1206689 - SERVICE_WORKER_LIFE_TIME histogram change detected for the Nightly build of 2015-09-17

Status: RESOLVED INVALID

(Core :: DOM: Core & HTML, defect)

Description

[Roberto Agostino Vitillo (:rvitillo)]

Alert details: http://alerts.telemetry.mozilla.org/index.html#/detectors/1/metrics/1336/alerts/?from=2015-09-17&to=2015-09-17 

Changeset for 20150917030229: https://hg.mozilla.org/mozilla-central/pushloghtml?fromchange=3e8dde8f8c174cce2c0b65c951808f88e35d1875&tochange=e7d613b3bcfe1e865378bfac37de64560d1234ec

Comment 1

[Vladan Djeric (:vladan)]

Ehsan: any ideas on what could cause this change?

Comment 2

[(no longer active)]

I have no idea what this bug means.  The range in comment 0 is bogus.  This has nothing to do with the code changes, if I'm understanding the telemetry probe correctly.

Comment 3

[Roberto Agostino Vitillo (:rvitillo)]

Ehsan, that's an automatic alert generated by cerberus [1]. It says that the distribution for the  SERVICE_WORKER_LIFE_TIME histogram changed in the build-id of the 2015-09-17 compared to how the distribution looked like in the previous N build-ids.

The range provided might not necessarily contain the cause for it, but in general it's likely to do so.

[1] https://github.com/mozilla/cerberus

Comment 4

[(no longer active)]

I see, thanks.

The telemetry probe in question is measuring a time value from user's machines, so its changes can be very rarely attributed to the code.

What is this bug asking for?  I'm not really sure how I can help.

Comment 5

[Roberto Agostino Vitillo (:rvitillo)]

I filed this bug to let the owner(s) of the probe know that this distribution changed and to determine firstly if we should care about it and secondly, if a change in the code might have triggered this.

Comment 6

[(no longer active)]

(In reply to Roberto Agostino Vitillo (:rvitillo) from comment #5)
> I filed this bug to let the owner(s) of the probe know that this
> distribution changed and to determine firstly if we should care about it and
> secondly, if a change in the code might have triggered this.

OK.  I don't think any action is needed here.