New issue
Advanced search Search tips

Issue 876947 link

Starred by 1 user

Issue metadata

Status: Fixed
Owner:
Closed: Dec 7
Components:
EstimatedDays: ----
NextAction: ----
OS: Windows
Pri: 3
Type: Bug



Sign in to add a comment

RenderThreadImplBrowserTest.NonResourceDispatchIPCTasksDontGoThroughScheduler failing on Windows with Network Service enabled

Project Member Reported by mattm@chromium.org, Aug 23

Issue description

RenderThreadImplBrowserTest.NonResourceDispatchIPCTasksDontGoThroughScheduler currently can fail for two different reasons on windows:

1) Consistently failing to find the utility service process exe on Windows, like:
[FATAL:utility_process_host.cc(263)] Check failed: false. Unable to get utility process binary name.

2) If #1 is fixed, flakily failing if the test cleans up and destroys the TestBrowserThreadBundle while ChildProcessLauncherHelper is running on the ProcessLauncherTaskRunner, when ChildProcessLauncherHelper::PostLaunchOnLauncherThread calls mojo::OutgoingInvitation::Send which eventually ends up trying to PostTask back to the IO task runner which has already been reset.
[8368:4188:0822/124450.321:11590265:FATAL:browser_thread_impl.cc(152)] Check failed: base::subtle::NoBarrier_Load(&globals.states[identifier]) >= BrowserThreadState::RUNNING (0 vs. 1)
Backtrace:
	base::debug::StackTrace::StackTrace [0x00007FF663F0D5E4+36]
	logging::LogMessage::~LogMessage [0x00007FF663F29E12+98]
	content::BrowserThread::PostTask [0x00007FF6627E50EE+414]
	content::BrowserThread::GetCurrentThreadIdentifier [0x00007FF6627E54EB+171]
	base::TaskRunner::PostTask [0x00007FF663F8010F+63]
	mojo::core::NodeController::SendBrokerClientInvitation [0x00007FF662686356+614]
	mojo::core::Core::SendInvitation [0x00007FF662679818+1992]
	MojoSendInvitationImpl [0x00007FF66267FA9D+45]
	MojoSendInvitation [0x00007FF663FCB638+120]
	mojo::OutgoingInvitation::Send [0x00007FF663FC7961+545]
	mojo::OutgoingInvitation::Send [0x00007FF663FC77B9+121]
	content::internal::ChildProcessLauncherHelper::PostLaunchOnLauncherThread [0x00007FF662817AB0+624]
	content::internal::ChildProcessLauncherHelper::LaunchOnLauncherThread [0x00007FF662817792+354]
	base::debug::TaskAnnotator::RunTask [0x00007FF664A3E588+376]
	base::internal::TaskTracker::RunOrSkipTask [0x00007FF6648BC584+1188]
	base::internal::TaskTracker::RunAndPopNextTask [0x00007FF6648BB7BB+507]
	base::internal::SchedulerWorker::RunWorker [0x00007FF666143700+1136]
	base::internal::SchedulerWorker::RunDedicatedWorker [0x00007FF6661431F0+32]
	base::PlatformThread::SetCurrentThreadPriority [0x00007FF663F80A9C+572]
	BaseThreadInitThunk [0x00007FFD994C2774+20]
	RtlUserThreadStart [0x00007FFD9A8D0D51+33]
 
Labels: Hotlist-KnownIssue
Project Member

Comment 2 by bugdroid1@chromium.org, Aug 24

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/e0a83f964850d26afe558c5a907a1215909ce593

commit e0a83f964850d26afe558c5a907a1215909ce593
Author: Matt Mueller <mattm@chromium.org>
Date: Fri Aug 24 23:50:42 2018

Move RegisterPathProvider() from ContentTestSuite to ContentTestSuiteBase

Fixes network_service_content_browsertests RenderThreadImplBrowserTest.NonResourceDispatchIPCTasksDontGoThroughScheduler
failing to find the utility service process exe on Windows, like:
[FATAL:utility_process_host.cc(263)] Check failed: false. Unable to get utility process binary name.

Bug:  876947 
Cq-Include-Trybots: luci.chromium.try:linux_mojo
Change-Id: Ia7f51366e4b4db27ec8c3bbee3d57866c1bd67b9
Reviewed-on: https://chromium-review.googlesource.com/1185628
Commit-Queue: Matt Mueller <mattm@chromium.org>
Reviewed-by: John Abd-El-Malek <jam@chromium.org>
Cr-Commit-Position: refs/heads/master@{#586063}
[modify] https://crrev.com/e0a83f964850d26afe558c5a907a1215909ce593/content/public/test/content_test_suite_base.cc
[modify] https://crrev.com/e0a83f964850d26afe558c5a907a1215909ce593/content/test/content_test_suite.cc

Project Member

Comment 3 by bugdroid1@chromium.org, Aug 25

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/cbdabdff5ab1f19b993e36f2958720d6295ea829

commit cbdabdff5ab1f19b993e36f2958720d6295ea829
Author: Wez <wez@chromium.org>
Date: Sat Aug 25 22:29:35 2018

Revert "Move RegisterPathProvider() from ContentTestSuite to ContentTestSuiteBase"

This reverts commit e0a83f964850d26afe558c5a907a1215909ce593.

Reason for revert: ContentTestSuiteBase is used as the base for both
ContentTestSuite and ContentBrowserTestSuite, as well as for Chrome and
extensions browser-test suites, so putting RegisterPathProvider() in the
base class causes duplicate-registration errors in browser-tests. e.g
see: 
https://ci.chromium.org/p/chromium/builders/luci.chromium.ci/Linux%20Tests%20(dbg)(1)/73970

Original change's description:
> Move RegisterPathProvider() from ContentTestSuite to ContentTestSuiteBase
> 
> Fixes network_service_content_browsertests RenderThreadImplBrowserTest.NonResourceDispatchIPCTasksDontGoThroughScheduler
> failing to find the utility service process exe on Windows, like:
> [FATAL:utility_process_host.cc(263)] Check failed: false. Unable to get utility process binary name.
> 
> Bug:  876947 
> Cq-Include-Trybots: luci.chromium.try:linux_mojo
> Change-Id: Ia7f51366e4b4db27ec8c3bbee3d57866c1bd67b9
> Reviewed-on: https://chromium-review.googlesource.com/1185628
> Commit-Queue: Matt Mueller <mattm@chromium.org>
> Reviewed-by: John Abd-El-Malek <jam@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#586063}

TBR=jam@chromium.org,mattm@chromium.org

Change-Id: I649f105c58d9643dd4cab5aa685b3c6e4556097e
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug:  876947 
Cq-Include-Trybots: luci.chromium.try:linux_mojo
Reviewed-on: https://chromium-review.googlesource.com/1189444
Reviewed-by: Wez <wez@chromium.org>
Commit-Queue: Wez <wez@chromium.org>
Cr-Commit-Position: refs/heads/master@{#586152}
[modify] https://crrev.com/cbdabdff5ab1f19b993e36f2958720d6295ea829/content/public/test/content_test_suite_base.cc
[modify] https://crrev.com/cbdabdff5ab1f19b993e36f2958720d6295ea829/content/test/content_test_suite.cc

Project Member

Comment 4 by bugdroid1@chromium.org, Nov 6

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/712deda11ebc1d1de5783911c107f0217f5ea2d9

commit 712deda11ebc1d1de5783911c107f0217f5ea2d9
Author: Matt Mueller <mattm@chromium.org>
Date: Tue Nov 06 21:53:44 2018

Re-enable RenderThreadImplBrowserTest.NonResourceDispatchIPCTasksDontGoThroughScheduler

Bug:  827318 , 876947 
Change-Id: I72ead221b797a90c19d7742422d65a29ed5f1741
Reviewed-on: https://chromium-review.googlesource.com/c/1318849
Reviewed-by: John Abd-El-Malek <jam@chromium.org>
Commit-Queue: Matt Mueller <mattm@chromium.org>
Cr-Commit-Position: refs/heads/master@{#605833}
[modify] https://crrev.com/712deda11ebc1d1de5783911c107f0217f5ea2d9/testing/buildbot/filters/mojo.fyi.network_content_browsertests.filter

Project Member

Comment 5 by bugdroid1@chromium.org, Nov 7

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/afc8b5307d59f96be9668baa6937ac32db091aaf

commit afc8b5307d59f96be9668baa6937ac32db091aaf
Author: Matt Mueller <mattm@chromium.org>
Date: Wed Nov 07 20:16:57 2018

Revert "Re-enable RenderThreadImplBrowserTest.NonResourceDispatchIPCTasksDontGoThroughScheduler"

This reverts commit 712deda11ebc1d1de5783911c107f0217f5ea2d9.

Reason for revert: Still fails on the waterfall.

Original change's description:
> Re-enable RenderThreadImplBrowserTest.NonResourceDispatchIPCTasksDontGoThroughScheduler
> 
> Bug:  827318 , 876947 
> Change-Id: I72ead221b797a90c19d7742422d65a29ed5f1741
> Reviewed-on: https://chromium-review.googlesource.com/c/1318849
> Reviewed-by: John Abd-El-Malek <jam@chromium.org>
> Commit-Queue: Matt Mueller <mattm@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#605833}

TBR=jam@chromium.org,mattm@chromium.org

Change-Id: I8181253eef71bad3b5a78e633778a9a110c4695c
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug:  827318 ,  876947 
Reviewed-on: https://chromium-review.googlesource.com/c/1324211
Reviewed-by: Matt Mueller <mattm@chromium.org>
Commit-Queue: Matt Mueller <mattm@chromium.org>
Cr-Commit-Position: refs/heads/master@{#606143}
[modify] https://crrev.com/afc8b5307d59f96be9668baa6937ac32db091aaf/testing/buildbot/filters/mojo.fyi.network_content_browsertests.filter

Project Member

Comment 6 by bugdroid1@chromium.org, Dec 5

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/78425a16bd7890ead789f71e602138bd97176b4b

commit 78425a16bd7890ead789f71e602138bd97176b4b
Author: Matt Mueller <mattm@chromium.org>
Date: Wed Dec 05 18:43:22 2018

Re-enable RenderThreadImplBrowserTest.NonResourceDispatchIPCTasksDontGoThroughScheduler on network service tests again.

Passing again now that 3575eb72d8073cbf789dc5436446fe5f888eea52 re-landed.

Bug:  827318 , 876947 
Change-Id: Icf910e02d0ab76fd62f596ecc2253277ba269b0b
Reviewed-on: https://chromium-review.googlesource.com/c/1361868
Reviewed-by: John Abd-El-Malek <jam@chromium.org>
Commit-Queue: Matt Mueller <mattm@chromium.org>
Cr-Commit-Position: refs/heads/master@{#614038}
[modify] https://crrev.com/78425a16bd7890ead789f71e602138bd97176b4b/testing/buildbot/filters/mojo.fyi.network_content_browsertests.filter

Status: Fixed (was: Started)

Sign in to add a comment