Looks like the bots are actually Win8, so this may be related to https://codereview.chromium.org/2060043002 . I'm already reverting that due to another issue.

Whhops, looks like these weren't fixed, and are timing out, causing whole the whole WebGL 2 step to fail:

https://build.chromium.org/p/chromium.gpu.fyi/builders/Win7%20Release%20%28NVIDIA%29/builds/25526/steps/webgl2_conformance_tests%20on%20NVIDIA%20GPU%20on%20Windows%20on%20Windows-2008ServerR2-SP1/logs/stdio

https://build.chromium.org/p/chromium.gpu.fyi/builders/Win7%20Release%20%28ATI%29/builds/21515/steps/webgl2_conformance_tests%20on%20ATI%20GPU%20on%20Windows%20on%20Windows-2008ServerR2-SP1/logs/stdio

[ RUN      ] WebglConformance.conformance2_textures_image_bitmap_from_video_tex_3d_rgba32f_rgba_float
(INFO) 2016-06-15 02:08:01,369 cache_temperature.EnsurePageCacheTemperature:55  PageCacheTemperature: any
[2668:2844:0615/020801:WARNING:webmediaplayer_impl.cc(335)] Using MultibufferDataSource
Traceback (most recent call last):
  File "c:\users\chrome~1\appdata\local\temp\isolated_runyhpqnr\third_party\catapult\telemetry\telemetry\internal\story_runner.py", line 84, in _RunStoryAndProcessErrorIfNeeded
    state.RunStory(results)
  File "c:\users\chrome~1\appdata\local\temp\isolated_runyhpqnr\content\test\gpu\gpu_tests\gpu_test_base.py", line 122, in RunStory
    RunStoryWithRetries(DesktopGpuSharedPageState, self, results)
  File "c:\users\chrome~1\appdata\local\temp\isolated_runyhpqnr\content\test\gpu\gpu_tests\gpu_test_base.py", line 72, in RunStoryWithRetries
    super(cls, shared_page_state).RunStory(results)
  File "c:\users\chrome~1\appdata\local\temp\isolated_runyhpqnr\third_party\catapult\telemetry\telemetry\page\shared_page_state.py", line 306, in RunStory
    self._current_page.Run(self)
  File "c:\users\chrome~1\appdata\local\temp\isolated_runyhpqnr\third_party\catapult\telemetry\telemetry\page\__init__.py", line 95, in Run
    shared_state.page_test.RunNavigateSteps(self, current_tab)
  File "c:\users\chrome~1\appdata\local\temp\isolated_runyhpqnr\third_party\catapult\telemetry\telemetry\page\legacy_page_test.py", line 191, in RunNavigateSteps
    page.RunNavigateSteps(action_runner)
  File "c:\users\chrome~1\appdata\local\temp\isolated_runyhpqnr\content\test\gpu\gpu_tests\webgl_conformance.py", line 192, in RunNavigateSteps
    'webglTestHarness._finished', timeout_in_seconds=300)
  File "c:\users\chrome~1\appdata\local\temp\isolated_runyhpqnr\third_party\catapult\telemetry\telemetry\internal\actions\action_runner.py", line 186, in WaitForJavaScriptCondition
    self._tab.WaitForJavaScriptExpression(condition, timeout_in_seconds)
  File "c:\users\chrome~1\appdata\local\temp\isolated_runyhpqnr\third_party\catapult\telemetry\telemetry\internal\browser\web_contents.py", line 136, in WaitForJavaScriptExpression
    e.message + '\n' + debug_message)
TimeoutException: Timed out while waiting 300s for IsJavaScriptExpressionTrue.
Console output:

John can help help here?

Win7 Release NVIDIA hasn't flaked since Jun 15th. I don't know why Win7 Release ATI is still flaking - it seems to have been happening since June 10th.

https://build.chromium.org/p/chromium.gpu.fyi/builders/Win7%20Release%20%28NVIDIA%29/builds/25526 failed because the WebGL 2.0 tests took more than 1 hour to run. Sharding these ( Issue 352807 ) will fix this.

The only recent flake I see in the WebGL 2.0 conformance tests on this bot was WebglConformance.deqp_functional_gles3_shadertexturefunction_textureoffset from https://build.chromium.org/p/chromium.gpu.fyi/builders/Win7%20Release%20%28NVIDIA%29/builds/25654 . Maybe this has been fixed?

A bunch more of these timeouts today. The log you linked in #c5, it said the video test timed out. Is this expected when the harness itself breaks 1hr? Just wanted to check that there wasn't a regression in speed for these tests that were causing them to take 5min+.

eg recent failure today:
https://build.chromium.org/p/chromium.gpu.fyi/builders/Win7%20Release%20%28NVIDIA%29/builds/25673/steps/webgl2_conformance_tests%20on%20NVIDIA%20GPU%20on%20Windows%20on%20Windows-2008ServerR2-SP1/logs/stdio

Yes, the problem's that the overall run is taking 1 hour. It can be seen by going to the bottom of the log and looking for "Duration: 3600.0s". If that duration is hit the step is going to fail.

The sharding is being put in place in https://codereview.chromium.org/2077443002 . Hopefully it will land today.

Seems like this is fixed by sharding?

Hopefully. Let's duplicate this into the other bug and reopen if it's seen again.