Kalin, could you help to check on one of these devices whether this is camera HW related issue?

Richard, how does the lab team usually process DUT related issues? Decommission the DUTs?

Can we have the source of this report - where is it coming from? How is camera being detected in lab? Is this for auto-labeling purposes? Were these boards just deployed? Were they failing camera detection before?

+nxia as deputy

kevcheng@, can you comment on #2? should we replace the dut, or is there way to remove the camera HW label?

I'd like to see the answers to Kalin's questions. Specifically:
- How is the internal camera detected?  
- How do we know if it's a sw or hw issue?

We run avtest_label_detect on all DUTs and found there is no camera on those DUTs.(But there should be an internal one.)
I originally thought it is a hardware issue, but I am not so sure now. Today, I rerun avtest_label_detect and found this issue is gone on some of the machines.

[flaky]
    chromeos4-row9-rack7-host1: SKATE B2A-A3C-I7M 9125.0.0-rc1
    chromeos4-row9-rack7-host11: SKATE B2A-A3C-I7M 9125.0.0-rc1

Back to the original issue, take the following two springs as an example:

    chromeos6-row2-rack16-host20
    chromeos6-row2-rack16-host2

Both have the same HWID "SPRING G9A-D3I-A2A. But there is no "/dev/video0" on chromeos6-row2-rack16-host2, 
so avtest_label_detect said there is no webcam on this device.

It sounds like a potential sw issue.

Owen, are you the right person to dive deeper into this issue?

OK. Will try to see if there is any error logs.