New issue
Advanced search Search tips

Issue 804892 link

Starred by 1 user

Issue metadata

Status: Assigned
Owner:
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Android
Pri: 2
Type: Bug



Sign in to add a comment

ParameterizedWebFrameTest.CallingPostPausableTaskWhilePaused flaked on Android

Project Member Reported by ellyjo...@chromium.org, Jan 23 2018

Issue description

C  424.219s Main  [ RUN      ] All/ParameterizedWebFrameTest.CallingPostPausableTaskWhilePaused/0
C  424.219s Main  [INFO:SkFontMgr_android_parser.cpp(625)] [SkFontMgr Android Parser] '/system/etc/fonts.xml' could not be opened
C  424.219s Main  
C  424.219s Main  [INFO:SkFontMgr_android_parser.cpp(625)] [SkFontMgr Android Parser] '/vendor/etc/fallback_fonts.xml' could not be opened
C  424.219s Main  
C  424.219s Main  ../../third_party/WebKit/Source/core/exported/WebFrameTest.cpp:648: Failure
C  424.219s Main  Value of: callback_helper.result()
C  424.219s Main    Actual: false
C  424.219s Main  Expected: true
C  424.219s Main  [  FAILED  ] All/ParameterizedWebFrameTest.CallingPostPausableTaskWhilePaused/0, where GetParam() = false (101 ms)
 

Comment 1 by joedow@chromium.org, Jan 24 2018

Components: Blink>Infra
We've seen this sporadically for several days now so I am going to disable the test.
Project Member

Comment 2 by bugdroid1@chromium.org, Jan 24 2018

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/c3627e8a96c6e29083e376ebd3fb8f18c35ab9bf

commit c3627e8a96c6e29083e376ebd3fb8f18c35ab9bf
Author: Joe Downing <joedow@chromium.org>
Date: Wed Jan 24 23:02:35 2018

Disabling flaky test: CallingPostPausableTaskWhilePaused on Android

I see sporadic failures going back several days so I am going to
disable this test.

TBR=jbudorick@chromium.org

BUG=804892

Change-Id: I55f9af5935e7eb200c3cf963df2d7ce05dc3b991
Reviewed-on: https://chromium-review.googlesource.com/884392
Reviewed-by: Joe Downing <joedow@chromium.org>
Commit-Queue: Joe Downing <joedow@chromium.org>
Cr-Commit-Position: refs/heads/master@{#531733}
[modify] https://crrev.com/c3627e8a96c6e29083e376ebd3fb8f18c35ab9bf/third_party/WebKit/Source/core/exported/WebFrameTest.cpp

Comment 3 by foolip@chromium.org, Jan 25 2018

Components: -Blink>Infra Platform>Extensions
Owner: rdevlin....@chromium.org
This test was added in https://chromium-review.googlesource.com/807346, so assigning to  rdevlin.cronin@ and guessing the component to get it out of Blink>Infra triage.
Status: Assigned (was: Untriaged)
[Extensions Triage] Marking as assigned.
Cc: haraken@chromium.org jbroman@chromium.org dcheng@chromium.org
+dcheng@, haraken@, jbroman@ from the original patch.  I sadly have no idea about android + blink - any ideas here?
Without having dug too deeply, two thoughts:

1. Maybe something about the way the timer task is scheduled (through the various indirections implied by the timer task runner) cause the exit-run-loop task to sometime be ordered before the timer task.

2. (less likely) Maybe the exit-run-loop task is targeting some inner run loop somewhere.

If it's just (1), one fix might be to have the test complete by using ExitRunLoop (though really we ought to be capturing the QuitClosure of the real base run loop, to be less deprecated).

If it's (2), then maybe tweaking RunPendingTasks to explicitly make a base::RunLoop and post its quit closure instead of the one that looks up the present top RunLoop would work.

Of course, reproing would determine exactly what is happening.

Sign in to add a comment