[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [P2PSIP] New versions of RELOAD and sip draft

Michael,

1:
I was actually working in that area ATM. There is an exclusion "Only present for type "relay"", however I'm not certain as to why that is. I agree it MUST be as you've stated . 

2:
There also must be a codes for app_attachlite_req = 29 and app_attachlite_ans = 30. 

3:

No comment.

4:
Seeing that ICE is required for RELOAD as is TLS over TCP I cannot think of any other protocol off the top of my head that is more dependent on that draft than this one. 

Julian

On Oct 12, 2009, at 6:48 PM, Michael Chen wrote:


Cullen,
I posted several issues regarding the -03 draft late July but received no reply. I don't see changes related to these issues in the new -04 draft. I want to make sure they are not overlooked: 

 http://www.ietf.org/mail-archive/web/p2psip/current/msg05019.html
1) Section 5.5.1.1, the definition IceCandidate.rel_addr_port does not conform with section 15.1 of the current mmusic-ice-19 draft, which has the following text: 

"<rel-addr> and <rel-port>:  convey transport addresses related to the
     candidate, useful for diagnostics and other purposes. <rel-addr>
     and <rel-port> MUST be present for server reflexive, peer
     reflexive and relayed candidates."

Therefore, the definition of IceCandidate should be:

        struct {
          ...
          select (type){
            case host:
              ;          /* Nothing */
            case srflx:
            case prflx:
            case relay:
              IpAddressPort     rel_addr_port;
          }
          ...
        } IceCandidate;

  rel_addr_port
     corresponds to the related address and port of ICE, which MUST
     present for type "srflx", "prflx" and "relay".
2) In Section 13.6, IANA message codes for AppAttachReq and AppAttachReq are not added. I suggest fill these two values (app_attach_req=5, app_attach_ans=6) in the holes left by the abandoned tunnel_req and tunnel_ans.
3) In Section 6.4.1.2, if the store request has multiple kinds, the current answer structure will send replica node IDs multiple times. For efficiency, I suggest moving the replica node IDs to the outer layer as: 

        struct {
          KindId                  kind;
          uint64                  generation_counter;
        } StoreKindResponse;

        struct {
          StoreKindResponse       kind_responses<0..216-1>;
          NodeId                  replicas<0..216-1>;
        } StoreAns;
4) It seems that mmusic-ice-tcp-07 has been abandoned. The key members of this draft should address the lack of ice-tcp standard in the next draft, meeting or on the mailing list. 

That is it so far. Thanks

--Michael

Cullen Jennings wrote:


I just submitted
draft-ietf-p2psip-base-04
and
draft-ietf-p2psip-sip-02
These contain technical changes but do not have a lot of editorial changes. At some point we need to go and reorganize the documents with editorial changes. 

Cullen <in my individual contributor role>

_______________________________________________
P2PSIP mailing list
P2PSIP at ietf.org
https://www.ietf.org/mailman/listinfo/p2psip



_______________________________________________
P2PSIP mailing list
P2PSIP at ietf.org
https://www.ietf.org/mailman/listinfo/p2psip