Closed Bug 898317 Opened 11 years ago Closed 7 years ago

Intermittent browser_newtab_bug735987.js | Test timed out | browser_newtab_drag_drop.js | Test timed out

Categories

(Infrastructure & Operations Graveyard :: CIDuty, task)

x86_64
macOS
task
Not set
normal

Tracking

(Not tracked)

RESOLVED WORKSFORME

People

(Reporter: ttaubert, Unassigned)

References

Details

(Keywords: intermittent-failure, Whiteboard: [buildduty] [test disabled on OS X] [leave open][maybe tied to flaky slaves])

Attachments

(3 files)

https://tbpl.mozilla.org/php/getParsedLog.php?id=25748372&tree=Mozilla-Inbound https://tbpl.mozilla.org/php/getParsedLog.php?id=25762240&tree=Mozilla-Central These test timeouts seem to be caused by popup dialogs displayed on the test slaves, see the attached screenshot. They're saying: "Your computer was restarted because of a problem." Can we somehow disable those dialogs? Or make sure they're not displayed when running tests?
Sorry, wrong component.
Component: General → Release Engineering: Platform Support
Product: Firefox → mozilla.org
QA Contact: coop
Version: Trunk → other
Product: mozilla.org → Release Engineering
Kim, this seems relevant to work you've done in the past. Any chance you could take a look? :)
Flags: needinfo?(kmoir)
Kim, this is rapidly becoming one of our top oranges. Please can you look into this?
I looked at this and it seems that /System/Library/CoreServices/ReportPanic is the one causing the popup. I assume it's okay to just disable sending reports to Apple. I'll test a Puppet patch to remove this notification which seems to be enabled by System/Library/LaunchAgents/com.apple.ReportPanic.plist
Flags: needinfo?(kmoir)
Assignee: nobody → kmoir
Attached patch bug898317.patchSplinter Review
Attachment #800948 - Flags: review?(dustin)
Attachment #800948 - Flags: review?(dustin) → review+
Attachment #800948 - Flags: checked-in+
So it looks like this my patch didn't fix the issue completely. I'm investigating further.
Beginning to wonder if these are slave specific issues. These failures in the last five days have been from talos-r4-lion-003 (4) and talos-r4-lion-082 which I put back into production last night.
Assignee: kmoir → nobody
Whiteboard: [orange][buildduty]
Depends on: 933945
Interesting, this slave ended up in an hdiutil error state after one of these failures earlier today.
All talos-r4-lion-082 - disabling in slavealloc.
Disabling test on OS X for too many intermittent failures (whilst this bug seems to be a platform issue, we're not getting anywhere and the failures are annoying).
Whiteboard: [orange][buildduty] → [buildduty] [test disabled on OS X] [leave open]
Summary: Intermittent OSX 10.7 browser_newtab_bug735987.js | Test timed out | browser_newtab_drag_drop.js | Test timed out → Intermittent browser_newtab_bug735987.js | Test timed out | browser_newtab_drag_drop.js | Test timed out
Tim, any idea why this has been so much more failure-prone on Win8 lately?
Flags: needinfo?(ttaubert)
(In reply to Ryan VanderMeulen [:RyanVM UTC-4] from comment #431) > Tim, any idea why this has been so much more failure-prone on Win8 lately? I think this may have been caused by bug 895359. It has been backed out so let's see if the failure rate decreases.
Flags: needinfo?(ttaubert)
I suspect that this is also tied to flaky slaves, so we should keep an eye on t-snow-r4-0069 and t-snow-r4-0031.
Whiteboard: [buildduty] [test disabled on OS X] [leave open] → [buildduty] [test disabled on OS X] [leave open][maybe tied to flaky slaves]
Status: NEW → RESOLVED
Closed: 7 years ago
Resolution: --- → WORKSFORME
Component: Platform Support → Buildduty
Product: Release Engineering → Infrastructure & Operations
Product: Infrastructure & Operations → Infrastructure & Operations Graveyard
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: