TabModelMergingTests consistently fail on N |
||
Issue descriptionhttps://codesearch.chromium.org/chromium/src/chrome/android/javatests/src/org/chromium/chrome/browser/tabmodel/TabModelMergingTest.java?rcl=9ab6f1801d197cda79b2f1dbf8f81d4f8d5f4884&l=98 fails because the tab model selector in the second activity hasn't been initialized. e.g. https://luci-logdog.appspot.com/v/?s=chromium%2Fbb%2Fchromium.android.fyi%2FNougat_Phone_Tester%2F2819%2F%2B%2Frecipes%2Fsteps%2Fchrome_public_test_apk_on_Android%2F0%2Flogs%2Forg.chromium.chrome.browser.tabmodel.TabModelMergingTest_testMergeOnColdStart%2F0 (from https://build.chromium.org/p/chromium.android.fyi/builders/Nougat%20Phone%20Tester/builds/2819)
,
Jun 16 2017
that was my first guess too, but it failed for me locally even after reverting to the junit3 version. this may not have started recently
,
Jun 16 2017
What's our target for getting the N bot healthy and online? The change that throws an illegal state exception while trying to access the tab model selector, while fairly old, was made after the last non-mechanical change to TabModelMergingTest. Last non-mechanical TabModelMergingTest on Aug 15, 2016: https://chromium.googlesource.com/chromium/src/+/1382473e75f7ae9fc20ce45356e1c94855ede461 Change to throw IllegalStateException in #getTabModelSelector on Oct 22, 2016: https://chromium.googlesource.com/chromium/src/+/820a95b0b81d33e42712f9198c215f703412e1a1 We can probably change the criteria helper to rely on ChromeActivity#areTabModelsInitialized() instead of #getTabModelSelector()#isTabStateInitialized() Stack trace: Caused by: java.lang.IllegalStateException: Attempting to access TabModelSelector before initialization at org.chromium.chrome.browser.ChromeActivity.getTabModelSelector(ChromeActivity.java:1480) at org.chromium.chrome.browser.tabmodel.TabModelMergingTest$1.isSatisfied(TabModelMergingTest.java:98) at org.chromium.content.browser.test.util.CriteriaHelper$1.call(CriteriaHelper.java:113) at org.chromium.content.browser.test.util.CriteriaHelper$1.call(CriteriaHelper.java:110) at java.util.concurrent.FutureTask.run(FutureTask.java:237) at android.os.Handler.handleCallback(Handler.java:751) at android.os.Handler.dispatchMessage(Handler.java:95) at android.os.Looper.loop(Looper.java:154) at android.app.ActivityThread.main(ActivityThread.java:6077) at java.lang.reflect.Method.invoke(Native Method) at com.android.internal.os.ZygoteInit$MethodAndArgsCaller.run(ZygoteInit.java:865) at com.android.internal.os.ZygoteInit.main(ZygoteInit.java:755)
,
Jun 16 2017
There are two issues. The first is the IllegalStateException, which can be fixed by checking #areTabModelsInitialized() before TabModelSelector#isTabStateInitialized(). The second is the changes I made a while back to ChromeTabbedActivity#isStartedUpCorrectly() to check if there are multiple instances running in fullscreen mode. In the TabModelMerging tests, isStartedUpCorrectly() returns false and the second activity's launch is aborted. Fix out for review: https://chromium-review.googlesource.com/c/538071/
,
Jun 19 2017
Fix landed: [Android] Fix TabModelMergingTests The TabModelMergingTests have been failing on an IllegalStateException because they are trying to access the tab models before they have been initialized. The CriteriaHelper accessing the tab models has been updated to check that that are initialized before accessing them. The tests are also failing because ChromeTabbedActivity#isStartedUpCorrectly() returns false when two fullscreen ChromeTabbedActivity classes are showing. This change also adds a call to ChromeTabbedActivity#onMultiInstanceModeStarted() so that the #isStartedUpCorrectly() check doesn't fail. BUG= 733890 Change-Id: I4d3e857f357756f68dfcf7a3efb5567c0d50fa31 Reviewed-on: https://chromium-review.googlesource.com/538071 Reviewed-by: Ted Choc <tedchoc@chromium.org> Commit-Queue: Theresa <twellington@chromium.org> Cr-Commit-Position: refs/heads/master@{#480464}
,
Jun 19 2017
,
Jun 19 2017
Thanks for fixing this. To answer your question from #3, we don't have a set timeline at the moment but the ongoing instability in both this bot and other N+ bots is something I'd like to address sooner rather than later -- probably Q3. That I've let the N bot languish in its current state is something I'm not proud of :( |
||
►
Sign in to add a comment |
||
Comment 1 by twelling...@chromium.org
, Jun 16 2017Labels: -Pri-3 Pri-2
Owner: twelling...@chromium.org
Status: Assigned (was: Untriaged)