Closed Bug 1298424 Opened 8 years ago Closed 7 years ago

Measure and graph release48 ER per OS

Categories

(Toolkit :: Telemetry, defect, P3)

defect

Tracking

()

RESOLVED FIXED

People

(Reporter: chutten, Unassigned)

Details

Like https://sql.telemetry.mozilla.org/dashboard/client-count-firefox-e10s-release- but one for each of Windows, Linux , and Darwin.

Or possibly one graph with six lines: test and control for each of the top three.

Why? Darwin and Linux have been unthrottled for 48 updates since earlier this week, so they are now representative populations for those FF populations on those OSes.

Also this information may help us identify sub-population issues with e10s, which will inform e10s rollout.
How's this look? https://sql.telemetry.mozilla.org/queries/1097/source#1910

test is above control for all three OSs surveyed. Linux control is the least engaged. All the test branches are within .15 of each other.

There is an inflection in the Mac population at Aug 21 (see the drop in Aug 22's numbers) which is likely exactly when we unstoppered updates. I'm not sure why Linux doesn't have as obvious a feature at the same time. Maybe the trend will be more obvious in a couple days.

I added a DAU plot [1] and a MAU plot [2] to look into it, and it seems as though the Linux population hasn't really grown at all. Are we sure it was given permission to update? 

[1]: https://sql.telemetry.mozilla.org/queries/1097/source#1911
[2]: https://sql.telemetry.mozilla.org/queries/1097/source#1912
Priority: -- → P3
Is it worrisome that Linux DAU showed no appreciable upswing around the time Mac did? (See https://sql.telemetry.mozilla.org/queries/1097/source#1911)
Flags: needinfo?(sledru)
I don't know. We tested and updates are working fine on GNU/Linux.
Maybe the eligible population is smaller?
Flags: needinfo?(sledru)
That would make sense, thank you.
Status: NEW → RESOLVED
Closed: 7 years ago
Resolution: --- → FIXED
You need to log in before you can comment on or make changes to this bug.