New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.

Issue 646836 link

Starred by 1 user

Issue metadata

Status: Fixed
Owner:
Closed: Nov 2016
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Linux
Pri: 2
Type: Bug



Sign in to add a comment

NavigationControllerBrowserTest.FrameNavigationEntry_NewSubframe flaky

Project Member Reported by grt@chromium.org, Sep 14 2016

Issue description

For example, from https://build.chromium.org/p/chromium.linux/builders/Linux%20Tests/builds/46219/steps/content_site_isolation_browsertests%20on%20Ubuntu-12.04/logs/stdio:

[ RUN      ] NavigationControllerBrowserTest.FrameNavigationEntry_NewSubframe
[32299:32299:0914/053346:8163233107:WARNING:audio_manager.cc(317)] Multiple instances of AudioManager detected
[32299:32299:0914/053346:8163233161:WARNING:audio_manager.cc(278)] Multiple instances of AudioManager detected
Xlib:  extension "RANDR" missing on display ":9".
[32299:32299:0914/053347:8163428287:WARNING:render_frame_host_impl.cc(2023)] OnDidStopLoading was called twice.
../../content/browser/frame_host/navigation_controller_impl_browsertest.cc:2419: Failure
Value of: visualizer.DepictFrameTree(root)
  Actual: " Site A ------------ proxies for B C\n   |--Site A ------- proxies for B C\n   +--Site B ------- proxies for A C\nWhere A = http://127.0.0.1/\n      B = http://baz.com/\n      C = http://foo.com/"
Expected: " Site A ------------ proxies for B\n" "   |--Site A ------- proxies for B\n" "   +--Site B ------- proxies for A\n" "Where A = http://127.0.0.1/\n" "      B = http://baz.com/"
Which is: " Site A ------------ proxies for B\n   |--Site A ------- proxies for B\n   +--Site B ------- proxies for A\nWhere A = http://127.0.0.1/\n      B = http://baz.com/"
With diff:
@@ -1,5 +1,6 @@
- Site A ------------ proxies for B
-   |--Site A ------- proxies for B
-   +--Site B ------- proxies for A
+ Site A ------------ proxies for B C
+   |--Site A ------- proxies for B C
+   +--Site B ------- proxies for A C
 Where A = http://127.0.0.1/
       B = http://baz.com/
+      C = http://foo.com/

[  FAILED  ] NavigationControllerBrowserTest.FrameNavigationEntry_NewSubframe, where TypeParam =  and GetParam() =  (446 ms)
 
Project Member

Comment 1 by bugdroid1@chromium.org, Sep 14 2016

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/fe9d10edbfa16ecba98bbfe1bbf7889f3b9795b6

commit fe9d10edbfa16ecba98bbfe1bbf7889f3b9795b6
Author: grt <grt@chromium.org>
Date: Wed Sep 14 14:03:26 2016

Disable FrameNavigationEntry_NewSubframe on Linux due to flakes.

BUG= 646836 
TBR=alexmos@chromium.org
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation

Review-Url: https://codereview.chromium.org/2338963003
Cr-Commit-Position: refs/heads/master@{#418549}

[modify] https://crrev.com/fe9d10edbfa16ecba98bbfe1bbf7889f3b9795b6/content/browser/frame_host/navigation_controller_impl_browsertest.cc

This test is also flaky on Android - it failed on https://build.chromium.org/p/chromium.fyi/builders/Site%20Isolation%20Android/builds/363 (and on 367, but not on 365, 364, 362, 361, 360).  I'll put together a CL to disable.

Comment 3 by creis@chromium.org, Oct 25 2016

Cc: lukasza@chromium.org nick@chromium.org nasko@chromium.org
Components: UI>Browser>Navigation
I think this is just a matter of waiting for the C process to exit before that check.

Nick, did you have a bug on file for your new test framework plans that would take care of this transparently?  Would be good to leave a breadcrumb here, though I might just fix this test for the time being.
Project Member

Comment 4 by bugdroid1@chromium.org, Oct 25 2016

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/7aafd756bccbb4a289b57870906e44e6a866d7a7

commit 7aafd756bccbb4a289b57870906e44e6a866d7a7
Author: lukasza <lukasza@chromium.org>
Date: Tue Oct 25 17:49:00 2016

Disable FrameNavigationEntry_NewSubframe on Android due to flakes.

BUG= 646836 
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation

Review-Url: https://codereview.chromium.org/2439263002
Cr-Commit-Position: refs/heads/master@{#427404}

[modify] https://crrev.com/7aafd756bccbb4a289b57870906e44e6a866d7a7/content/browser/frame_host/navigation_controller_impl_browsertest.cc

Project Member

Comment 5 by bugdroid1@chromium.org, Nov 7 2016

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/b2d5f1f334cfeab18119f38708d0cd191e4eac2f

commit b2d5f1f334cfeab18119f38708d0cd191e4eac2f
Author: creis <creis@chromium.org>
Date: Mon Nov 07 23:25:05 2016

Fix more flaky OOPIF tests by waiting for the old RenderFrame to be deleted.

This fixes the following tests in --site-per-process mode:
NavigationControllerBrowserTest.FrameNavigationEntry_NewSubframe
FrameTreeBrowserTest.NavigateGrandchildToBlob

BUG=662550,  646836 
TEST=Bots stay green.
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation

Review-Url: https://codereview.chromium.org/2479963002
Cr-Commit-Position: refs/heads/master@{#430412}

[modify] https://crrev.com/b2d5f1f334cfeab18119f38708d0cd191e4eac2f/content/browser/frame_host/frame_tree_browsertest.cc
[modify] https://crrev.com/b2d5f1f334cfeab18119f38708d0cd191e4eac2f/content/browser/frame_host/navigation_controller_impl_browsertest.cc

Comment 6 by creis@chromium.org, Nov 11 2016

Status: Fixed (was: Assigned)

Sign in to add a comment