Open Bug 1963076 Opened 14 days ago Updated 14 days ago

xpcom/tests/unit/test_getTimers.js fails often locally

Categories

(Core :: XPCOM, defect, P3)

defect

Tracking

()

ASSIGNED

People

(Reporter: jstutte, Assigned: jstutte)

Details

Attachments

(1 file)

To me and some others, running this test locally often results in failures like this:

0:01.36 INFO (xpcshell/head.js) | test MAIN run_test pending (1)
 0:01.36 INFO "CCGCScheduler::EnsureGCRunner: 250ms, 0"
 0:01.37 FAIL run_test - [run_test : 51] there should be no timer at xpcshell startup - 1 == 0
/home/jens/src/mozilla-unified/obj-x86_64-pc-linux-gnu/_tests/xpcshell/xpcom/tests/unit/test_getTimers.js:run_test:51
/home/jens/src/mozilla-unified/testing/xpcshell/head.js:_execute_test:590
-e:null:1
 0:01.38 INFO exiting test
...
 0:01.64 TEST_END: Test FAIL, expected PASS. Subtests passed 0/1. Unexpected 1
 0:01.65 INFO INFO | Result summary:
 0:01.65 INFO INFO | Passed: 0
 0:01.65 INFO INFO | Failed: 1
 0:01.65 INFO INFO | Todo: 0
 0:01.65 INFO INFO | Retried: 0
 0:01.65 SUITE_END

It seems to be timing dependent if we hit this or not and are currently just lucky in CI ?

I assume the test should never rely on timers.length at all.

Severity: -- → S4
Priority: -- → P3

Given we seem to not hit this often enough in CI to see a bug filed, I assume this has a low severity.

Assignee: nobody → jstutte
Status: NEW → ASSIGNED
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: