Hmm, the bot sees a device in the right port:

$ cat /sys/bus/usb/devices/1-2.1/serial
HAD1DL5H

I'm pretty sure that was its serial num before it went MIA. Looking at the syslogs, I see "usb-storage 1-2.1:1.0: USB Mass Storage device detected" which sounds eerily similar to the problems I'm seeing in bug 780559. I was going to swing by the lab later; I'll take a look at the device when I'm in there.

Hmmm. Device appeared to have powered itself down (maybe some kind of hibernation?)

Had to press/hold power button on the visor.

List of devices attached
HAD1DL5H	device

I'm not aware of any special sleep/hibernation behavior on vega, but then I can't say I've explicitly tested for it, either.

Looks like there might be a temperature sensor or two that needs to be ignored. There are a number of sane-looking temps (20's and 30's), but a bunch in the 400's, with an average reported temperature in the 300's.

Nevermind, looks like the average temperature isn't what it actually uses to decide when to quarantine something. It eventually cooled down enough and picked up a task.

Haven't seen this go offline randomly since this was filed, although it does seem to be having temperature issues. It's been "In Maintenance" for over 12 hours due to temperatures https://chromium-swarm.appspot.com/bot?id=build370-a9&sort_stats=total%3Adesc

We can add a cooling tray fan to try and dissipate the heat. Unclear how well this will work on this platform, but worth a try.

It'd also be worth checking to make sure that the screen is actually off - it has to use a non-standard way of doing so since the screen is normally turned on/off with the proximity sensor, not a power button. It's possible it got stuck in an on state and is doing actual work while idle since it renders a 3D environment.

Right, that's my comment at https://bugs.chromium.org/p/chromium/issues/detail?id=889196#c14

Uploaded https://chrome-internal-review.googlesource.com/c/infradata/config/+/722450 a bit ago for the fix. Let me land that and see if it fixes our 🌡️ problem.

The following revision refers to this bug:
  https://chrome-internal.googlesource.com/infradata/config/+/bba417a5d3d9df77d0469036dd8beb599ba26314

commit bba417a5d3d9df77d0469036dd8beb599ba26314
Author: Ben Pastene <bpastene@chromium.org>
Date: Wed Dec 05 22:59:25 2018

The following revision refers to this bug:
  https://chrome-internal.googlesource.com/infradata/config/+/13721dc1fc86261b669901fb7d9b5b7801890d21

commit 13721dc1fc86261b669901fb7d9b5b7801890d21
Author: Ben Pastene <bpastene@chromium.org>
Date: Thu Dec 06 00:26:37 2018

The following revision refers to this bug:
  https://chrome-internal.googlesource.com/infradata/config/+/4c7e82833f2401af609d022aef129959feb178af

commit 4c7e82833f2401af609d022aef129959feb178af
Author: Ben Pastene <bpastene@chromium.org>
Date: Thu Dec 06 21:13:23 2018

Tentatively closing this as fixed after #12. Please reopen if you see it MIA again