1884140 - Enable DTLS1.3 in WebRTC for Firefox Release

Status: RESOLVED FIXED

(Core :: WebRTC, enhancement, P3)

Description

[Anna Weine]

Hi,

We (NSS) have recently improved the support of DTLS1.3. The DTLS1.3 code is now in Release.
What do you think about enabling DTLS1.3 in WebRTC for Release? Does it sound like it's worth doing? Any problems you expect?

Thanks

P.s. as far as I can see, we only need to change the pref: https://searchfox.org/mozilla-central/source/modules/libpref/init/all.js#342

Comment 1

[Byron Campen [:bwc]]

I think that would be reasonable to do. Martin, do you feel like it is time?

Comment 2

[Martin Thomson [:mt:]]

I don't see any reason not to. This isn't without risk, but as long as Anna is comfortable helping with any tricky protocol-level diagnosis, offering this will put us in a strong position.

We likely won't see a lot of performance gain - there aren't that many direct Firefox-to-Firefox sessions. Most are terminated at servers.

Before doing this, I recommend re-activating the WEBRTC_DTLS_PROTOCOL_VERSION telemetry probe and maybe WEBRTC_DTLS_{CLIENT|SERVER}_FAILURE_TIME (a failure reason would be even better (see SSL_HANDSHAKE_RESULT). We let these probes expire, but having information about how adoption is going would seem to be wise.

Comment 3

[Byron Campen [:bwc]]

Any reason not to convert those telemetry probes to Glean? I don't see us wanting to compare the long expired telemetry with the newer version.

Comment 4

[Martin Thomson [:mt:]]

By all means use Glean, if that is easier or better. Comparing with old behaviour isn't that useful, I agree.

Comment 5

[Anna Weine]

I will be happy to help if anything goes wrong :)

Comment 6

[Anna Weine]

Hi, do you have any update on this?

Comment 7

[Byron Campen [:bwc]]

Should we go ahead and remove expired telemetry like WEBRTC_SRTP_CIPHER and WEBRTC_DTLS_CIPHER? Or start gathering that again with glean?

Comment 8

[Martin Thomson [:mt:]]

Removal seems sensible. As to whether you want to continue tracking it, that's up to you. I would. I've found that having this data is critical to understanding what to do with new ciphers (and it looks like there are some new ciphers coming).

Comment 9

[Byron Campen [:bwc]]

Ok, new ciphers coming up definitely justifies some telemetry, I'll go ahead and add that in.

Comment 10

[Byron Campen [:bwc]]

Attachment: Bug 1884140: Glean telemetry for webrtc DTLS. r?mjf

Comment 11

[Byron Campen [:bwc]]

Attachment: Bug 1884140: Allow the use of DTLS 1.3 on release/beta. r?mjf

Depends on D208147

Comment 12

[Byron Campen [:bwc]]

Attachment: data-review-1884140.md

Comment 13

[Byron Campen [:bwc]]

Try looks fine:

https://treeherder.mozilla.org/jobs?repo=try&revision=474020c625bbbd08fa35f6d0eb2dbac5b3bbbbd3
https://treeherder.mozilla.org/jobs?repo=try&revision=b26873c2b15ee173f05dc702bfba57d533c0ee00

Comment 14

[Chris H-C :chutten]

Comment on attachment 9397907 [details]
data-review-1884140.md

DATA COLLECTION REVIEW RESPONSE:

Is there or will there be documentation that describes the schema for the ultimate data set available publicly, complete and accurate?

Yes.

Is there a control mechanism that allows the user to turn the data collection on and off?

Yes. This collection can be controlled through the product's preferences.

If the request is for permanent data collection, is there someone who will monitor the data over time?

No. This collection will expire in Firefox 135.

Using the category system of data types on the Mozilla wiki, what collection type of data do the requested measurements fall under?

Category 1, Technical.

Is the data collection request for default-on or default-off?

Default on for all channels.

Does the instrumentation include the addition of any new identifiers?

No.

Is the data collection covered by the existing Firefox privacy notice?

Yes.

Does the data collection use a third-party collection tool?

No.

---

Result: datareview+

Comment 15

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/d09724a12a23
Glean telemetry for webrtc DTLS. r=mjf
https://hg.mozilla.org/integration/autoland/rev/bd103d0b429a
Allow the use of DTLS 1.3 on release/beta. r=mjf

Comment 16

[Cristian Tuns]

https://hg.mozilla.org/mozilla-central/rev/d09724a12a23
https://hg.mozilla.org/mozilla-central/rev/bd103d0b429a

Comment 17

[Pascal Chevrel:pascalc]

Is there a user impact that should translate into mentioning that change in our general, developer, or enterprise release notes? Thanks

Comment 18

[Byron Campen [:bwc]]

I think it is worth mentioning that webrtc will be using the latest version of the DTLS spec for encryption.

Comment 19

[Anna Weine]

Attachment: Bug 1884140 - Adding WebRTC DTLS1.3 to FeatureManifest

Comment 20

[Phabricator Automation]

Comment on attachment 9400444 [details]
Bug 1884140 - Adding WebRTC DTLS1.3 to FeatureManifest

Revision D209674 was moved to bug 1895498. Setting attachment 9400444 [details] to obsolete.

Comment 21

[Phabricator Automation]

Comment on attachment 9400444 [details]
Bug 1884140 - Adding WebRTC DTLS1.3 to FeatureManifest

Revision D209674 was moved to bug 1895498. Setting attachment 9400444 [details] to obsolete.

Comment 22

[Phabricator Automation]

Comment on attachment 9400444 [details]
Bug 1884140 - Adding WebRTC DTLS1.3 to FeatureManifest

Revision D209674 was moved to bug 1895498. Setting attachment 9400444 [details] to obsolete.