Closed Bug 1731580 Opened 3 years ago Closed 11 months ago

browser-chrome test on Linux asan often fails because worker lost

Categories

(Toolkit :: Password Manager, defect, P3)

defect

Tracking

()

RESOLVED INCOMPLETE
Tracking Status
firefox-esr78 --- unaffected
firefox-esr91 --- unaffected
firefox92 --- unaffected
firefox93 --- unaffected
firefox94 --- wontfix
firefox95 --- wontfix

People

(Reporter: aryx, Assigned: serg)

References

(Regression)

Details

(Keywords: intermittent-failure, regression, Whiteboard: [retriggered] [stockwell unknown])

browser-chrome tests on Linux ASan are observing frequent "retries" for tasks (= task runs again, e.g. because the previous run of the task had the worker executing stopped responding to the CI).

This failed task indicates an issue with the tests in toolkit/components/passwordmgr/test/browser/ (see "Test Groups" tab).

Emilio, could this be triggered by bug 1731347 (scroll down link above to see the increase in blue, retried tasks). The backfills don't identify a push, this could be explained because they inherit the scheduled test manifests (at least).

Flags: needinfo?(emilio)
Summary: browser-chrome test on Linux asan often fails because worker los → browser-chrome test on Linux asan often fails because worker lost

I don't see how since that patch only touches a test that should be run in dt (and should be a no-op on nightly / early beta).

Flags: needinfo?(emilio)

Backfills point to bug 1727158 as regressor.

Flags: needinfo?(nika)
Keywords: regression
Regressed by: 1727158
Has Regression Range: --- → yes

There are 71 total failures in the last 7 days on linux1804-64-asan-qr opt

Whiteboard: [retriggered][stockwell needswork:owner]

Hi Chris,
Is someone in the fission team can take a look at this bug?
This seems to affect the whole password manager browser test.

Flags: needinfo?(cpeterson)

(In reply to Dimi Lee [:dimi] from comment #9)

Is someone in the fission team can take a look at this bug?
This seems to affect the whole password manager browser test.

We might need to back out bug 1727158 for Fx94 and fix these regressions in Nightly 95.

Fission Milestone: --- → MVP
Flags: needinfo?(cpeterson)

Changing severity to N/A because this is a task not a bug.

Assignee: nobody → sgalich
Severity: -- → N/A
Status: NEW → ASSIGNED
Type: defect → task
Priority: -- → P1
Type: task → defect

This test issue doesn't need to block Fission's staged rollout. Clearing needinfo for Nika.

Is this test problem also happening in Beta 94?

Flags: needinfo?(nika)
Whiteboard: [retriggered][stockwell needswork:owner] → [retriggered][stockwell needswork:owner] fission-soft-blocker
Fission Milestone: MVP → Future
Whiteboard: [retriggered][stockwell needswork:owner] fission-soft-blocker → [retriggered][stockwell needswork:owner] [fission:m95?]

It hasn't been observed for the 5 pushes to Beta which use Gecko version 94. The issue also got much less frequent on trunk.

Fission Milestone: Future → ---
Whiteboard: [retriggered] [fission:m95?][stockwell unknown] → [retriggered] [stockwell unknown]

Lowering to P3 because this is not being worked on any time soon and failure rate over last month is low.

Priority: P1 → P3
Severity: N/A → S4
Status: ASSIGNED → RESOLVED
Closed: 11 months ago
Resolution: --- → INCOMPLETE
You need to log in before you can comment on or make changes to this bug.