Out of curiosity, why is this bug assigned to johnme@?

I took a look at global-interface-listing and debugger-proto-property and couldn't find any recent changes to cause the flakiness. I can create bugs to mark the tests as flaky. I haven't looked at the others yet. 

johnme@ has been making changes in notification-properties.html. I readily admit it's almost certainly not his fault, but the bug needs to be assigned to somebody, and perhaps johnme@ can find a better owner for this.
I am not sure if marking all these tests as flaky is the right approach - it seems like there's a LOT of tests that are flaking, so I am wondering if there is a deeper problem here.

I'm going to duplicate  Issue 649792  into this. From that bug report:

These test flakes are blocking the CQ. Upgrading to P1.

https://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_rel_ng?numbuilds=200

Example:

https://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_rel_ng/builds/303239

Failed:

fast/frames/frame-limit.html
fast/frames/cached-frame-counter.html
virtual/gpu-rasterization/fast/images/color-profile-background-clip-text.html

without_patch failed:

unexpected_failures:
fast/frames/cached-frame-counter.html
virtual/gpu-rasterization/fast/images/color-profile-background-clip-text.html

so:

fast/frames/frame-limit.html

failed.

It looks like the test timed out randomly 3 times in a row.

 Issue 649792  has been merged into this issue.

Sheriff-o-matic isn't seeing any issues with the first two tests, so I'm not sure if they aren't flaking often enough or what. I'm currently investigating color-profile-background-clip-text failure. It's being tracked in a separate bug (649589)

Adding next shift's sheriffs

re comment 8: From my experience, sheriff-o-matic isn't particularly great at identifying flakes that are not happening at the moment. If you look at the build history though, you can see plenty of red https://uberchromegw.corp.google.com/i/chromium.webkit/builders/WebKit%20Win7%20%28dbg%29?numbuilds=200

+Eric, could be related to a change he's reverting, https://codereview.chromium.org/2366033002/

At least some of these flakes (including the ones linked in #10, which seem 100% consistent), started before the patch mentioned in #11 - so likely unrelated

On further inspection, I think my patch is responsible for some of these flakes, but not all of them. My revert is in, but probably worth continuing to investigate the Win7 flakes which started with this build: https://uberchromegw.corp.google.com/i/chromium.webkit/builders/WebKit%20Win7%20%28dbg%29/builds/7442

fast/backgrounds/border-radius-split-background-image.html
fast/backgrounds/border-radius-split-background.html
fast/borders/border-styles-split.html

From looking at these flakes, I think we just missed/need a Win7 rebaseline - the changes seem pretty harmless.

From looking at the change log, my guess is that these were introduced by:
https://chromium.googlesource.com/skia.git/+/bac104605ef3d9a8ed0022694990f00518b809e9

It seems like suppressions were landed for LayoutTests here: https://codereview.chromium.org/2360393002, but Win7 specific issues may have been missed.

Assigning to caryclark who checked in that CL.

Those test failures are being tracked in a separate bug (https://bugs.chromium.org/p/chromium/issues/detail?id=649631#c4). After some back and forth we ended up just disabling them, but I can imagine that they needed to be rebased instead.

I think that the particular tests that this bug was opened for no longer fail, so I'm going to mark this bug as fixed.

debugger-proto-property.html just failed in the most recent run: https://uberchromegw.corp.google.com/i/chromium.webkit/builders/WebKit%20Win7%20%28dbg%29/builds/7458

It seems to me that consistent failures of the three tests from  issue 649631 , and the occasional flakes of a bunch of tests described in this bug are two separate issues.

I think this issue should still be looked at, and we should either disable all the flaking tests, or fix the root causes that's causing the flake (if any).

Let me remove Sheriff-Chromium label since every bot does not look flaky as of now.

Now I don't observe flakiness of any HTML-related tests.

I can't access https://build.chromium.org/p/tryserver.chromium.linux/builders/linux_chromium_rel_ng to confirm whether this seems to be affecting the CQ. If it isn't then perhaps this should just be closed.

There is some flakiness of some layout tests still observable on https://uberchromegw.corp.google.com/i/chromium.webkit/builders/WebKit%20Win7%20%28dbg%29?numbuilds=200 .
virtual/threaded/animations/3d/transform-origin-vs-functions.html
seems to show up often, but there are a few other flaky tests observed there.