content_browsertests AccessibilityFullscreenBrowserTest.InsideIFrame is flaky on multiple CQ trybots |
||||
Issue description
Fails then passed on retry:
[ RUN ] AccessibilityFullscreenBrowserTest.InsideIFrame
DevTools listening on ws://127.0.0.1:37913/devtools/browser/4489fe7c-7f4a-465a-b6e4-9f7adbcc5347
Fontconfig warning: "/etc/fonts/fonts.conf", line 146: blank doesn't take any effect anymore. please remove it from your fonts.conf
../../content/browser/accessibility/fullscreen_browsertest.cc:141: Failure
Expected equality of these values:
2
CountLinks(manager->GetRoot())
Which is: 1
[ FAILED ] AccessibilityFullscreenBrowserTest.InsideIFrame, where TypeParam = and GetParam() = (297 ms)
Logs:
https://logs.chromium.org/v/?s=chromium%2Fbb%2Ftryserver.chromium.linux%2Fcast_shell_linux%2F534801%2F%2B%2Frecipes%2Fsteps%2Fcontent_browsertests__with_patch_%2F0%2Fstdout
Flakiness Dashboard:
https://test-results.appspot.com/dashboards/flakiness_dashboard.html#testType=content_browsertests&tests=AccessibilityFullscreenBrowserTest.InsideIFrame
,
Mar 5 2018
foolip@, can you PTAL? It seems to me that your r533391 has introduced a flaky test (I see that it also flakes on https://ci.chromium.org/buildbot/chromium.fyi/Site%20Isolation%20Android/?limit=200). FWIW, it is probably too late to revert. It is a pity that back when opening this bug we didn't trace the problem to r533391 (which added content/test/data/accessibility/fullscreen/iframe.html). Looking at the test code I can't spot anything obviously wrong, but apparently either 1) the test doesn't wait long enough or 2) the product code doesn't detect the new links in some cases. Some notes / questions: document.onwebkitfullscreenchange = function() { const iframe = document.createElement('iframe'); iframe.srcdoc = '<a href="/">link inside iframe</a>'; document.body.appendChild(iframe); iframe.onload = function() { document.webkitExitFullscreen(); document.onwebkitfullscreenchange = function() { <- why add event listener *after* triggering an event? button.setAttribute('aria-label', 'Done'); <- are all "link" ax nodes guaranteed to be known to the browser at this point? } } }
,
Mar 11 2018
Also on win10_chromium_x64_rel_ng: https://ci.chromium.org/buildbot/tryserver.chromium.win/win10_chromium_x64_rel_ng/98905
,
Mar 14 2018
Issue 821701 has been merged into this issue.
,
Apr 5 2018
Marking a11y-secondary since this appears to be automation related. Please leave a comment if this is actually Windows UI based. |
||||
►
Sign in to add a comment |
||||
Comment 1 by ynovikov@chromium.org
, Feb 23 2018Labels: OS-Android