Migrate mission-control v2 to use telemetry.crash instead of crash_summary
Categories
(Cloud Services :: Mission Control, task)
Tracking
(Not tracked)
People
(Reporter: wlach, Unassigned)
References
Details
Mission Control v2 (https://github.com/mozilla/missioncontrol-v2) is the last known user of the crash summary dataset.
:sguha, :wcbeard: could one of you migrate it to use telemetry.crash as soon as practicable? The data should now be complete back one year (see bug 1568042). This is blocking deprecation of the crash summary dataset (bug 1572069).
|
Reporter | |
Updated•5 years ago
|
|
||
Comment 1•5 years ago
|
||
I'm working on this and seeing discrepancies in the numbers (crash > crash_summary). Here's an example query
https://sql.telemetry.mozilla.org/queries/65888/source
Hoping to get feedback from someone in fx-metrics on if this is expected
|
|
Reporter | |
Comment 2•5 years ago
|
||
I did up a quick analysis of the nightly discrepencies on October 22nd (a few hundred pings in telemetry.crash which were not in crash_summary):
https://iodide.telemetry.mozilla.org/notebooks/322/?viewMode=report
It appears as if those pings that were in telemetry.crash but not crash_summary were submitted rather late in the day (11pm UTC or later), leading me to think that for some reasons these late-submitted pings were simply not being picked up by crash_summary. I'm not sure if this is worth much investigation given that crash_summary is going away -- the important thing is that telemetry.crash does not have this defect.
|
|
Reporter | |
Comment 3•5 years ago
|
||
Hey Chris! I noticed you'd modified missioncontrol-v2 to use telemetry.crash. Please go ahead and close this issue when you're satisfied that things are working as you like, and I'll turn off the crash_summary job
|
|
||
Comment 4•5 years ago
|
||
It's been running for a few days with no perceptible breaks. Closing
Description
•