Add additional metrics to Glean crash ping to reach parity with Telemetry
Categories
(Toolkit :: Crash Reporting, enhancement)
Tracking
()
| Tracking | Status | |
|---|---|---|
| firefox128 | --- | fixed |
People
(Reporter: afranchuk, Assigned: afranchuk)
References
Details
Attachments
(1 file)
This may necessitate new metrics types, e.g. for stack traces.
| Assignee | ||
Comment 1•3 years ago
|
||
There are many ways to organize the metrics. For instance, the platform-specific ones could all be under crash.<platform>. However, I don't see much practical advantage to that, since they are all optional anyway. Another option is to group all the boolean flags together in a labeled_boolean metric (just for the sake of reducing verbosity), however that has a disadvantage of not having separate descriptions for each flag. So for now I'm mostly keeping everything as separate metrics, one for each Telemetry metadata entry (though maybe a few of those are no longer needed?).
Comment 2•3 years ago
|
||
As a first step I'd say we could add the annotations that were in the legacy crash ping minus the stack traces - which are the biggest hurdle.
| Assignee | ||
Comment 3•3 years ago
|
||
Yeah I've essentially done that in my working copy. I'm currently missing stack traces, asyncShutdownTimeout, and the environment (which is a lot of info).
Comment 4•3 years ago
|
||
AFAIK the environment is important and it's one of the other sticky bits aside from the stack trace. Let's put that and the stack traces in separate bugs.
| Assignee | ||
Comment 5•2 years ago
|
||
Comment 7•2 years ago
|
||
Backed out with Bug 1830098 for causing bustage on metrics.rs related to brokenAddBlockers.
Comment 9•2 years ago
|
||
Comment 10•2 years ago
|
||
| Assignee | ||
Updated•2 years ago
|
Comment 11•2 years ago
|
||
| bugherder | ||
Description
•