New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.

Issue 906626 link

Starred by 1 user

Issue metadata

Status: Fixed
Owner:
Closed: Nov 19
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Android
Pri: 3
Type: Bug



Sign in to add a comment

GPU FYI Nexus 9 devices are offline

Project Member Reported by geoffl...@chromium.org, Nov 19

Issue description

The following devices have gone offline:
build127-b4--device1
build127-b4--device2
build127-b4--device3
build127-b4--device4
build127-b4--device5
build127-b4--device6
build127-b4--device7
https://chromium-swarm.appspot.com/botlist?f=device_os%3AM&f=os%3AAndroid&f=pool%3AChrome-GPU&f=device_type%3Aflounder

This causes the Android FYI Release (Nexus 9) ( https://ci.chromium.org/p/chromium/builders/luci.chromium.ci/Android%20FYI%20Release%20(Nexus%209) ) bot to fail all its jobs.


 
Owner: jo...@chromium.org
Status: Assigned (was: Untriaged)
These are very old devices.  Any chance you would consider upgrading to something newer?

Passing this on to johnw to a look at this.
Cc: ynovikov@chromium.org kbr@chromium.org
CCing others who are more familiar with GPU Android testing.

Have the Nexus 9s been very troublesome in the past?  Should we keep supporting this one-off config?

Either way, it looks like the host for these devices went down somehow and I hope we can get them running again in the meantime.
Since Nexus 9 can support Nougat, and Chromium is in the process of dropping support for Jelly Bean now, I'd expect that we'll need to test on these devices for 4 more years - my guess for when Chromium deprecates Nougat.

These bots vere never troublesome in the past, and looks like the issue is not the devices, but the blade server they are connected to.

One more candidate to switch to Skia's RPIs :)
Looks like there was some sort of BIOS level EFI/USB error preventing bootup. A hard remote-powercycle seems to have un-wedged it. 

Waiting to see if the devices recover or if they need any interaction. 
Labels: -Pri-1 Pri-3
These are all back with the exception of --device7 which looks like it has a low voltage condition. Will inspect in person later today.
note: ditto for --device3
These tablets are all using older/unreliable usb interconnects. Going to swap them out for more reliable ones. 
Status: Fixed (was: Assigned)
Charging looks better on these now. Hopefully the new interconnects improve overall stability.
Sorry for the late reply but thanks for the really fast attention this got!

Sign in to add a comment