Open
Bug 885824
Opened 12 years ago
Updated 2 years ago
Additional telemetry for diagnostics
Categories
(Firefox :: General, defect)
Firefox
General
Tracking
()
NEW
People
(Reporter: taras.mozilla, Unassigned)
Details
* Whether last update download failed
* Updater type(this might be needed, might be inferred from existing UPDATER_SERVICE_* histograms)...:
** installed from zip or installed in user-writable-dir
** installed as system service for a limited account
** installed a system service for an admin account.
![]() |
||
Comment 1•12 years ago
|
||
(In reply to Taras Glek (:taras) from comment #0)
> * Whether last update download failed
we already report this info.
> * Updater type(this might be needed, might be inferred from existing
> UPDATER_SERVICE_* histograms)...:
What is an updater type?
> ** installed from zip or installed in user-writable-dir
zip would need to be done by Firefox and is not going to be done by app update.
user writable dir is already reported by app update.
> ** installed as system service for a limited account
Fairly certain that we already report info that this can be inferred.
> ** installed a system service for an admin account.
we already report this info.
Moving to Firefox General since the only thing here that I am sure is not reported is zip which will need to be done by Firefox itself and not the installer.
Component: Installer → General
Reporter | ||
Comment 2•12 years ago
|
||
Rob, thanks for clarifying. I'll ping you later for details on what we collect and how to interpret it.
(In reply to Robert Strong [:rstrong] (do not email) from comment #1)
> (In reply to Taras Glek (:taras) from comment #0)
> > * Whether last update download failed
> we already report this info.
How? I'm interested in details like incomplete download, http error code, etc.
> > * Updater type(this might be needed, might be inferred from existing
> > UPDATER_SERVICE_* histograms)...:
> What is an updater type?
update-situation might be a better term.
![]() |
||
Comment 3•12 years ago
|
||
(In reply to Taras Glek (:taras) from comment #2)
> Rob, thanks for clarifying. I'll ping you later for details on what we
> collect and how to interpret it.
>
> (In reply to Robert Strong [:rstrong] (do not email) from comment #1)
> > (In reply to Taras Glek (:taras) from comment #0)
> > > * Whether last update download failed
> > we already report this info.
>
> How? I'm interested in details like incomplete download, http error code,
> etc.
Some of what you want might not be covered but some already is via the update last error.
>
> > > * Updater type(this might be needed, might be inferred from existing
> > > UPDATER_SERVICE_* histograms)...:
> > What is an updater type?
>
> update-situation might be a better term.
This is too vague to implement and is likely covered by other telemetry data points.
There is already a bug filed in app update for additional telemetry data points and those should be added as dependent bugs of bug 715942 as the missing data points are identified.
Summary: Additional telemetry for updater diagnostics → Additional telemetry for diagnostics
![]() |
||
Comment 5•12 years ago
|
||
(In reply to Doug Turner (:dougt) from comment #4)
> squarely in rob's area.
See comment #1 and comment #3. The only thing here that isn't already implemented or covered by existing bugs is the request to report when installed from zip as previously mentioned.
Assignee: robert.bugzilla → nobody
![]() |
||
Comment 6•12 years ago
|
||
(In reply to Taras Glek (:taras) from comment #2)
> Rob, thanks for clarifying. I'll ping you later for details on what we
> collect and how to interpret it.
>
> (In reply to Robert Strong [:rstrong] (do not email) from comment #1)
> > (In reply to Taras Glek (:taras) from comment #0)
> > > * Whether last update download failed
> > we already report this info.
>
> How? I'm interested in details like incomplete download, http error code,
> etc.
Taras, I've been trying to come up with a decent way to report error codes where the error code could be an http error code as well as NS_ERROR_* codes in the same histogram. Is there a reasonable way of doing so without trying to map them all?
Updated•2 years ago
|
Severity: normal → S3
You need to log in
before you can comment on or make changes to this bug.
Description
•