IEEE 802/IETF Liaison Report
IEEE 802/IETF Relationship Document
An updated version of the document has been posted to the IETF archive:
Feedback has been requested from the IESG & IAB, but so far, none has been received. Next step is to send the document to the IEEE 802 ExComm for review.
MIB Transfer Document
Dave Harrington has put together a rough draft of the MIB transfer document and there is ongoing discussion on how to refine it. The goal is to have a document that the IESG, IAB & IEEE 802 can review sometime in late September or early October.
One of the issues that has arisen is what rights the IETF has that it can transfer. A discussion will need to be arranged with Jorge Contreras (the IETF legal counsel) to understand this.
Since IETF 64 and the November IEEE 802 plenary are occurring in the same city (Vancouver, CA) one week apart, it has been proposed that this would be a good time to schedule a meeting to wrap up any loose ends relating to the MIB transfer document. Meeting times/dates TBD.
From Dan Romascanu:
The IEEE 802.1 WG is increasing the pace of the work for the MSTP MIB, which according to the draft agreement between the IETF and IEEE was started in the IETF Bridge MIB Working group and is being continued in IEEE 802.1, based on the initial IETF Internet-Drafts.
There has been discussion of whether the TRILL WG requires a liaison with IEEE 802.1. The TRILL WG does not depend on any IEEE 802.1 work-in-progress and IEEE 802.1 does not depend on any IETF work-in-progress, so it is not clear that a liaison is required. IEEE 802 provides free access to standards on the web via the “Get IEEE 802” process, so that a liaison is not required to provide TRILL WG participants with access to IEEE 802 standards.
A liaison letter was received from IEEE 802.16, thanking the IETF for its security review of IEEE 802.16eD8 and requesting a review of IEEE 802.16eD10, to see whether the issues found earlier have been addressed.
The TRILL WG Charter requires IEEE 802 review of several documents. At the IEEE 802 plenary in San Francisco in July, IEEE 802.1 Vice-Chair Paul Congdon lead a discussion on how to provide the requested review. This seems likely to be handled via formation of an adhoc group, similar to the way IEEE 802.11 handled the review of CAPWAP WG documents.
During the IESG review of draft-ietf-pwe3-ethernet-encap-10.txt, two IESG DISCUSS comments related to whether the document had been reviewed by IEEE 802. The PWE3 WG Charter did not require review by IEEE 802 and so that review had not been done. There is now ongoing discussion between the IESG, PWE3 WG Charter and document authors on whether IEEE 802 review is necessary, and if so, how it should be handled.
Jumbo Frame Discussion
At the IEEE 802 plenary, Alex Zinin and Bernard Aboba arranged a meeting with the IEEE 802.3 Chair (Bob Grow), IEEE 802 Vice-Chair (Pat Thaler) and interested participants to discuss how to develop a document on Jumbo Frames.
The overall conclusion was that such a document could be developed in the IETF, but needed to reflect a balanced perspective, given the large amount of work that has been done in IEEE 802.3 on the subject. In particular, such a document could draw upon previous analysis of frame error rates, CRC effectiveness, etc. so as to make clear the potential risks involved.
Minutes of the meeting are included below.
- Bernard Aboba (IETF/IEEE 802 liaison)
- Bob Grow (IEEE 802.3 chair)
- Pat Thaler (IEEE 802 vice chair)
- Shimon Muller (IEEE 802.3 contributor)
- Alex Zinin (IETF RTG Area director)
Opened the meeting by reviewing the history of IETF/IEEE 802 discussion wrt jumbo frames.
IEEE 802 folks explained the technical challenges they see with making jumbo frames a generic technology, specifically with legacy equipment.
IEEE 802 folks noted the 802.3as project that elevates the maximum MTU size for Ethernet to 2000 octets.
IEEE 802.3as is in Sponsor Ballot and is expected to be approved soon. The purpose of IEEE 802.3as is to allow space for protocols that add to the segment/frame after the original segmentation (e.g. 802.1x, IPSec tunneling) so that those protocols don’t need to fragment segments. Using the extra space to source bigger segments would defeat the purpose of the extension.
Alex & Bernard articulated reasons for the IETF’s interest in documenting jumbo frames: documentation of the de-facto mechanism to avoid problems due to misinterpretation in IP networks, outline of implementation considerations, identification of applicability constraints and operational requirements for deployment of the technology.
After some discussion, the sides agreed that it would be appropriate for the IETF to produce an informational document describing the above. Bob Grow agreed to help recruit a Technical Advisor for the IETF and it was proposed to organize an “adhoc group” within IEEE 802.3 in order to provide feedback on specific document(s), if requested by the IETF. Adhoc groups function under IEEE 802 rules and procedures so that the resulting review is voted on and issued as the opinion of IEEE 802.3.
- Bob Grow of IEEE 802.3 to send a short summary of our conversation to the IEEE 802.3 list and solicit interested people for the 802.3 “adhoc” group. Bob will also solicit a technical advisor for the IETF.
- IETF to decide on the organizational form of the effort (Design Team/ separate WG/existing WG). This would probably include WG charter modifications to appoint an IEEE 802 advisor and create a milestone for document development and review by IEEE 802.3.
- IETF to request comment from IEEE 802 on the proposed Charter change via the “New Work” mechanism (e.g. New Work announcement forwarded to the IEEE 802.1 & .3 mailing lists)
- IETF to send a liaison request to 802.3 requesting review of Jumbo Frame documents, as well as access to the IEEE 802.3as work in progress.