2.4.8 Ethernet Interfaces and Hub MIB (hubmib)

NOTE: This charter is a snapshot of the 51st IETF Meeting in London, England. It may now be out-of-date. Last Modified: 31-Jul-01


Dan Romascanu <dromasca@avaya.com>

Operations and Management Area Director(s):

Randy Bush <randy@psg.com>
Bert Wijnen <bwijnen@lucent.com>

Operations and Management Area Advisor:

Bert Wijnen <bwijnen@lucent.com>

Mailing Lists:

General Discussion:hubmib@ietf.org
To Subscribe: hubmib-request@ietf.org
In Body: subscribe your_email_address
Archive: http://www.ietf.org/mail-archive/working-groups/hubmib/current/maillis

Description of Working Group:

The Ethernet Interfaces and Hub MIB WG is Chartered to define a set of managed objects that instrument devices, MAUs and interfaces that conform to the IEEE 802.3 standard for Ethernet. This set of objects should be largely compliant with, and even draw from IEEE 802.3, although there is no requirement that any specific object be present or absent. The MIB object definitions produced will be for use by SNMP and will be adequately consistent with other SNMP objects, standards and conventions.

The WG will define new MIB objects to cover the following 802.3 technologies:

P802.3ae - 10 Gb/s Ethernet P802.3af - DTE Power via MDI

The documents structure and their relationship with the current WG documents is to be decided

Goals and Milestones:



Meet at the 41st IETF to discuss implementation experience of RFC 2108 and RFC 2239, and to consider future extensions for Full Duplex operation and 1 Gigabit Ethernet Speeds



Gather implementation experience feedback concerning RFC 2108 and RFC 2239



Post Internet-Draft(s) for Full Duplex and 1 Gigabit Ethernet MIB extensions



Meet at the 42nd IETF to discuss the Internet-Draft(s) and issue recomendations concerning advancement of RFC 2108 and RFC 2239 on the standards track



Post revised Internet-Draft(s)



Conduct WG Last Call on Internet-Draft(s)



Submit final version of the Internet-Draft(s) to the IESG for consideration as Proposed Standards



Submit revised version of the Internet-Drafts, following the Area Directorate review



Submit final versions of the MAU MIB and Ethernet-like Interfaces MIB Internet-Draft(s) to the IESG for consideration as Proposed Standards



Submit the Ethernet Chipsets document to IESG for consideration as an Informational RFC

May 01


Begin identifying new work items for future work

Jun 01


Issue WG Drafts for MIBs for P802.3ae & P802.3af

Jul 01


Gather implementation experience concerning WG documents already on the standards track

Oct 01


Issue revised WG Drafts for MIBs for P802.3ae/P802.3af

Oct 01


Issue revised WG drafts for existing stds track documents if so required by the implementation reports

Nov 01


WG Last Call All documents

Jan 02


Forward Internet-Drafts to the AD/IESG Stds track considerations

Request For Comments:






Definitions of Managed Objects for IEEE 802.3 Medium Attachment Units (MAUs)



Definitions of Managed Objects for IEEE 802.3 Repeater Devices using SMIv2



Definitions of Managed Objects for the Ethernet-like Interface Types



Definitions of Object Identifiers for Identifying Ethernet Chip Sets



Definitions of Managed Objects for IEEE 802.3 Medium Attachment Units (MAUs) using SMIv2

Current Meeting Report

51st IETF Hubmib Working group London meeting minutes Thursday 9 Aug, 2001

1. Agenda by Dan

2. WIS MIB status by Dan +IBM-

a. Joint session held on Tuesday

b. Design team put in place from editors from both working groups and drafts

c. Three major issues:

3. IEEE 802.3 Status by Dan

a. One plenary and one interim meetings have happened since last IETF

b. 802.3ae on schedule (Last editing ends by Sep this year)

i. Last chance to provide input either via your company

or via Dan through the IEEE voting process

ii. Standard expected to be done by March 2002

c. 802.3af is late (more than one year+ACE-)

i. draft 2.0 9 2001

ii. will have an impact on our schedule

d. 802.3ah (Ethernet For the First Mile) PAR approved

i. Includes EPON, Ethernet over VDSL and point to point fiber

ii. Expect to take 2 years at IEEE

iii. Expect to take 1 year to come to IETF for MIB definition

iv. Maybe a tutorial in the next IETF meeting based on the interest on the list

e. Request from IEEE 802.3 WG chair +IBM-

i. Cooperation between two organizations

ii. If any objects needed to be added, this working group will let IEEE know and to consider

iii. Questions raised regarding IETF +IBM- IEEE process and Dan/Bert clarifies:

1. Use DAN to voice any technical issue to expedite the changes that might arise from this working group

2. This has actually not happened before

iv. Questions raised regarding timing between two groups +IBM- chicken and egg problem

1. Change so that they consider to add them to the draft standard to +IBw-to their draft standard+IB0-

v. The proposal for additional text in the Charter will be taken by Bert to the IESG for approval'

4. Working Document Status by Dan

a. Power mib first draft out

b. Changes to the Ethernet interface/MAU mib draft 3 out

c. WIS mib first draft out

5. Implementation report status by Yasha

a. Not present at the meeting (but sent a representative)

b. Yasha has not received any implementation experience from the working group

c. Dan suggested sending Yasha+IBk-s email address again and solicit for report from the mailing list

d. What are the issues and what is not implemented?

e. About 3 +AL0- vendors in the meeting can provide implementation report

6. Ethernet/MAU MIB status by John Flick

a. Added 64 bit counters +IBM- wrap up time is specified in the DESCRIPTION (most of them can roll over within 5 minutes)

b. IfMtu +IBM- larger frame sizes not just IEEE 1500 +IBM- call for clarification in the DESCRIPTION that this frame size is viewed from the network layer

c. 802.3ad does provide ifStackTable mapping. This needs work in the draft.

d. Open Issues for Ethernet MIB:

i. StretchRatio is read-writable in 802.3 draft.

This is for debug and detecting WIS interface.

(The latter is supported by other object already).

+IBM- leave it open and take it to the list.

ii. Which correct value for ifHighSpeed?

1. should be 9294 Mb/s since it is viewed from the PCS layer.

iii. Loopback test needs to be added to this MIB since ifTestTable is obsoleted

iv. IF-MIB packet/octet counter mappings +IBM-

1. Outbound discards are counted twice

2. Inbound discards are counted once

3. PAUSE frame is counted in OCTETs but not frames

4. Dan suggested posting the diagram (by John) to this working group as well as IEEE and call for clarification

e. Test table will be added to the Ethernet MIB not MAU MIB

f. Open Issues for MAU MIB:

i. The auto-configuration for speed, duplex and pause are too complex. Suggestion by John to use a single object that enum all the combinations.

ii. Dan agreed that if most people implemented the MIB this way then we should adopt the change and deprecate the old objects

g. Next draft is pending implementation report and WIS MIB draft (2 months)

7. Power Ethernet mib by Avi Berger

a. Deliver power over RJ45

b. For smaller device (no laptop yet+ACE-)

c. Question on how to provide PowerCDlassification object +IBM- hardware detects the chip and power class defined in IEEE

d. In overCurrent condition, let go of the lower priority line and maintain the regular ones.

e. LowPower and power saving concern - raise it to the working group

f. IEEE will finish the draft by end of this month (1.3)

g. New IETF draft will be based on the new IEEE document

8. 802.17 new standard +IBM- do we do it in this working group?

a. A tutorial was presented in the iporpr WG

b. IPRPR claims it is not Ethernet

c. Hence it is not done in this working group

9. Close

Joint Session (hubmib & atommib)

HUBMIB (Ether)/AToMMIB Joint Session
chaired by Dan Romascanu <dromasca@avaya.com> & Nathan Kohn <mvnk@lucent.com>

Minutes by John Flick & Nathan Kohn

Attendance about 20, mostly hubmib.

Objective: Discuss cooperated efforts in defining a MIB that will cover 802.3ae WAN interface sublayer. Data rate and format compatible with SONET STS-192c.

- Discuss the potential for next revision of RFC 2558 to cover
- Or a new MIB that will cover both SONET and 10 Gbs Ethernet interfaces using SONET

Current internet draft: draft-ietf-hubmib-wis-mib-00.txt

This draft generated several comments both from SONET community and IEEE. Dan presented both the IEEE and HUBMIB status and schedules. IEEE began work in January 2000, scheduled to finish work in March, 2002. Ethernet MIB WG would like to be ready with MIBS at roughly the same time as IEEE finishes its standard. Dan is a voting memeber of IEEE and feels that work can be done effeciently in parallel.

IEEE Status:
- Draft D3.1 published, draft D3.2 to be published in the next few weeks
- Dan can provide passwords to our WG members to access the IEEE drafts
- Next ballot with all 802 members
- Content quite stable, reduced number of open issues
- Dan met with IEEE folks at last plenary that resulted in a set of inputs to the hubmib

Independently, Mike Heard and Kaj Tesnik have issued an alternate individual internet-draft for WIS. Unfortunately, neither was able to attend this meeting.

Dan asked how many had read WIS draft - very few hands went up

Mike Ayers reviews current draft - current draft was primarily to get comments. WIS is somewhat of a challenge because it has its feet in both the LAN world and the WAN (SONET) world. Many of the things in the SONET MIB were just copied into the WIS MIB. Draft was based on IEEE D3.0, did not include updates from D3.1.

Two major questions:

- If this is manageable as both Ethernet and SONET, how do we arrange the MIB to be usable by both Ethernet and SONET management systems simultaneously, without confusing either. WIS is similar, but not exactly SONET.

- IfStack usage: one proposal is etherCsmacd/sonetPath/sonet: would we confuse SONET managers if we did this, since you cannot configure multiple paths in WIS?

- Several questions about specific management objects; example, is the SONET object SONET-SES-OTHER-SET, which allows four different ways of collecting stats. The question was whether it was prescriptive or descriptive. After discussion it was decided that it is descriptive. However we need to address the top-level question of how we represent the interface first

Need input from management app developers of how various representations would affect them.

Current document does not provide mapping to IEEE objects. Dan asked if we had gone thru the exercise of mapping between the IEEE clause 30 objects and 2558 objects. Mike answered that we need to understand the differences between the IEEE objects and SONET objects. An example is the reset and power options - are they only applicable for the section? Should these objects only be for WIS? Another

Dan summarized 3 main issues:

- Need to describe our interface model, what the stack layerslooks like, and how it is managed.

- Need to map IEEE clause 30 objects to SONET and/or WIS objects. Are all necessary, are any missing?

- How do we reconcile counter models used by Ethernet world and SONET world. How are the statistics for counters veiwed from the higher layers?

Need input from SONET folks on this. Particularly on the future direction for SONET MIBs, i.e. provisioning.

Can we put together a design team with current WIS editors along with Mike and Kaj to have them come up with answers to these three major issues.

Chairs of hubmib and atommib will put together list of questions for design team, which will consist of current hubmib editors, Mike and Kaj who submitted counter-proposal. Kam Lam (co-author of the optical mib in AToMMIB) also volunteered for design team. This will be a joint design team which will feed back to both WGs. HUBMIB and AToMMIB are both meeting on thursday and will discuss the proposal within each WG.

Here are the proposed questions (supplied by Dan after the meeting):


Ethernet Interfaces and Hub MIB WG and AToM MIB WG Joint Meeting
Ethernet/MAU MIB Status
IEEE 802.3ae Status of WIS MIB Work