Add `extensions.legacy.enabled` pref to Telemetry Aggregates

RESOLVED FIXED

Status

P2
normal
RESOLVED FIXED
a year ago
a year ago

People

(Reporter: bmiroglio, Unassigned)

Tracking

Details

Attachments

(1 attachment, 1 obsolete attachment)

(Reporter)

Description

a year ago
*This is for Nightly 57 ONLY*

Having this pref in TMO would help the add-ons team 

1.) track blowback on Nightly since we just set extensions.legacy.enabled = false by default. 
2.) filter out Users on Nightly 57 that reset the pref back to true

The alternative is generating custom reports, however if adding the pref to TMO is trivial then let's do that :). 

Frank--what's your take on this?
(Reporter)

Updated

a year ago
Summary: Add `extensions.legacy.enabled` pref as a filter Telemetry Aggregates → Add `extensions.legacy.enabled` pref to Telemetry Aggregates

Updated

a year ago
Assignee: nobody → fbertsch
Status: NEW → ASSIGNED
Priority: -- → P1
We've decided that this is too specific of an ask for the aggregator. Instead we should add this information to main_summary, where we can easily query it in re:dash and build reports there.

The schema needs updated: https://github.com/mozilla/telemetry-batch-view/blob/master/src/main/scala/com/mozilla/telemetry/views/MainSummaryView.scala#L799

As well as the function to retrieve the prefs: https://github.com/mozilla/telemetry-batch-view/blob/master/src/main/scala/com/mozilla/telemetry/views/MainSummaryView.scala#L307

And then it will show. Does this work for you, Ben?
Assignee: fbertsch → nobody
Status: ASSIGNED → NEW
Component: Datasets: Telemetry Aggregates → Datasets: Main Summary
Flags: needinfo?(bmiroglio)
(Reporter)

Comment 2

a year ago
Yes, that works for me. Thanks :)
Flags: needinfo?(bmiroglio)

Comment 3

a year ago
Frank, do you have time to work on this?
Flags: needinfo?(fbertsch)
Not immediately, I'm bogged down with experiments + heavy_users stuff.
Flags: needinfo?(fbertsch)

Comment 6

a year ago
User prefs are stored as a nested struct in main_summary.

Unfortunately, adding new fields to a nested struct causes schema evolution to fail (and hence will break the dataset in re:dash).

The above PR will need to be re-worked to use a different approach - I suggest we flatten all the prefs into top-level fields named "user_pref_<sanitized pref name>"

Frank, what do you think of this approach (it's similar to what we do for scalars and histograms currently)?
Flags: needinfo?(fbertsch)

Updated

a year ago
Points: --- → 2
Priority: P1 → P2

Updated

a year ago
Attachment #8918977 - Attachment is obsolete: true
Good plan, happy to see you kept the old prefs as they were. I'm reviewing now.
Flags: needinfo?(fbertsch)

Comment 9

a year ago
This has been added - a new top-level field called "user_pref_extensions_legacy_enabled" will appear in tomorrow's table.
Status: NEW → RESOLVED
Last Resolved: a year ago
Resolution: --- → FIXED
You need to log in before you can comment on or make changes to this bug.