Report cloud replay crashes to middleman
Categories
(Core Graveyard :: Web Replay, enhancement)
Tracking
(firefox74 fixed)
Tracking | Status | |
---|---|---|
firefox74 | --- | fixed |
People
(Reporter: bhackett1024, Assigned: bhackett1024)
References
Details
Attachments
(1 file)
When a local replaying process crashes we report it to the middleman so that the latter can attempt crash recovery. When a cloud replaying process crashes we don't do any reporting to the middleman, and activity stalls until the middleman treats the non-responsive process as hanged and marks it as crashed. It would be better if cloud replaying process crashes were immediately reported to the middleman.
Cloud replaying processes use a different mechanism to detect crashes than local replaying processes (the two cases are running on different operating systems, after all). The cloud's mechanism is implemented in the translation layer running there (https://github.com/bhackett1024/ReplayLinker) so all we need in m-c is an API which the translation layer can call to report the crash.
Assignee | ||
Comment 1•6 years ago
|
||
Comment 3•6 years ago
|
||
bugherder |
Updated•5 years ago
|
Description
•