Alexandre Poirot [:ochameau] Assignee
	Description • 10 years ago

In bug 1072181, we are about to turn app.update.silent to true for the dev edition.
That ends up breaking various update test that rely on this pref to be false by default.

	Alexandre Poirot [:ochameau] Assignee
	Comment 1 • 10 years ago

Here is a patch to just ensure this pref is false when we run update tests.

	Alexandre Poirot [:ochameau] Assignee
	Comment 2 • 10 years ago

https://treeherder.mozilla.org/ui/#/jobs?repo=try&revision=79d182418c32

	Robert Strong (they/them - no direct email)
	Comment 3 • 10 years ago

Comment on attachment 8505342 [details] [diff] [review]
patch

Note that this pref was added ages ago for testing purposes. Then people started using it to prevent notifications. There is every chance in the world that there are cases where it does notify. There is also nothing we can do to notify them of an update if it still works and they need to opt-in to an update due to extension compatibility, etc.

	Robert Strong (they/them - no direct email)
	Comment 4 • 10 years ago

You might consider changing the amount of time until the notification to restart is shown instead. You could notify after a week or two instead.

	Alexandre Poirot [:ochameau] Assignee
	Comment 5 • 10 years ago

Panos, I think you want to know about the last two comments.
app.update.silent may not be the best shot for silent updates...

In the meantime, I think we can land this patch that will make the test just more robust.

	Robert Strong (they/them - no direct email)
	Comment 6 • 10 years ago

If you want to feel free to schedule some time with me to discuss the options available to you.

	Panos Astithas (he/him) [:past] (please ni?)
	Comment 7 • 10 years ago

Thanks, I didn't know about any of this. In bug 1080406 I'm trying to use a less intrusive mechanism to notify about updates, and in my testing I haven't actually seen any update notifications on nightly with the pref enabled.

Joe, could we schedule a meeting with Rob and anyone else who would be interested?

	Carsten Book [:Tomcat]
	Comment 8 • 10 years ago

https://hg.mozilla.org/integration/fx-team/rev/074a7010a1ad

	Carsten Book [:Tomcat]
	Comment 9 • 10 years ago

https://hg.mozilla.org/mozilla-central/rev/074a7010a1ad

	Panos Astithas (he/him) [:past] (please ni?)
	Comment 10 • 10 years ago

Comment on attachment 8505342 [details] [diff] [review]
patch

Approval Request Comment
[Feature/regressing bug #]: it's a test improvement that will be required once we uplift some other patches that will make the failure permanent
[User impact if declined]: none, only test failures
[Describe test coverage new/current, TBPL]: existing test
[Risks and why]: none, it's a test-only change
[String/UUID change made/needed]: none

	Robert Strong (they/them - no direct email)
	Comment 11 • 10 years ago

Comment on attachment 8505342 [details] [diff] [review]
patch

Test only changes to aurora can land as follows
https://wiki.mozilla.org/Tree_Rules
"Exception: If patches only make changes to tests, test harnesses or anything else that does not affect the shipped builds, they may land with self approval (use a=testonly, a=npotb etc)."

	Panos Astithas (he/him) [:past] (please ni?)
	Comment 12 • 10 years ago

https://hg.mozilla.org/releases/mozilla-aurora/rev/a1366e157fb1

	Panos Astithas (he/him) [:past] (please ni?)
	Comment 13 • 10 years ago

Thanks for the cluebat Rob!