
180883456[1003428c0]: [Socket Thread|WebrtcVideoSessionConduit] VideoConduit.cpp:1120: ReceivedRTPPacket RTP Processing Failed 0
180883456[1003428c0]: [Socket Thread|WebrtcVideoSessionConduit] VideoConduit.cpp:1120: ReceivedRTPPacket RTP Processing Failed 0
279 INFO Analyzing element: pcRemote_remote_video
280 INFO TEST-FAIL | /tests/dom/media/tests/mochitest/test_peerConnection_basicH264Video.html | The author of the test has indicated that flaky timeouts are expected.  Reason: untriaged 
180883456[1003428c0]: [Socket Thread|WebrtcVideoSessionConduit] VideoConduit.cpp:1120: ReceivedRTPPacket RTP Processing Failed 0
180883456[1003428c0]: [Socket Thread|WebrtcVideoSessionConduit] VideoConduit.cpp:1120: ReceivedRTPPacket RTP Processing Failed 0
180883456[1003428c0]: [Socket Thread|WebrtcVideoSessionConduit] VideoConduit.cpp:1120: ReceivedRTPPacket RTP Processing Failed 0
180883456[1003428c0]: [Socket Thread|WebrtcVideoSessionConduit] VideoConduit.cpp:1120: ReceivedRTPPacket RTP Processing Failed 0
281 INFO Analyzing element: pcRemote_remote_video
282 INFO TEST-FAIL | /tests/dom/media/tests/mochitest/test_peerConnection_basicH264Video.html | The author of the test has indicated that flaky timeouts are expected.  Reason: untriaged 
180883456[1003428c0]: [Socket Thread|WebrtcVideoSessionConduit] VideoConduit.cpp:1120: ReceivedRTPPacket RTP Processing Failed 0
180883456[1003428c0]: [Socket Thread|WebrtcVideoSessionConduit] VideoConduit.cpp:1120: ReceivedRTPPacket RTP Processing Failed 0

Comment 3

[Ethan Hugg [:ehugg]]

This may be a duplicate of bug 1110976 - there's a patch there from Byron that should fix.

Comment 4

[Andreas Pehrson [:pehrsons]]

Ah. Yes. Sweet! Works for me with his patch as well.

Thanks Ethan.