Now that we've gone WGLC, let's turn this sucker on to get some real interop testing! This should probably only be on EARLY_BETA_OR_EARLIER

Now with less beta-ness

And yes, I will update the commit message before landing :)

https://tbpl.mozilla.org/?tree=Try&rev=6b5871d2e4d0 (just to make sure nothing goes crazy)

Comment on attachment 8466460 [details] [diff] [review] patch brave new world. checking after (or with) h2-14. and please don't have it hit a build until monday (so checkin on sunday at earliest) in case of badness. awesome.

Comment on attachment 8466460 [details] [diff] [review] patch we should get david's sign off on the alpn pref change..

remote: https://hg.mozilla.org/integration/mozilla-inbound/rev/72f85a52a2ca

[Tracking Requested - why for this release]: Nicholas, so, this is a preference that will only turn on when Firefox 34 is in beta? Is there anything in particular we should be aware of in QA for testing this - or just have a look at it when it goes to beta? Thanks!

I'm running the 20140805 (rev https://hg.mozilla.org/mozilla-central/rev/7f81be7db528 ) nightly, yet "network.http.spdy.enabled.http2draft" is still set to false ?

(In reply to Peja Stija from comment #10) > I'm running the 20140805 (rev > https://hg.mozilla.org/mozilla-central/rev/7f81be7db528 ) nightly, yet > "network.http.spdy.enabled.http2draft" is still set to false ? self service: https://hg.mozilla.org/mozilla-central/file/7f81be7db528/modules/libpref/src/init/all.js#l1139

(In reply to Liz Henry :lizzard from comment #9) > [Tracking Requested - why for this release]: > > Nicholas, so, this is a preference that will only turn on when Firefox 34 is > in beta? no - this pref is simply on. > Is there anything in particular we should be aware of in QA for > testing this - or just have a look at it when it goes to beta? Thanks! We'll be tracking interop opportunities - right now the only thing to be aware of are general regressions. I'm not sure why you marked this for tracking.

(In reply to Peja Stija from comment #10) > I'm running the 20140805 (rev > https://hg.mozilla.org/mozilla-central/rev/7f81be7db528 ) nightly, yet > "network.http.spdy.enabled.http2draft" is still set to false ? That's one day too early. The newest nightly (20140806, rev 6cbdd4d523a7) is the one you want.

Backed out by request, for causing bug 1050063. Nightlies respun. remote: https://hg.mozilla.org/mozilla-central/rev/2f198e81ed98

Here's the version that has been checked in, and must be checked in again...

Landed direct to m-c because it already landed just fine once except for the twitter issue: https://hg.mozilla.org/mozilla-central/rev/bd44d84142e8

So, for the record, this WILL make tonight's Nightly build.

(In reply to Wes Kocher (:KWierso) from comment #17) > So, for the record, this WILL make tonight's Nightly build. thanks ;) I feel like I'm in an on again off again relationship. interestingly twitter has deployed a server side fix instead of rolling back h2-14. Their implementation actually launched after this was enabled the first time. (perhaps before it hit the nightly build though - unsure on that.)

Release Note Request (optional, but appreciated) [Why is this notable]: The next era in Web tech has begun. Seriously. [Suggested wording]: Implemented HTTP/2 and APLN. [Links (documentation, blog post, etc)]: :hurley might have a blog post?

(In reply to Florian Bender from comment #19) > [Links (documentation, blog post, etc)]: :hurley might have a blog post? I don't (yet), though I'm planning on writing one. Perhaps I should get on that :)

Florian, what is APLN? The thing that I found is this bug. Nicholas, is this going to ship in 34? Thanks

(In reply to Sylvestre Ledru [:sylvestre] from comment #21) > Florian, what is APLN? The thing that I found is this bug. > alpn is a tls extension that allows negotiation of http/2. > Nicholas, is this going to ship in 34? Thanks http2-draft14 support will ship as part of 34. That's the same thing that IE announced in their developer preview this week. Support for the final http/2 will follow when the IETF has decided the format is final. If that that turns out to be the same thing as http2-draft14 then we will probably just uplift the final "http2" label to channels that already have working http2-draft14 support. hth

OK. Thanks for the info. So, since it is not a final standard, maybe we should write in the release note Implementation of HTTP/2 (draft14) and APLN

I guess my ni? is covered by Comment 22, then :)

Added to the release notes with "Implementation of HTTP/2 (draft14) and APLN" as wording. A blog post or documentation would be appreciated!

(In reply to Sylvestre Ledru [:sylvestre] from comment #25) > Added to the release notes with "Implementation of HTTP/2 (draft14) and > APLN" as wording. typo - you want ALPN

Sorry. Fixed. thanks!

So looking at Firefox 34 Beta 10 (BuildID=20141117202603), I see the following preferences set by default: network.http.spdy.enabled.http2draft = false security.ssl.enable_alpn = true Wasn't the http2 pref supposed to be true? Nicholas, am I missing something here?

see bug 1088910 - we held back http2draft after early_beta on 34 to give the standards process a little more breathing room and server operators a little more chance to mature their -14 implementations if they choose to offer them.

Thanks Patrick for clarifying this! Setting as "qe-verify-" since there doesn't seem to be anything for the QA team to do here.