Open Bug 1732768 Opened 4 years ago Updated 1 year ago

Intermittent timeouts or seemingly timeout-induced crashes under /encoding/legacy-mb-*

Categories

(Core :: Internationalization, defect)

Product:
Core
Component:
Internationalization
Type:
defect

Tracking

()

People

(Reporter: hsivonen, Unassigned)

References

Details

The WPTs for legacy CJK encodings (under testing/web-platform/tests/encoding/legacy-mb-schinese/ , testing/web-platform/tests/encoding/legacy-mb-tchinese/ , testing/web-platform/tests/encoding/legacy-mb-japanese/ , and testing/web-platform/tests/encoding/legacy-mb-korean/ ) run tests in 1000-character slices.

It seems that occasionally VM execution progress relative to the wall clock is unlucky and these either time out properly or a process goes away and another process experiences the resulting IPC errors as a crash.

We should eventually do something that reduces the alarms that this raises. The most actionable option is probably making these run in 500-character slices.

Blocks: 1686889
No longer blocks: 686889

Henri,

This seems like a relatively low priority intermittent WPT failure with regard to triage. I'm thinking maybe S4. Do you agree?

Flags: needinfo?(hsivonen)

Yeah, let's mark this S4.

Severity: -- → S4
Flags: needinfo?(hsivonen)
Blocks: 1752551
You need to log in before you can comment on or make changes to this bug.