Issue metadata
Sign in to add a comment
|
smoothness.tough_webgl_cases failing on chromium.perf/Mac Air 10.11 Perf |
||||||||||||||||||||||
Issue descriptionsmoothness.tough_webgl_cases failing on chromium.perf/Mac Air 10.11 Perf Regression ranges: chromium: 468663 - 468785 Builders failed on: - Mac Air 10.11 Perf: https://build.chromium.org/p/chromium.perf/builders/Mac%20Air%2010.11%20Perf
,
May 4 2017
It seems to be passing again here: https://build.chromium.org/p/chromium.perf/builders/Mac%20Air%2010.11%20Perf/builds/795 I think something was going wrong with Mac Air 10.11 Perf and it's since been resolved I'm gonna go ahead and mark this obsolete, please reopen if you feel there's more work that can be done
,
May 4 2017
I'm going to take this opportunity to add retries to the browser bringup sequence in shared_page_state.py.
,
May 4 2017
Ken: if desktop browser startup is deemed to be unreliable. I think we probably want to add retries in a deeper level, probably in DesktopBrowserBackend? (https://github.com/catapult-project/catapult/blob/master/telemetry/telemetry/internal/backends/chrome/desktop_browser_backend.py#L303)
,
May 4 2017
In https://cs.chromium.org/chromium/src/content/test/gpu/gpu_tests/gpu_integration_test.py?q=gpu_integration_test.py+package:%5Echromium$&dr&l=25 we found it necessary to try fetching the first tab in order to determine whether the launched browser was alive, and to call Browser.Close() if launching failed. The desktop_browser_backend will be too low a level to determine whether the launched browser is actually viable. I would ideally like to do this in the Browser class. It would have been ideal if starting the browser were separate from Browser.__init__. Looking into refactoring this.
,
May 4 2017
How about we do it in PossibleDesktopBrowser.Create()? https://github.com/catapult-project/catapult/blob/60a406396e81bc09500e854916c51657972ad033/telemetry/telemetry/internal/backends/chrome/desktop_browser_finder.py#L66
,
May 7 2017
,
May 7 2017
Thanks, Ned, good suggestion. Up for review: https://codereview.chromium.org/2868703002/
,
May 9 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/cffa5eef397958bc2fbc0500af130ebecd07a8c9 commit cffa5eef397958bc2fbc0500af130ebecd07a8c9 Author: catapult-deps-roller@chromium.org <catapult-deps-roller@chromium.org> Date: Tue May 09 08:59:00 2017 Roll src/third_party/catapult/ a263e5566..b8a441dac (4 commits) https://chromium.googlesource.com/external/github.com/catapult-project/catapult.git/+log/a263e5566dcc..b8a441daccc8 $ git log a263e5566..b8a441dac --date=short --no-merges --format='%ad %ae %s' 2017-05-08 bpastene devil: Add devil to path when running find_usb_devices.py standalone. 2017-05-08 dtu [pinpoint] Make Change.patch more structured. 2017-05-08 kbr Retry browser startup in desktop_browser_finder. 2017-05-08 benjhayden Use the same colors for dashboard graphs. Created with: roll-dep src/third_party/catapult BUG= 718300 Documentation for the AutoRoller is here: https://skia.googlesource.com/buildbot/+/master/autoroll/README.md If the roll is causing failures, see: http://www.chromium.org/developers/tree-sheriffs/sheriff-details-chromium#TOC-Failures-due-to-DEPS-rolls CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.android:android_optional_gpu_tests_rel TBR=sullivan@chromium.org Change-Id: Ibd3c820f751661198724c7483b52cb4e9f105a2d Reviewed-on: https://chromium-review.googlesource.com/499429 Reviewed-by: <catapult-deps-roller@chromium.org> Commit-Queue: <catapult-deps-roller@chromium.org> Cr-Commit-Position: refs/heads/master@{#470267} [modify] https://crrev.com/cffa5eef397958bc2fbc0500af130ebecd07a8c9/DEPS
,
May 16 2017
,
May 17 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/b1226755d23b14fe273acd625806455a0a00e199 commit b1226755d23b14fe273acd625806455a0a00e199 Author: kbr <kbr@chromium.org> Date: Wed May 17 05:05:39 2017 Clean up logging around browser relaunch in gpu_integration_test.py. Though this was mostly subsumed into Telemetry's desktop_browser_finder, the gpu_integration_test_unittests in particular demonstrate that a retry loop is still needed here in some situations. BUG= 718300 TBR=nednguyen@google.com CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.android:android_optional_gpu_tests_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel Review-Url: https://codereview.chromium.org/2888713004 Cr-Commit-Position: refs/heads/master@{#472334} [modify] https://crrev.com/b1226755d23b14fe273acd625806455a0a00e199/content/test/gpu/gpu_tests/gpu_integration_test.py
,
May 17 2017
|
|||||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||||
Comment 1 by kbr@chromium.org
, May 4 2017Components: Tests>Telemetry
Labels: OS-Mac Pri-1 Type-Bug-Regression
Owner: kbr@chromium.org
Status: Assigned (was: Available)