New issue
Advanced search Search tips

Issue 670295 link

Starred by 1 user

Issue metadata

Status: Fixed
Owner:
Closed: Dec 2016
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 3
Type: Bug
Proj-Servicification

Blocking:
issue 669357



Sign in to add a comment

virtual/mojo-loading/http/tests/preload/external_css_import_preload.html is very flaky

Project Member Reported by jbroman@chromium.org, Dec 1 2016

Issue description

Failure looks like:

FAIL We should scan external css for @imports and preload them assert_less_than: expected a number less than 1064.295 but got 1064.51

Flakiness dashboard:
https://test-results.appspot.com/dashboards/flakiness_dashboard.html#tests=virtual%2Fmojo-loading%2Fhttp%2Ftests%2Fpreload%2Fexternal_css_import_preload.html&testType=webkit_tests

Flakiness began when the test was enabled:
https://codereview.chromium.org/2505933002

Will land a TestExpectations change for now.
 
Cc: yhirano@chromium.org
Owner: csharrison@chromium.org
I can take ownership, thanks for updating test expectations. Moving yhirano to cc.
Project Member

Comment 2 by bugdroid1@chromium.org, Dec 1 2016

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/38debab426d3f491185fc23666877a12b7a39d5d

commit 38debab426d3f491185fc23666877a12b7a39d5d
Author: jbroman <jbroman@chromium.org>
Date: Thu Dec 01 15:25:13 2016

Mark virtual/mojo-loading/http/tests/preload/external_css_import_preload.html as flaky.

BUG= 670295 
TBR=yhirano@chromium.org
NOTRY=true

Review-Url: https://codereview.chromium.org/2549513002
Cr-Commit-Position: refs/heads/master@{#435611}

[modify] https://crrev.com/38debab426d3f491185fc23666877a12b7a39d5d/third_party/WebKit/LayoutTests/TestExpectations

Project Member

Comment 3 by bugdroid1@chromium.org, Dec 5 2016

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/dff0384e2ab5ade040a08a38e72eece575ea0aee

commit dff0384e2ab5ade040a08a38e72eece575ea0aee
Author: csharrison <csharrison@chromium.org>
Date: Mon Dec 05 16:24:43 2016

Don't remove CSSPreloaderResourceClient for unused speculative markup preloads

The CSS @import scanner attaches a passive resource client to a css
preload request. This passive client should not affect the policy
decisions of the preload and should just observe notifications passively.

This patch fixes a bug where removing a passive client from an otherwise
unused preload ends up cancelling it, which removes the preload from
memory cache. This is very wrong behavior, and causes the optimization
to be less effective, and report bad metrics.

Simply not removing the client will not cause the resource to live
longer than necessary, because the client holds only weak references
to the resource.

BUG= 670295 , 662999 

Review-Url: https://codereview.chromium.org/2542183002
Cr-Commit-Position: refs/heads/master@{#436312}

[modify] https://crrev.com/dff0384e2ab5ade040a08a38e72eece575ea0aee/third_party/WebKit/LayoutTests/TestExpectations
[modify] https://crrev.com/dff0384e2ab5ade040a08a38e72eece575ea0aee/third_party/WebKit/Source/core/html/parser/CSSPreloadScanner.cpp

Status: Fixed (was: Assigned)
This should have fixed the flakiness. If it starts occurring again we should re-open the issue.

Comment 5 by laforge@google.com, Nov 7 2017

Components: Internals>Network>Service

Comment 6 by laforge@google.com, Nov 7 2017

Components: -Internals>Network>Service Internals>Services>Network
Apologies, applied the wrong component in bulk.

Sign in to add a comment