WebRtcWebcamBrowserTest.MANUAL_TestAcquiringAndReacquiringWebcam frequently timing out in Linux Tester |
||||||
Issue descriptionLast green run: https://build.chromium.org/p/chromium.webrtc/builders/Linux%20Tester/builds/17601 First failing run: https://build.chromium.org/p/chromium.webrtc/builders/Linux%20Tester/builds/17602 I don't see any changes from the WebRTC repo between those runs. On the Chrome side, I see r396818, but I'm not 100% certain that is to blame. Still, it's the closest thing I have to a clue. :) grunell@ (CL author) and/or guidou@ (current sheriff), can you take a look?
,
Jun 1 2016
,
Jun 1 2016
The bot failed after the revert, so I relanded the CLs. Perhaps this is a hardware failure of the webcam in the bot? Note that no other bot has exhibited this failure so far.
,
Jun 3 2016
kjellander@ - Do you think we should we route this to the lab folks? https://build.chromium.org/p/chromium.webrtc/builders/Linux%20Tester?numbuilds=200 is still mostly red.
,
Jun 6 2016
Yes, I believe this must be hardware related. Especially since it doesn't seem to affect the almost identical (except that WebRTC it ToT revision on it): https://build.chromium.org/p/chromium.webrtc.fyi/builders/Linux%20Tester Labs team: can you look into replacing the webcam on build50-b1?
,
Jun 6 2016
,
Jun 6 2016
,
Jun 7 2016
I think I have found 2 spare webcams. Anyway I power cycled build50-b1 (sometimes this does wonders). It's effective with https://build.chromium.org/p/chromium.webrtc/builders/Linux%20Tester/builds/17790 I'll replace it if badness returns.
,
Jun 8 2016
The bot is very green now :) Thanks Peter. I'm curious about you mentioning power cycling the machine - I thought all our bots were rebooting between builds, so what difference could that do?
,
Jun 8 2016
I find the best way to quiesce hardware is to pull power. In retrospect what I should have done first is just reset the camera. |
||||||
►
Sign in to add a comment |
||||||
Comment 1 by guidou@chromium.org
, Jun 1 2016