ExecuteScriptApiTest/DestructiveScriptTest.MacrotaskRemoval/0 flaky |
|||||||
Issue descriptionExecuteScriptApiTest/DestructiveScriptTest.MacrotaskRemoval/0 is very flaky on release builders. 1) Go to https://build.chromium.org/f/chromium/flakiness/ 2) Search for ExecuteScriptApiTest/DestructiveScriptTest.MacrotaskRemoval 3) Look at linux_chromium_rel_ng Here is a sample output: [20226:20226:0515/211511.766899:FATAL:optional.h(219)] Check failed: !storage_.is_null_. #0 0x0000028dfaec base::debug::StackTrace::StackTrace() #1 0x0000028f94c1 logging::LogMessage::~LogMessage() #2 0x0000008cf452 _ZNKR4base8OptionalIN7content8Manifest11ShareTargetEE5valueEv #3 0x000002a622b5 CorePageLoadMetricsObserver::OnFirstContentfulPaintInPage() #4 0x000002a68c02 page_load_metrics::(anonymous namespace)::DispatchObserverTimingCallbacks() #5 0x000002a689a5 page_load_metrics::PageLoadTracker::UpdateSubFrameTiming() #6 0x000002a5f4de page_load_metrics::MetricsWebContentsObserver::OnTimingUpdated() #7 0x000002a5f33d _ZN3IPC8MessageTI37PageLoadMetricsMsg_TimingUpdated_MetaSt5tupleIJN17page_load_metrics14PageLoadTimingENS3_16PageLoadMetadataEEEvE8DispatchINS3_26MetricsWebContentsObserverES9_N7content15RenderFrameHostEMS9_FvPSB_RKS4_RKS5_EEEbPKNS_7MessageEPT_PT0_PT1_T2_ #8 0x000002a5f244 page_load_metrics::MetricsWebContentsObserver::OnMessageReceived() #9 0x0000019ab738 content::WebContentsImpl::OnMessageReceived() #10 0x000001787f3f content::RenderFrameHostImpl::OnMessageReceived() #11 0x00000382c07c IPC::ChannelProxy::Context::OnDispatchMessage() #12 0x0000005d3d1e _ZNO4base8CallbackIFvvELNS_8internal8CopyModeE1ELNS2_10RepeatModeE1EE3RunEv #13 0x000002997ec4 base::debug::TaskAnnotator::RunTask() #14 0x0000029023c9 base::MessageLoop::RunTask() #15 0x00000290264f base::MessageLoop::DeferOrRunPendingTask() #16 0x000002902a3d base::MessageLoop::DoWork() #17 0x0000029046e9 base::MessagePumpGlib::Run() #18 0x000002902152 base::MessageLoop::RunHandler() #19 0x00000292dfd9 base::RunLoop::Run() #20 0x000002e5411b content::RunThisRunLoop() #21 0x000005e7f755 extensions::ResultCatcher::GetNextResult() #22 0x000000c99ffa ExtensionApiTest::RunExtensionTestImpl() #23 0x000000c9a2cf ExtensionApiTest::RunExtensionSubtest() #24 0x0000008058b5 DestructiveScriptTest::RunSubtest() #25 0x000000805bc2 DestructiveScriptTest_MacrotaskRemoval_Test::RunTestOnMainThread() #26 0x000002e312d5 content::BrowserTestBase::ProxyRunTestOnMainThreadLoop() #27 0x0000029eda72 ChromeBrowserMainParts::PreMainMessageLoopRunImpl() #28 0x0000029ecc06 ChromeBrowserMainParts::PreMainMessageLoopRun() #29 0x00000104589a chromeos::ChromeBrowserMainPartsChromeos::PreMainMessageLoopRun() #30 0x0000016b6a71 content::BrowserMainLoop::PreMainMessageLoopRun() #31 0x00000198ca3a content::StartupTaskRunner::RunAllTasksNow() #32 0x0000016b51e7 content::BrowserMainLoop::CreateStartupTasks() #33 0x0000016b9313 content::BrowserMainRunnerImpl::Initialize() #34 0x0000016b2cb8 content::BrowserMain() #35 0x00000289844f content::ContentMainRunnerImpl::Run() #36 0x000004131437 service_manager::Main() #37 0x0000028973ef content::ContentMain() #38 0x000002e31064 content::BrowserTestBase::SetUp() #39 0x0000029c3698 InProcessBrowserTest::SetUp() #40 0x0000012553ed testing::Test::Run() #41 0x000001255b13 testing::TestInfo::Run() #42 0x000001255f32 testing::TestCase::Run() #43 0x000001259c26 testing::internal::UnitTestImpl::RunAllTests() #44 0x000001259921 testing::UnitTest::Run() #45 0x0000029d1e41 base::TestSuite::Run() #46 0x0000028d4243 ChromeTestSuiteRunner::RunTestSuite() #47 0x000002e52200 content::LaunchTests() #48 0x0000028d283f main #49 0x7fd7d1eaaf45 __libc_start_main #50 0x000000579b49 <unknown> [ RUN ] ExecuteScriptApiTest/DestructiveScriptTest.MacrotaskRemoval/0
,
May 16 2017
The first failure is from a try job on https://codereview.chromium.org/2872793002. Though it seems like a stretch, the CL does modify some page_load_metrics test expectations, and the failures for this browser test are consistently hitting some DCHECK in page_load_metrics IPC handling. Tentatively assigning to zhenw@ to investigate.
,
May 16 2017
(You may want to try reverting that CL to see if the flake clears, given the P0)
,
May 16 2017
+ emea sheriff. zhenw@ is MTV timezone, maybe we can get a faster solution to unblock EMEA?
,
May 16 2017
I will take a look. Thanks for pointing out.
,
May 16 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/28e83a520012c28bb95b0b41cac7a03071001a4f commit 28e83a520012c28bb95b0b41cac7a03071001a4f Author: kolos <kolos@chromium.org> Date: Tue May 16 14:51:44 2017 Disable flaky DestructiveScriptTest.MacrotaskRemoval TBR=jochen@chromium.org,zhenw@chromium.org BUG= 722703 Review-Url: https://codereview.chromium.org/2887683002 Cr-Commit-Position: refs/heads/master@{#472103} [modify] https://crrev.com/28e83a520012c28bb95b0b41cac7a03071001a4f/chrome/browser/extensions/execute_script_apitest.cc
,
May 16 2017
zhen, I have a change in review that will address this issue: https://codereview.chromium.org/2885053002 - once this lands, you should be able to re-land your change.
,
May 16 2017
I'm confident that the rollback of Zhen's change should address the test flake, so I'm going to revert the disabling of the affected test now.
,
May 16 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/e56b2dd082428125bac359c3511a2912d5ca3229 commit e56b2dd082428125bac359c3511a2912d5ca3229 Author: bmcquade <bmcquade@chromium.org> Date: Tue May 16 16:12:01 2017 Revert of Disable flaky DestructiveScriptTest.MacrotaskRemoval (patchset #1 id:1 of https://codereview.chromium.org/2887683002/ ) Reason for revert: Revert https://codereview.chromium.org/2885773002/ should address the flake, so this test can now be re-enabled. Original issue's description: > Disable flaky DestructiveScriptTest.MacrotaskRemoval > > TBR=jochen@chromium.org,zhenw@chromium.org > BUG= 722703 > > Review-Url: https://codereview.chromium.org/2887683002 > Cr-Commit-Position: refs/heads/master@{#472103} > Committed: https://chromium.googlesource.com/chromium/src/+/28e83a520012c28bb95b0b41cac7a03071001a4f TBR=kolos@chromium.org # Skipping CQ checks because original CL landed less than 1 days ago. NOPRESUBMIT=true NOTREECHECKS=true NOTRY=true BUG= 722703 Review-Url: https://codereview.chromium.org/2889633002 Cr-Commit-Position: refs/heads/master@{#472123} [modify] https://crrev.com/e56b2dd082428125bac359c3511a2912d5ca3229/chrome/browser/extensions/execute_script_apitest.cc
,
May 16 2017
,
May 16 2017
Thanks for taking care of this! I will reland my CL after Bryan's CL.
,
May 16 2017
,
May 16 2017
Lowering priority and removing sheriff label since this is no longer blocking the tree.
,
May 16 2017
zhen, my patch just landed, so you should be able to re-land your frame timing patch now.
,
Dec 16 2017
|
|||||||
►
Sign in to add a comment |
|||||||
Comment 1 by machenb...@chromium.org
, May 16 2017Labels: -Pri-3 Sheriff-Chromium Pri-0
Owner: ----
Status: Available (was: Assigned)