+++ This bug was initially created as a clone of Bug #1428790 +++ https://treeherder.mozilla.org/logviewer.html#?job_id=157904426&repo=mozilla-inbound#L12306 TEST-START | tp6_facebook_heavy 22:29:40 INFO - Initialising browser for tp6_facebook_heavy test... 22:29:54 INFO - Local copy of 'simple' is fresh enough 22:29:54 INFO - 0 days old 22:29:54 INFO - Cloning profile located at /Users/cltbld/.mozilla/profiles/simple 22:29:54 INFO - => bookmarkbackups 22:29:54 INFO - => browser-extension-data 22:29:54 INFO - => browser-extension-data/screenshots@mozilla.org 22:29:54 INFO - => cache2 22:29:54 INFO - => cache2/doomed 22:29:54 INFO - => cache2/entries [taskcluster 2018-01-23T07:03:33.227Z] Aborting task - max run time exceeded!

Somebody should look into the history of this job and see if the time to run has been increasing. It's suspicious that it started failing on Windows, then Linux, and now OS X as well.

This bug has failed 37 times in the last 7 days only on OS X 10 affecting opt build type. Failing tests: opt-talos-tp6-e10, opt-talos-h2-e10s. Lin to the latest log:https://treeherder.mozilla.org/logviewer.html#?repo=mozilla-central&job_id=160022649&lineNumber=1419 Part of that log: INFO - mozcrash Saved minidump as /Users/cltbld/tasks/task_1517546859/build/blobber_upload_dir/D8BF1E5B-57B4-45A7-938B-4041F9376B0A.dmp 22:18:42 INFO - PROCESS-CRASH | tp6_google_heavy | application crashed [unknown top frame] 22:18:42 INFO - Crash dump filename: /var/folders/fd/kt26_m457rl243vm27v2bssr00000w/T/tmpZq0gcF/profile/minidumps/D8BF1E5B-57B4-45A7-938B-4041F9376B0A.dmp 22:18:42 INFO - stderr from minidump_stackwalk: 22:18:42 INFO - 2018-02-01 22:18:39: minidump.cc:4359: INFO: Minidump opened minidump /var/folders/fd/kt26_m457rl243vm27v2bssr00000w/T/tmpZq0gcF/profile/minidumps/D8BF1E5B-57B4-45A7-938B-4041F9376B0A.dmp 22:18:42 INFO - 2018-02-01 22:18:39: minidump.cc:4808: ERROR: ReadBytes: read 0/32 22:18:42 INFO - 2018-02-01 22:18:39: minidump.cc:4453: ERROR: Minidump cannot read header 22:18:42 INFO - 2018-02-01 22:18:39: stackwalk.cc:133: ERROR: Minidump /var/folders/fd/kt26_m457rl243vm27v2bssr00000w/T/tmpZq0gcF/profile/minidumps/D8BF1E5B-57B4-45A7-938B-4041F9376B0A.dmp could not be read 22:18:42 INFO - 2018-02-01 22:18:39: minidump.cc:4331: INFO: Minidump closing minidump 22:18:42 INFO - minidump_stackwalk exited with return code 1 22:18:42 INFO - TEST-UNEXPECTED-ERROR | tp6_google_heavy | Found crashes after test run, terminating test :jmaher could you please take a look, because the test is still failing even if it is disabled.

oh, this is almost perma fail, lets disable on osx.

ok, finally off for osx.

Comment on attachment 8947926 [details] [diff] [review] disable h2 on osx for perma fail on timeout Review of attachment 8947926 [details] [diff] [review]: ----------------------------------------------------------------- Thanks!

Pushed by jmaher@mozilla.com: https://hg.mozilla.org/integration/mozilla-inbound/rev/1149bbf594f5 disable h2 on osx for near perma fail timing out. r=rwood

In the last 7 days we have 38 failures. They occur on macosx64-nightly (opt) and OS X 10.10 (opt and debug). A recent log example: https://treeherder.mozilla.org/logviewer.html#?repo=mozilla-central&job_id=165557481

I think this is the nightly profiling jobs acting up- outside of that it looks like tp6, lets see what things look like with the profiling jobs working.

Ah, yes again the tp6 failure is the same issue as seen in Bug 1439979 - a problem with non-replay 404 requests. Follow along in that bug for progress. Once that's fixed it should remove several of these tp6 intermittents.