1196890 - Report the value of xpinstall.signatures.required to FHR

Status: RESOLVED FIXED

Comment 1

[Dave Townsend [:mossop] (he/him)]

Attachment: patch

This adds a few prefs to telemetry:

xpinstall.signatures.required: Tells us if signature checking has been turned off
extensions.autoDisableScopes: Tells us if something has changed the pref controling some sideloading controls
extensions.enabledScopes: Tells us if something has changed the locations we look for add-ons
AddonManager.checkCompatibility: Tells us if application version compatibility checks have been disabled (adding this mostly because I suspect no-one uses it anymore and we might want to remove support for it).

Comment 2

[Benjamin Smedberg]

Comment on attachment 8650673 [details] [diff] [review]
patch

settings.addonCompatibilityCheckEnabled can't change during runtime, right? If so r+datareview=me

Comment 3

[Dave Townsend [:mossop] (he/him)]

(In reply to Benjamin Smedberg  [:bsmedberg] from comment #2)
> Comment on attachment 8650673 [details] [diff] [review]
> patch
> 
> settings.addonCompatibilityCheckEnabled can't change during runtime, right?
> If so r+datareview=me

It can if a user modifies the prefs in about:config. What do I need to do in that case?

Comment 4

[Benjamin Smedberg]

Does that matter to you or anyone doing analysis? i.e. do you really want to split subsessions if somebody changes the pref? I'll assume no, and this is fine.

Comment 5

[Dave Townsend [:mossop] (he/him)]

(In reply to Benjamin Smedberg  [:bsmedberg] from comment #4)
> Does that matter to you or anyone doing analysis? i.e. do you really want to
> split subsessions if somebody changes the pref? I'll assume no, and this is
> fine.

The only impact it might have is that it could change which add-ons are active when changed at runtime, but I'm guessing that changing add-ons already creates a new session anyway. Mostly I just want to know how many have this pref set.

Comment 6

[Pulsebot]

https://hg.mozilla.org/integration/fx-team/rev/aac9d5e8a700

Comment 7

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/aac9d5e8a700

Comment 8

[Dave Townsend [:mossop] (he/him)]

Comment on attachment 8650673 [details] [diff] [review]
patch

Given our concerns over whether or not add-on signing should become mandatory we'd like to understand how many users are manually disabling the pref in as many branches as possible.

Approval Request Comment
[Feature/regressing bug #]: No bug
[User impact if declined]: None
[Describe test coverage new/current, TreeHerder]: On nightly for a week, data is showing up in telemetry
[Risks and why]: Very low risk, this patch just adds some values to the telemetry ping.
[String/UUID change made/needed]: None

Comment 9

[Ritu Kothari (:ritu)]

Comment on attachment 8650673 [details] [diff] [review]
patch

Adding additional telemetry probes on add-on signing. The risk is low, let's uplift to Aurora42 and Beta41.

Comment 10

[Wes Kocher (:KWierso)]

https://hg.mozilla.org/releases/mozilla-aurora/rev/cef90ae4d228

Comment 11

[Wes Kocher (:KWierso)]

https://hg.mozilla.org/releases/mozilla-beta/rev/8b8b71abe67b

Comment 12

[Wes Kocher (:KWierso)]

And a followup on Aurora to fix xpcshell test failures: https://hg.mozilla.org/releases/mozilla-aurora/rev/091859006f2b
And the same for Beta: https://hg.mozilla.org/releases/mozilla-beta/rev/e4ac4ddeef7f

Comment 13

[Wes Kocher (:KWierso)]

And more followups because I apparently cut out a line from the patch while uplifting it:
https://hg.mozilla.org/releases/mozilla-aurora/rev/48cce9628816
https://hg.mozilla.org/releases/mozilla-beta/rev/600951187865
https://hg.mozilla.org/releases/mozilla-release/rev/c2124b4c256b