1304546 - (e10s-multi-aurora) [meta] Turn e10s-multi on in Aurora-DevEdition

Status: RESOLVED FIXED

(Core :: General, defect)

Description

[Blake Kaplan (:mrbkap) (inactive)]

This is a placeholder bug to turn e10s-multi on (for some number of processes to be determined at a later date) in Aurora.

Comment 1

[Gabor Krizsanits (INACTIVE)]

Attachment: Enable 4 content processes on aurora

Comment 2

[Blake Kaplan (:mrbkap) (inactive)]

Comment on attachment 8854045 [details] [diff] [review]
Enable 4 content processes on aurora

Approval Request Comment
[Feature/Bug causing the regression]: e10s-multi
[User impact if declined]: n/a
[Is this code covered by automated tests?]: Yes.
[Has the fix been verified in Nightly?]: Yes.
[Needs manual test from QE? If yes, steps to reproduce]: No.
[List of other uplifts needed for the feature/fix]: We're uplifting as we go.

Comment 3

[Blake Kaplan (:mrbkap) (inactive)]

Note to sheriffs: this change has no effect on Nightly (we've already enabled e10s-multi on Nightly).

Comment 4

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/mozilla-inbound/rev/5a0fce48c63f
Turning on e10s-multi in Aurora. r=mrbkap

Comment 5

[Wes Kocher (:KWierso) (Not reading bugmail; email directly if needed)]

https://hg.mozilla.org/mozilla-central/rev/5a0fce48c63f

Comment 6

[Liz Henry (:lizzard Please n-i to RyanVM, jcristau, or pascal)]

Tweaking the status flags so we won't lose the uplift request

Comment 7

[Gerry Chang [:gchang]]

Comment on attachment 8854045 [details] [diff] [review]
Enable 4 content processes on aurora

Enable e10s-multi on aurora and see how it goes. Aurora54+.

Comment 8

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/releases/mozilla-aurora/rev/29eab893bad0

Comment 9

[Carsten Book [:Tomcat]]

(In reply to Ryan VanderMeulen [:RyanVM] from comment #8)
> https://hg.mozilla.org/releases/mozilla-aurora/rev/29eab893bad0

it seems this is causing leaks like https://treeherder.mozilla.org/logviewer.html#?job_id=88700020&repo=mozilla-aurora 

i did some bisecting in https://hg.mozilla.org/try/rev/110f708df669971abf55dbab23056701e2fbb380 and that try run is entirely green.

Comment 10

[Carsten Book [:Tomcat]]

gabor, blake: since this is a very important change i guess i we either need to fix or back this change out from aurora

Comment 11

[Gabor Krizsanits (INACTIVE)]

I'm going to try to fix this in the test, but I wonder why this might be happening only on aurora and not on central...

Sotaro, I think you worked on a similar issue before in bug 1285238, do you know if there is any diversion between aurora and nightly that can cause this leak?

(In reply to Carsten Book [:Tomcat] from comment #9)
> (In reply to Ryan VanderMeulen [:RyanVM] from comment #8)
> > https://hg.mozilla.org/releases/mozilla-aurora/rev/29eab893bad0
> 
> it seems this is causing leaks like
> https://treeherder.mozilla.org/logviewer.html#?job_id=88700020&repo=mozilla-
> aurora 
> 
> i did some bisecting in
> https://hg.mozilla.org/try/rev/110f708df669971abf55dbab23056701e2fbb380 and
> that try run is entirely green.

Comment 12

[Gabor Krizsanits (INACTIVE)]

RyanVM|mtg> gabor, mrbkap: 
3:21 PM https://treeherder.mozilla.org/logviewer.html#?job_id=88717085&repo=mozilla-aurora is also happening frequently on Windows and it wouldn't shock me if it turned out to be the same root issue

Dão, are you aware of any diversion between aurora and nightly that could cause this failure?

Comment 13

[Dão Gottwald [::dao]]

(In reply to Gabor Krizsanits [:krizsa :gabor] from comment #12)
> RyanVM|mtg> gabor, mrbkap: 
> 3:21 PM
> https://treeherder.mozilla.org/logviewer.html#?job_id=88717085&repo=mozilla-
> aurora is also happening frequently on Windows and it wouldn't shock me if
> it turned out to be the same root issue
> 
> Dão, are you aware of any diversion between aurora and nightly that could
> cause this failure?

Nope... redirecting to Florian who wrote this test and the feature being tested.

Comment 14

[Ryan VanderMeulen [:RyanVM]]

Let's move the leak discussion over to bug 1347625, which I believe is the same root cause. Regarding the Windows failures in browser_devices_get_user_media_anim.js, I believe Gabor is still investigating.