881388 - [B2G] [Inari] [YouTube] Unsupported format/invalid file error message appears while playing a video

Status: VERIFIED INVALID

(Core :: Audio/Video, defect, P1)

Description

[Chris Kreinbring [:CKreinbring]]

Attachment: Screenshot of error message during YouTube video play

Description:
While watching a YouTube video, the user will sometimes see an error message.  This doesn't happen every time, and it tends to happen more often with longer videos.

Repro Steps:
1) Updated to Inari Build ID: 20130610070206
2) Launch the browser and navigate to YouTube.com
3) Select a video to view, the longer the better.
4) Watch the video through its full length.

Actual:
The video plays through its full length with no errors.

Expected:
At some point in the video the user will see the error message "Video cannot be played: the format is unsupported or the file is invalid".  When the user taps the OK button they are returned to the YouTube browser page.

Environmental Variables
Occurs on Inari 1.1 commercial RIL
Gecko: http://hg.mozilla.org/releases/mozilla-b2g18/rev/8e3f39363c54
Gaia: ce3b99781d182ad550a325206990c249b0dbcf0e
Platform Version: 18.0

Notes:
Repro frequency: 70%
Q Analysts Team Priority: Pri 2
See attached screenshot

Comment 1

[Jason Smith [:jsmith]]

It would be interesting to know if this reproduces when testing https://m.youtube.com/?tsp=1&player=html5&monetize=1.

Comment 2

[Ben Francis [:benfrancis]]

Hopefully someone from Video won't mind diagnosing this, even if it turns out not to be a video bug.

Comment 3

[Jason Smith [:jsmith]]

The error thrown from YouTube in the video app appears to not be specific to the Video app after looking at the Gaia video source. This sounds like a bug where we will stop playing h264 media content with an error for the file type being non-supported. On that regard, I have a strong feeling this could reproduce in other cases outside of the YouTube in video app use case.

Sotaro probably knows more of the technical implications here though.

Comment 4

[Sotaro Ikeda [:sotaro]]

There is 2 possibilities "netowrk error" and "decode error". If either error happens, error is thrown and video playback is shutdown. From the screen shot, it is not clear which error happens.

Comment 5

[Sotaro Ikeda [:sotaro]]

I tried to reproduce the problem by using v1.1 leo. I can not reproduce it until now.

Comment 6

[Jason Smith [:jsmith]]

QAWanted - I need to know exactly what video you reproduced this and a confirmation this is still reproducible with that video.

Comment 7

[Sotaro Ikeda [:sotaro]]

Adb logcat might help to analyze the problem.

Comment 8

[cajbir (:cajbir)]

Does the video originally play and this error appears during successful playback? What are you using to watch the video? Is it the youtube site in the browser, the youtube app, or the video app?

Comment 9

[cbarker]

Correct, when the video is streamed on the Youtube site in the browser, it will play for a time then will error out and cease playing.  

The amount of time that the video plays seems to be random.

Comment 10

[Jason Smith [:jsmith]]

QAWanted request isn't met here. You need the following information:

1. A link to a youtube video demonstrating the problem
2. Logcat included in the bug when the issue reproduces

Comment 11

[cbarker]

Attachment: Log Cat for Youtube Error

Comment 12

[cbarker]

My Mistake, I have attached the log.

This is reproducible on the Inari.

As to the question of which video this occurs on, this error can occur on any video on Youtube, it isn't 100% however.

Comment 13

[Sotaro Ikeda [:sotaro]]

(In reply to cbarker from comment #12)
> My Mistake, I have attached the log.
> 
> This is reproducible on the Inari.

Is it reproducible also on partner built rom?

Comment 14

[Sotaro Ikeda [:sotaro]]

(In reply to Sotaro Ikeda [:sotaro] from comment #13)
> (In reply to cbarker from comment #12)
> > My Mistake, I have attached the log.
> > 
> > This is reproducible on the Inari.
> 
> Is it reproducible also on partner built rom?

Sorry, partner built ROM is only 1.0.1.

Comment 15

[Sotaro Ikeda [:sotaro]]

(In reply to cbarker from comment #11)
> Created attachment 761725 [details]
> Log Cat for Youtube Error

I checked logcat. There is no error around codec.

Comment 16

[Sotaro Ikeda [:sotaro]]

From the log, HW codec was shut down normally. Around the shut down log, wlan log "wlan0: CTRL-EVENT-CONNECTED - Connection" was printed. Network Error could be possible.

----------------------------------------

06-12 17:48:22.259: I/Adreno200-EGLSUB(110): <GetImageAttributes:1161>: RGBX_8888
06-12 17:48:22.259: I/wpa_supplicant(354): wlan0: Associated with 08:60:6e:22:ae:80
06-12 17:48:22.269: I/Adreno200-EGLSUB(110): <CreateImage:896>: Android Image
06-12 17:48:22.269: I/Adreno200-EGLSUB(110): <CreateImage:896>: Android Image
06-12 17:48:22.269: I/Adreno200-EGLSUB(110): <GetImageAttributes:1279>: YV12
06-12 17:48:22.299: I/wpa_supplicant(354): wlan0: WPA: Key negotiation completed with 08:60:6e:22:ae:80 [PTK=CCMP GTK=CCMP]
06-12 17:48:22.299: I/wpa_supplicant(354): wlan0: CTRL-EVENT-CONNECTED - Connection to 08:60:6e:22:ae:80 completed (reauth) [id=0 id_str=]
06-12 17:48:22.309: E/VDL_RTOS(116): Flush VDL_stats_q: stats_ptr 0xe23c98
06-12 17:48:22.309: E/omx_vdec(116): Warning - previous ts > current ts. And both are non B-frames
06-12 17:48:22.379: D/memalloc(110): /dev/pmem: Allocated buffer base:0x4a900000 size:614400 offset:3072000 fd:156
06-12 17:48:22.409: I/Adreno200-EGLSUB(110): <CreateImage:896>: Android Image
06-12 17:48:22.409: I/Adreno200-EGLSUB(110): <GetImageAttributes:1161>: RGBX_8888
06-12 17:48:22.409: I/Adreno200-EGLSUB(110): <CreateImage:896>: Android Image
06-12 17:48:22.409: I/Adreno200-EGLSUB(110): <GetImageAttributes:1105>: RGBA_8888
06-12 17:48:22.419: D/memalloc(110): /dev/pmem: Allocated buffer base:0x4a900000 size:614400 offset:0 fd:173
06-12 17:48:22.429: I/Adreno200-EGLSUB(110): <CreateImage:896>: Android Image
06-12 17:48:22.429: I/Adreno200-EGLSUB(110): <GetImageAttributes:1161>: RGBX_8888
06-12 17:48:22.529: I/Adreno200-EGLSUB(110): <CreateImage:896>: Android Image
06-12 17:48:22.529: I/Adreno200-EGLSUB(110): <GetImageAttributes:1161>: RGBX_8888
06-12 17:48:22.539: E/QC_BASE(116): Killing timer thread...
06-12 17:48:22.539: E/QC_BASE(116): Timer thread exited
06-12 17:48:22.549: E/QC_BASE(116):  DEINIT COMPLETED
06-12 17:48:22.549: E/QC_BASE(116): component_deinit deinit completed
06-12 17:48:22.549: E/QC_BASE(116): OMX AAC component destroyed

Comment 17

[cbarker]

For reference this occurred on WiFi, with a strong 5 bar connection.  

Not sure if this has anything to do with it, but no SIM Card is being detected so it is continuously attempting to detect in the background.

Comment 18

[Jason Smith [:jsmith]]

Is this potentially a dupe of bug 878981?

Comment 19

[Sotaro Ikeda [:sotaro]]

(In reply to Jason Smith [:jsmith] from comment #18)
> Is this potentially a dupe of bug 878981?

From attachment 761725 [details], no information about it.

Comment 20

[Sotaro Ikeda [:sotaro]]

Current gaia's video app implementation handle MEDIA_ERR_DECODE and MEDIA_ERR_NETWORK as same 'videoinvalid'. I feel two error needs to be different error message. I am going to create a new bug for it.

Comment 21

[Sotaro Ikeda [:sotaro]]

(In reply to Sotaro Ikeda [:sotaro] from comment #20)
> Current gaia's video app implementation handle MEDIA_ERR_DECODE and
> MEDIA_ERR_NETWORK as same 'videoinvalid'. I feel two error needs to be
> different error message. I am going to create a new bug for it.

Created Bug 882780.

Comment 22

[bhavana bajaj [:bajaj]]

:sotaro, can we DUP this and mrk it resolved given https://bugzilla.mozilla.org/show_bug.cgi?id=880571#c13 is now fixed?

Comment 23

[Sotaro Ikeda [:sotaro]]

(In reply to bhavana bajaj [:bajaj] from comment #22)
> :sotaro, can we DUP this and mrk it resolved given
> https://bugzilla.mozilla.org/show_bug.cgi?id=880571#c13 is now fixed?

It it not clear which error happened on this bug. Check by QA is necessary.

Comment 24

[Sotaro Ikeda [:sotaro]]

Need to check which error is shown by the rom including Bug 880571.

Comment 25

[Natalya Kot [:nkot]]

Attachment: logcat 6/25

Retested on 6/25 v1-train build (after the fix in bug 880571 has been uplifted)

Reproduced the bug following STR from comment 0 
 - YouTube video stops playing 
 - current error notification "A network error caused the video failed to download."

*attaching logcat for more info

Inari
Build  20130625070217
Gecko  http://hg.mozilla.org/releases/mozilla-b2g18/rev/29933d1937db
Gaia   1436e2778b90bd74635b0b94d1cf8ccb0d71b60c

Comment 26

[Natalya Kot [:nkot]]

Attachment: another logcat, 6/25

Comment 27

[Naoki Hirata :nhirata (please use needinfo instead of cc)]

I'm not sure if I am seeing this correctly, but it looks like the wpa_supplicant caused the network to disconnect looking at the logcat...

Comment 28

[Sotaro Ikeda [:sotaro]]

comment #25 and attachment 767441 [details], the problem happens by network/WLAN problem.
It is device and ROM specific problem. It is better to be analyzed by people related to WLAN.

Comment 29

[Sotaro Ikeda [:sotaro]]

It is device and ROM specific problem. this bug do not need to be leo+ bug, I think. Necessary fix is already done in bug 880571.

Comment 30

[Sotaro Ikeda [:sotaro]]

This is WLAN problem, I un-assign myself.

Comment 31

[Sotaro Ikeda [:sotaro]]

(In reply to nkot from comment #25)
> Created attachment 767356 [details]
> logcat 6/25
> 
> Retested on 6/25 v1-train build (after the fix in bug 880571 has been
> uplifted)
> 
> Reproduced the bug following STR from comment 0 
>  - YouTube video stops playing 
>  - current error notification "A network error caused the video failed to
> download."
> 
> *attaching logcat for more info
> 
> Inari
> Build  20130625070217
> Gecko  http://hg.mozilla.org/releases/mozilla-b2g18/rev/29933d1937db
> Gaia   1436e2778b90bd74635b0b94d1cf8ccb0d71b60c

Does the problem happens when using 3G data connection via RIL?

Comment 32

[Jason Smith [:jsmith]]

Renom per comment 29

Comment 33

[Naoki Hirata :nhirata (please use needinfo instead of cc)]

nkot has issues trying to get mobile data connection in the office and has asked me to take a look into this issue.

There's an overall bad reception on the device due to bug 862155 for her office.

Comment 34

[Naoki Hirata :nhirata (please use needinfo instead of cc)]

I am blocked from testing this because my inari device is not getting a mac address and cannot scan the network on 1.1; bug 875714 and bug 889623.

Comment 35

[Naoki Hirata :nhirata (please use needinfo instead of cc)]

Attachment: logcat

Fixed the issue with no MAC address for the Inari phone.
Video that shows is : "beauty and the briefcase full movie"

I have not gotten any errors.  The picture gets frozen in place and the audio continues to run after some point.

Comment 36

[Naoki Hirata :nhirata (please use needinfo instead of cc)]

ignore comment 35.  I tested on the wrong phone.  I need to test on the inari; accidentally tested on the buri.

Comment 37

[Naoki Hirata :nhirata (please use needinfo instead of cc)]

Attachment: logcat

I was not able to reproduce the wpa_supplicant issue; having said that I did end up crashing the whole OS trying to play a long video.

It first disconnected and then reboots the phone.  I replaced the previous logcat.

Comment 38

[Jason Smith [:jsmith]]

With the switch to YouTube without a UA override, this bug is no longer valid, as we do not render YouTube videos in the Gaia Video app with the override removed.

Comment 39

[Chris Kreinbring [:CKreinbring]]

Verifying as invalid.