This is a test that we run on Chromebox for Meetings (CfMs). We have limited it to run only in our out-of-band CfM suites by setting:

ATTRIBUTES = "suite:hotrod, suite:bluestreak-pre-cq"

Please help me understand why failures in this test is a problem that requires attention by the cros-sheriff. Our understanding is that failures here should only be our own problem.

Unfortunately I don't have a Chromium account so I can't see info from the link so if someone could share the information in there in a way that I can access it would be appreciated. (just requested a Chromium account but it takes a couple of days).

Hi Kristoffer, I think the CQ runs "suite:bluestreak-pre-cq". If that fails, the CQ fails. Please refer to the attached HWTest__bluestreak-pre-cq for a bit more details. Thanks.

Deleted: HWTest__bluestreak-pre-cq 38.9 KB

HWTest__bluestreak-pre-cq 38.9 KB View Download

bluestreak-pre-cq was setup as a "private pre-cq" for CfM developers. It contains tests that definitely are not up to CQ standards. It should _not_ run on the regular CQ. As far as I know, it has not been executed as a part of CQ earlier. Did something change? What is the best way to remove it from the CQ again?

This attached file has better context about the issue. The log showed that

File "/usr/local/telemetry/src/third_party/catapult/telemetry/telemetry/internal/image_processing/image_util_numpy_impl.py", line 22, in Width
  return image.shape[1]
AttributeError: 'NoneType' object has no attribute 'shape

Deleted: client.0.DEBUG 143 KB

client.0.DEBUG 143 KB Download

Hi Harpreet, as Christopher mentioned in C#4, bluestreak-pre-cq was meant to use for developers only. Could you help remove this test suite from HWTest pool in CQ? Thanks!

Looks like egemih@ added this to cq??? https://chromium-review.googlesource.com/c/chromiumos/chromite/+/862997

I have a fix the first error we see in https://crrev.com/c/1284589. However, that error just hides a Chrome crash for this test, so it still fails with what might actually be a real regression.

That said, even if it is a real regression this test is not supposed to be in CQ, so anyone who knows how to remove the bluestreak-pre-cq suite from CQ, please go ahead ASAP.

Looks like egemih@ is no longer on the team.

Bernie, you reviewed the CL listed above. Can you please take a look. Is reverting this CL all that needs to be done to remove it from cq?

That CL made it experimental, and it appears to still be experimental https://cs.corp.google.com/chromeos_public/chromite/config/chromeos_config.py?l=2517 so this should not be breaking the CQ. 

I defer to the CFM folks on what they want to do about this, but it should not be critical to the CrOS team that this is failing (still not optimal though). 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/third_party/autotest/+/0dd2630971ce107e24e9ecbc6996177d8133a85e

commit 0dd2630971ce107e24e9ecbc6996177d8133a85e
Author: Kristoffer Erlandsson <kerl@google.com>
Date: Thu Oct 18 13:16:25 2018

Catch all exceptions when taking screenshots.

TEST=Against lab machine
BUG=chromium:895221

Change-Id: I2bc756ee368b23dc2f903850bc7911d6694ead6d
Reviewed-on: https://chromium-review.googlesource.com/1284589
Commit-Ready: Kristoffer Erlandsson <kerl@google.com>
Tested-by: Kristoffer Erlandsson <kerl@google.com>
Reviewed-by: Denis Tosic <dtosic@google.com>

[modify] https://crrev.com/0dd2630971ce107e24e9ecbc6996177d8133a85e/client/common_lib/cros/test_webrtc_peer_connection.py