Closed Bug 1125679 Opened 10 years ago Closed 10 years ago

Timeouts on Linux and Mac in runreftest.py aren't actually handled by killing the process, "NameError: global name 'proc' is not defined"

Categories

(Testing :: Reftest, defect)

Product:
Testing
Component:
Reftest
Type:
defect
Priority:
Not set
Severity:
normal

Tracking

(Not tracked)

Status:
RESOLVED DUPLICATE of bug 1130268

People

(Reporter: philor, Assigned: ted)

References

Details

(Keywords: regression)

Starting back in October when bug 915865 landed (see, e.g. bug 1011786 comment 122), instead of crashing the browser and getting a stack after 330 second timeouts in test suites run by the reftest harness, on Linux and Mac we instead get "NameError: global name 'proc' is not defined" and then sit around for another 1000 seconds waiting for the next layer in the stack to time us out. Pretty awesome that we haven't cared in the least until right now, when the persistent hangs in 10.10 crashtests finally forced me to notice it. And I'm still not going to notice the fact that starting at the exact same time, every Windows reftest harness hang prints a "NameError: global name 'printstatus' is not defined" error.
Status: NEW → RESOLVED
Closed: 10 years ago
Resolution: --- → DUPLICATE
You need to log in before you can comment on or make changes to this bug.