1113790 - pref change to true for network.http.altsvc.enabled and altsvc.oe

Status: RESOLVED FIXED

(Core :: Networking: HTTP, defect)

Description

[Patrick McManus [:mcmanus]]

The current status of these prefs:

#ifdef RELEASE_BUILD
pref("network.http.altsvc.enabled", false);
pref("network.http.altsvc.oe", false);
#else
pref("network.http.altsvc.enabled", true);
pref("network.http.altsvc.oe", true);
#endif
	
The intention has always been to enable them one release after we ship h2 at proposed standard level (i.e. not a draft).

gecko-36 is the first release to have enabled advertisement for h2 at non-draft status - therefore the alt svc changes should be done in 37.

Comment 1

[Patrick McManus [:mcmanus]]

Attachment: pref change to true for altsvc

Comment 2

[u408661]

Comment on attachment 8539476 [details] [diff] [review]
pref change to true for altsvc

Review of attachment 8539476 [details] [diff] [review]:
-----------------------------------------------------------------

<kool-aid-man>OH YEAH!</kool-aid-man>

Comment 3

[Patrick McManus [:mcmanus]]

https://hg.mozilla.org/integration/mozilla-inbound/rev/f7bd120de6e0

Comment 4

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/f7bd120de6e0

Comment 5

[Florian Bender]

Release Note Request (optional, but appreciated)
[Why is this notable]: AltSvc enables so-called opportunistic (unauthenticated) encryption
[Suggested wording]: Implemented HTTP/2 AltSvc support
[Links (documentation, blog post, etc)]: (:hurley?)

Comment 6

[u408661]

(In reply to Florian Bender from comment #5)
> [Links (documentation, blog post, etc)]: (:hurley?)

I suspect mcmanus has something in the works (or will soon), since he's been the one writing all the code for this.

Comment 7

[Lawrence Mandel [:lmandel] (use needinfo)]

relnoted as "New: Implemented HTTP/2 AltSvc support for opportunistic (unauthenticated) encryption"

mcmanus/hurley - I didn't find any Mozilla specific docs with a quick search. I also see that bug  	1003448 is still marked dev-doc-needed. Do you know of any doc that I can link to that further explains this feature?

Comment 8

[u408661]

I suppose, unless mcmanus knows of a better doc somewhere, the I-D might be better than nothing? (http://tools.ietf.org/html/draft-ietf-httpbis-alt-svc-04 is the version implemented in 37)

Comment 9

[Patrick McManus [:mcmanus]]

comment 8 is good. I'll also blog about this next week

Comment 10

[Lawrence Mandel [:lmandel] (use needinfo)]

I don't think the link in comment 8 is good for the release notes. It's simply too detailed for the audience. We need something that is more consumable starting with a bit of a lay description of the change. 

Patrick - Can you include a high level description of the feature in your blog post? If so, let's wait on the link until you have the post up.

Comment 11

[Patrick McManus [:mcmanus]]

https://bitsup.blogspot.com/2015/03/opportunistic-encryption-for-firefox.html