+sky@ - any chance you have some cycles to help look into this?

And just to clarify, these failures appear when applying this CL[1]. The CL does not break any guarantees made by Mojo bindings, but it can affect the timing of arbitrary events. The failures therefore strongly imply that incorrect ordering assumptions being made somewhere, and this has been the case for all other blocking bugs which have been fixed so far.

[1] https://chromium-review.googlesource.com/c/chromium/src/+/1145692

I will try to look today.

I presume the test is timing out? Any idea where it is timing out? Do you know if you had https://chromium-review.googlesource.com/c/chromium/src/+/1161171 applied (landed yesterday morning).

Thanks. The failure I have on record is from this try run[1], which pulled
revision bc9d7d70[2], so I am not sure if that linked CL would be relevant.

The actual failure is in [3] and doesn't look like a timeout.

[1]
https://ci.chromium.org/p/chromium/builders/luci.chromium.try/linux-chromeos-rel/64188
[2] https://crrev.com/bc9d7d70
[3]
https://logs.chromium.org/logs/chromium/buildbucket/cr-buildbucket.appspot.com/8938897002999411472/+/steps/ash_unittests__with_patch_/0/logs/AshServiceTest.OpenWindow/0

I ran out of time today. Will try to look tomorrow. Ken, feel free to disable the test if needed to land your change.

Thanks, no worries. There are enough other blockers outstanding that any
single one isn't that urgent. I'll disable this if it's still not fixed by
the time the other bugs are.

I have a (short term) patch out that should fix this.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/79be6d9228e61d0db480220cc05f4cfa8c79d47d

commit 79be6d9228e61d0db480220cc05f4cfa8c79d47d
Author: Scott Violet <sky@chromium.org>
Date: Thu Aug 09 22:28:07 2018

chromeos: fix ash unittest for changes to mojo

This test is creating an aura::Window and then embedding another client in
it. When this happens it is expected that clients set the type before doing
this.

I'm sending this to make sure we don't block some work being done to mojo. I
think the real fix is to remove EMBED_IN_OWNER (there is a bug filed on it).

BUG= 872071 
TEST=test only change

Change-Id: I6e0f6c5584ea9ec19b0e09fe944a83187a9f5ebf
Reviewed-on: https://chromium-review.googlesource.com/1169695
Reviewed-by: Xiyuan Xia <xiyuan@chromium.org>
Commit-Queue: Scott Violet <sky@chromium.org>
Cr-Commit-Position: refs/heads/master@{#581922}
[modify] https://crrev.com/79be6d9228e61d0db480220cc05f4cfa8c79d47d/ash/ash_service_unittest.cc