Last Modified: 2003-10-20
Routing protocols for VoIP allow intermediaries, such as SIP proxies and H.323 gatekeepers, to make call routing decisions based on reachability information learned from peer elements. The iptel group has already defined a protocol, Telephony Routing over IP (TRIP), RFC 3219, which solves one aspect of this problem. Specifically, it handles the case where calls need to be routed between domains. It allows for the exchange of routing information between these providers, so that policies can be applied to the resulting data to create a forwarding information base.
However, this protocol does not address all the scenarios of route information exchange between servers. One important scenario is the propagation of routing information between gateways and the signaling servers in front of them. This is also known as "gateway registration". It allows the signaling server to make a routing decision about which gateway to use based on dynamic information about the gateway resources. Vendors have deployed proprietary solutions for this communications interface. A standard is needed. The group will generate a standards track document that defines a protocol (possibly based on TRIP) for this interface.
TRIP and the gateway registration protocol are orthogonal to the DNS-based mechanisms specified in ENUM and RFC 3264. Those mechanisms are used to translate a URI, representing a name, to an address. If that address is a phone number in the telephone network, trip and tgrep can be used to assist in determining the right route (through various gateways) to that number.
The group will also generate a MIB document for TRIP.
Note that the group is not working on elevating TRIP to Draft Standard at this time.
Deliverables:
1. A proposed standard specification for gateway to server route exchange.
2. A proposed standard TRIP MIB specification, based heavily on the existing BGP-4 MIB documents.
3. A standards track update to the tel URI.
4. Standards track extensions to the tel URI for PSTN interoperability, such as number portability and trunk group identification.
| Done | Submit gateway location framework document to IESG for consideration as an RFC. | |
| Done | Submit call processing syntax framework document to IESG for consideration as an RFC. | |
| Done | Submit call processing syntax document to IESG for consideration as a Proposed Standard. | |
| Done | Submit gateway location protocol document to IESG for consideration as an RFC. | |
| Done | TRIP MIB Document submitted to IESG for consideration as proposed standard | |
| Jun 03 | Gateway to Server Route Exchange document submitted to IESG for consideration as proposed standard. | |
| Jul 03 | Tel URI revisions submitted to IESG | |
| Sep 03 | Number portability extensions submitted to IESG for consideration as proposed standard | |
| Nov 03 | Consider closing |
| RFC | Status | Title |
|---|---|---|
| RFC2824 | I | Call Processing Language Framework and Requirements |
| RFC2871 | I | A Framework for a Gateway Location Protocol |
| RFC3219 | PS | Telephony Routing over IP (TRIP) |
|
IETF 58 IPTEL Meeting Minutes Reported by Tom Taylor <taylor@nortelnetworks.com> The IPTEL Working Group met on Wednesday morning, 12 Novemeber. The meeting was chaired by Cullen Jennings <fluffy@cisco.com> and Jonathan Rosenberg <jdrosen@dynamicsoft.com>. The agenda was accepted as proposed:
StatusCullen Jennings presented.CPL (draft-ietf-iptel-cpl-08.txt) The draft was reworked to remove dependencies on caller prefs. The XML basis was changed to Schema from DTD. WG Last Call is complete. Will send to IESG for reinsertion into the RFC Editor's queue. RFC2806bis (draft-ietf-iptel-rfc2806bis-02.txt) The draft has completed WG Last Call. It will be submitted to the IESG unless the discussion regarding the user=phone parameter leads to changes. TGREP (draft-ietf-iptel-tgrep-02.txt) The draft is ready for final NITS review, WG Last Call. TRIP MIB (draft-ietf-iptel-trip-mib-09.txt) After lots of discussion back and forth between the editors and the MIB doctor, another revision has come out. The document is now ready for sending to the IESG. Other stuff: Calling Party's Category tel: URI parameter (draft-mahy-iptel-cpc-00.txt) tel: URI parameters (draft-yu-tel-url-08.txt) trunk group (draft-ietf-iptel-trunk-group-00.txt) Need expressions of interest to take these forward. Tom Taylor mentioned that the Yu and Mahy drafts were of particular interest to those working on interworking of SIP to the PSTN. Jonathan responded that the question is whether the proposed functionality is needed in the SIP network. Flemming Andreasen <fandreas@cisco.com> and Richard Stastny <Richard.Stastny@oefeg.at> indicated their interest in those drafts. Another speaker indicated his interest in the trunk group draft. Cullen said he would contact James Yu to see if he was still interested in carrying his draft forward. Jon Peterson indicated that James was still interested, but had some concern on the content of 2806bis. Jonathan emphasized that he was only prepared to add these drafts as WG items if he sees interest on the list. He asked for list comments and/or document revisions by mid-December, or we would not continue. A milestones update slide was presented. The chairs are looking to close the WG by March. There was no comment on the proposed milestones. user=phone discussionRohan Mahy <rohan@cisco.com> presented the issue based on slides worked out between him and Brian Rosen <Brian.Rosen@marconi.com>. These slides reflected the results of discussion on the SIPPING list, also captured by Rohan's draft (draft-mahy-sipping-user-equals-phone-00.txt). There is just one major point of disagreement. The points of agreement:
Disagreement: how to distinguish "dial strings" from phone numbers. The subsequent discussion was lively and wide ranging. The following notes capture a number of the points, but not all of them. Tom Taylor noted that number transformations occur at various points along the call path, not just at the first hop. Brian Rosen pointed out that user=phone is broken no matter what, and his proposed fix is one way to solve the problem. Two speakers expressed the view that interoperability is more likely to be achieved if the approach taken was to use separate phone contexts for telephone numbers and dial strings. Francois Audet <audet@nortelnetworks.com> asked why the proposed dialstring syntax used E and F rather than * and #. It appears that 2806bis doesn't allow * and #. Henning Schulzrinne <hgs@cs.columbia.edu> expressed a preference for explicit semantics hence, in this case, allowing the user=dialstring parameter. Someone remarked that if you don't want to put phone-context in, then you shouldn't set user=phone. Jonathan Rosenberg wondered how developer could distinguish a telephone number from a dial string if the IETF cannot. Context defines a name space. That should be adequate. Dean Willis <dean.willis@softarmor.com> made the point that "9,411" is actually a request for a gateway to provide a specific service -- outpulsing according to a specific cadence. This is a specific instance of the general class of service requests. Brian Rosen picked up Dean's point and elaborated on it. He has a further need: to provide guidance to a proxy server that serves multiple domains. Henning said this provided further support for user=dialstring; it is a parameter pointing to a specific operation. Richard Stastny added his support for user=dialstring because phones cannot dial E.164 numbers. Dick Knight <dick.rr.knight@bt.com> pointed out that this was effectively pushing the distinction problem back to the UA. Another person pointed out we already have two mechanisms to transmit the requirement to pause: KPML and 2833. We don't need another. Tom Taylor repeated his point that the network may need to do number translation beyond the first hop -- hence it is desirable to have a general mechanism to indicate that the user part is in a context-specific form. Richard Stastny agreed that digit manipulation is definitely needed at the first hop, but shouldn't be needed thereafter. Brian Rosen explained why KPML wouldn't work. It is easier to get user=dialstring to work. Francois Audet expressed his view that allowing user=dialstring not a big deal. Pause is a different discussion. At this point the chairs used a series of hums to determine the general opinion of the meeting.
Brian Rosen volunteered to write a draft defining user=dialstring. This will be an individual contribution to SIPPING, to be considered at the request of IPTEL. There was a split view on whether getting pauses to the gateway is a problem. The discussion will be taken to the list. |