O_O

Thanks for filing.

... back to untriaged to make sure I>C>A sees it.

This again? >.>

Ok, looking.

Oh, this bot's not swarmed. That's going to make it hard to reproduce the failures.

I think I'll just swarm it.

We still apparently have an issue w/ how non-swarmed provisioning keeps the screen on, which, given the similarities in the two, means we may also have issues in the swarming logic.

... alternatively, was this something that got fixed in swarming provisioning that didn't make it back to non-swarmed provisioning?

Given that it was transient, I'd still like to swarm it. If something like this happens again, it will be much easier to debug when we have the test isolates at our hands. (That way I can run the exact same code on locked vs unlocked screens)

The following revision refers to this bug:
  https://chrome-internal.googlesource.com/infra/infra_internal.git/+/18ae4aea479d62d246e41741fb4fbfc67d73770e

commit 18ae4aea479d62d246e41741fb4fbfc67d73770e
Author: bpastene <bpastene@google.com>
Date: Thu Oct 27 23:55:49 2016

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/e3fdf2ff9602ccee1b1d269c71c87f1fbb2a709c

commit e3fdf2ff9602ccee1b1d269c71c87f1fbb2a709c
Author: bpastene <bpastene@chromium.org>
Date: Fri Oct 28 21:06:14 2016

chromium.linux: Swarm the tests on Android Tests (dbg)

Will swap out the slave with a VM and throw its devices into the pool
once everything's stable.

BUG= 659869 

Review-Url: https://codereview.chromium.org/2456873003
Cr-Commit-Position: refs/heads/master@{#428489}

[modify] https://crrev.com/e3fdf2ff9602ccee1b1d269c71c87f1fbb2a709c/testing/buildbot/chromium.linux.json

Many of the tests in ContentViewCoreSelectionTest would also fail if the device's screen was not on. Should an assertScreenIsOn also be added to the setup of those tests?

Yes, it seems worth adding that assertion there too.  It will help diagnose the cause of flakiness.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/73584a85ececf0f877630c158fb90f1fa7a2cdc3

commit 73584a85ececf0f877630c158fb90f1fa7a2cdc3
Author: amaralp <amaralp@chromium.org>
Date: Tue Nov 01 18:34:36 2016

Asserting screen is on for all content shell tests

Many ContentShellTests will fail if the screen is not on.
This assert helps diagnose those failures.

BUG= 659869 

Review-Url: https://codereview.chromium.org/2465053002
Cr-Commit-Position: refs/heads/master@{#429061}

[modify] https://crrev.com/73584a85ececf0f877630c158fb90f1fa7a2cdc3/content/public/android/javatests/src/org/chromium/content/browser/input/ImeTest.java
[modify] https://crrev.com/73584a85ececf0f877630c158fb90f1fa7a2cdc3/content/shell/android/javatests/src/org/chromium/content_shell_apk/ContentShellTestBase.java

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/tools/build.git/+/a04a87716ec6eb0f2d4427b63ab97708534c0673

commit a04a87716ec6eb0f2d4427b63ab97708534c0673
Author: bpastene <bpastene@chromium.org>
Date: Mon Nov 07 18:15:53 2016

Move Android Tests (dbg) to a ccompute bot.

BUG= 659869 

Review-Url: https://codereview.chromium.org/2457973002

[modify] https://crrev.com/a04a87716ec6eb0f2d4427b63ab97708534c0673/masters/master.chromium.linux/slaves.cfg

The following revision refers to this bug:
  https://chrome-internal.googlesource.com/infradata/master-manager.git/+/52832963f23b061bf24bfaaef89f836edd920b24

commit 52832963f23b061bf24bfaaef89f836edd920b24
Author: bpastene <bpastene@google.com>
Date: Mon Nov 07 18:28:52 2016

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/tools/build.git/+/7f40ab150bc694d86c57e3e08d38441f8fd5c59c

commit 7f40ab150bc694d86c57e3e08d38441f8fd5c59c
Author: sergeyberezin <sergeyberezin@chromium.org>
Date: Mon Nov 07 19:47:04 2016

Revert of Move Android Tests (dbg) to a ccompute bot. (patchset #1 id:1 of https://codereview.chromium.org/2457973002/ )

Reason for revert:
Left the builder offline - needs to be landed closer to the master restart.

Original issue's description:
> Move Android Tests (dbg) to a ccompute bot.
>
> BUG= 659869 
>
> Committed: https://chromium.googlesource.com/chromium/tools/build/+/a04a87716ec6eb0f2d4427b63ab97708534c0673

TBR=jbudorick@chromium.org,stip@chromium.org,bpastene@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG= 659869 

Review-Url: https://codereview.chromium.org/2483933002

[modify] https://crrev.com/7f40ab150bc694d86c57e3e08d38441f8fd5c59c/masters/master.chromium.linux/slaves.cfg

The following revision refers to this bug:
  https://chrome-internal.googlesource.com/infradata/master-manager.git/+/76d6679002cc1ce11fef5cc9c99964eadf86ae49

commit 76d6679002cc1ce11fef5cc9c99964eadf86ae49
Author: bpastene <bpastene@google.com>
Date: Wed Nov 09 01:56:35 2016

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/tools/build.git/+/f15c7493d34d1a3e35696de75d37855449e9d267

commit f15c7493d34d1a3e35696de75d37855449e9d267
Author: bpastene <bpastene@chromium.org>
Date: Wed Nov 09 01:59:53 2016

Reland of "Move Android Tests (dbg) to a ccompute bot."

Original: https://codereview.chromium.org/2457973002/

NOPRESUBMIT=true
NOTRY=true
TBR=stip@chromium.org
BUG= 659869 

Review-Url: https://codereview.chromium.org/2480293007

[modify] https://crrev.com/f15c7493d34d1a3e35696de75d37855449e9d267/masters/master.chromium.linux/slaves.cfg

FYI, this bug is still there. Here are 2 failures on Dec. 10th:

https://build.chromium.org/p/chromium.android/builders/Lollipop%20Tablet%20Tester/builds/6110
https://build.chromium.org/p/chromium.android/builders/Android%20N5X%20Swarm%20Builder/builds/7103

Any update on this? We are trying to deflake IME tests, and this issue prevents us from identifying other issues (we need to click on individual failures on dashboard to understand the failures, and this is now becoming a major source.)

We had a bit of an outage the other week with our testing devices. It's since been resolved and everything, including their screens, should be on and working fine. Are these types of failures still happening? Can you point me to an example?

https://build.chromium.org/p/chromium.android/builders/Android%20N5X%20Swarm%20Builder?numbuilds=200 seems pretty green and doesn't seem to be having device issues.

Here are 2 examples:

ImeTest#testDeleteSurroundingTextWithRangeSelection and other tests failed on Dec 22nd:
https://build.chromium.org/p/chromium.android/builders/Lollipop%20Tablet%20Tester/builds/6264

Failed on Dec 17th:
https://build.chromium.org/p/chromium.android/builders/Lollipop%20Tablet%20Tester/builds/6208

re #6208

It looks as though the problem is with one device, 0af6646d. The log for provision_devices.py includes some interesting errors for that device:

E  413.751s ProvisionDevice(0af6646d)  Failed to configure local properties.
...
W  413.857s ProvisionDevice(0af6646d)  Unable to find database file to set lock screen settings.

Those are:
 - https://codesearch.chromium.org/chromium/src/build/android/provision_devices.py?rcl=0&l=349
 - https://codesearch.chromium.org/chromium/src/build/android/pylib/device_settings.py?rcl=0&l=81

respectively. The latter is particularly interesting given that it's trying to set lockscreen.disabled in one of a couple of sqlite tables.

re #6264

Again, one device, 0a084b87. It had the same issues in provisioning.

I don't think this is a problem anymore?