Re: [dnssd] WGLC for draft-ietf-dnssd-push

[Stuart Cheshire <[e-mail]>]

On 15 Apr 2019, at 10:07, David Schinazi <[e-mail]> wrote:

> This email starts another Working Group Last Call (WGLC) for draft-ietf-dnssd-push, which will last for two weeks until 2019-04-29. As a reminder, this document has already passed WGLC but some slight changes to the document warranted another WGLC.

Thank you David.

I have attached an HTML file of the output from <https://tools.ietf.org/rfcdiff> comparing draft-ietf-dnssd-push-16 with draft-ietf-dnssd-push-19.

There are a variety of minor textual changes, such as replacing references to draft-ietf-dnsop-session-signal with references to RFC 8490, now that this document is published.

In terms of substantive changes, there are two:

1. DNS Push Notification RECONFIRM

We realized that there is no useful response to a RECONFIRM request. We changed RECONFIRM to be a unidirectional DSO message, and that results in the deletion of the entire “RECONFIRM Response” section. This simplifies both code, and the document.

2. DNS Push Notification Updates

Previously, Tom Pusateri and I thought that borrowing the DNS Update encoding made sense. That was before DSO (RFC 8490).

When implementing the current specification of DNS Push Notifications over DSO, Ted Lemon discovered that it has no way to express the DNS CLASS of changes.

We need to address the problem Ted Lemon discovered, and now that DNS Push uses DSO TLV encoding instead of traditional DNS messages, the incentive to follow strictly the existing DNS Update encoding has disappeared.

To address this issue, we adopted a new, simpler, way of expressing changes. It’s broadly similar, but simpler.

Please take a look at these changes and let us know if you see any problems.

And if you *don’t* see any problems with these changes, please respond saying that too.

Stuart Cheshire