[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: Proposed liaison response to IEEE
Thanks Loa.....this sounds OK to me and I'll go with the liaison as-is.
However, one point we should bear in mind and that has not been stated
so far is that the VLAN field assumes a different semantic in a co-ps
mode layer network (be this PBT or VLAN-XC swapping) compared to its
original cl-ps mode Ethernet layer network semantic. In the co-ps mode
case it is either:
- part of the trail termination address in PBT that does not get
swapped during forwarding, ie the trail sink term address is the
combination of DMAC+VCID;
- a proxy for the trail termination address in VLAN-XC (BTW - I am
not clear what VLAN-XC layer network addresses are here) that could be
swapped on each link-connection.
AFAI understand the original IEEE intent of a VLAN was to create a
restricted broadcast domain within which a certain subset of the total
MAC address population reside.
These are quite different semantics. So asking for clarification of
what IEEE mean by (and the scope of) swapping the VLAN field would still
not really be comparable to such an action in a co-ps mode layer
network.
regards Neil
> -----Original Message-----
> From: owner-ccamp@ops.ietf.org
> [mailto:owner-ccamp@ops.ietf.org] On Behalf Of Loa Andersson
> Sent: 21 September 2006 13:03
> Cc: gels@rtg.ietf.org; bwijnen@lucent.com; ccamp@ops.ietf.org
> Subject: Re: Proposed liaison response to IEEE
>
>
> All,
>
> there have been a number of correct statements in this short thread.
>
> Neil is right (given the layered network model) that Ethernet
> is a layer network and as such connectionless and packet oriented.
>
> Nurit is right that if you configure a VLAN on only to parts
> of an Ethernet bridge the frame will go through without
> looking at the MAC address
>
> Don is right that the liaison does not capture an earlier
> comment he made on the gels-list. Neither does it capture a
> comment made by Attila.
>
> However, if I understand what is happening here is that we
> need to clarify a paragraph in the liaison form the IEEE802.1
> on translation of S-VIDs.
>
> I guess that this is something we all need to understand, and
> given the answer, it will be possible to follow up with
> further more precise questions.
>
> If for example the answer on S-VID translation is that it is
> not supported other than at domain borders, then the question
> on whether it needs to be done by taking the MAC-address into
> consideration or not will have very different meaning.
>
> Let agree on getting the liaison out, and then decide on what
> we do next depending on the answer.
>
> /Loa
>
> Nurit Sprecher wrote:
> > Hi Neil,
> > Thanks for your clarification.
> > Please note that the discussion below relates to the
> liaison that the
> > IETF wants to send to the IEEE in order to understand the standard
> > behavior of the Ethernet forwarding behavior. As I
> understand it, Don
> > was concerned that S-VID translation does not preclude the
> forwarding
> > decision which is based on the MAC address. My intention
> was to indicate
> > that for point-to-point VLAN connection (on a link), MAC
> learning is not
> > required by the 802.1Qrev-5 spec and the forwarding
> decision does not
> > rely on the MAC address. This is not a discussion on
> networking but on
> > the standard behavior of the 802.1Q bridges.
> > Regards,
> > Nuritl.
> >
> > ________________________________
> >
> > From: neil.2.harrison@bt.com [mailto:neil.2.harrison@bt.com]
> > Sent: Thursday, September 21, 2006 13:00
> > To: Nurit Sprecher; dwfedyk@nortel.com; adrian@olddog.co.uk
> > Cc: gels@rtg.ietf.org; bwijnen@lucent.com; ccamp@ops.ietf.org
> > Subject: RE: Proposed liaison response to IEEE
> >
> >
> > Nurit.....precision is indeed important. In a co mode
> layer network
> > connections can be p2p or p2mp in nature and each can be
> composed of
> > multiple concatenated subnetwork-connections (ie nodes in THIS layer
> > network) and link-connections......noting that a single p2p
> > link-connection hop between 2 bridges in a cl-ps Ethernet layer
> > network is NOT a connection in THAT (ie Ethernet) layer network
> > (though it will be supported by a trail which is provided by a
> > connection in some lower layer network).
> >
> > regards, Neil
> >
> > -----Original Message-----
> > From: owner-ccamp@ops.ietf.org
> [mailto:owner-ccamp@ops.ietf.org] On
> > Behalf Of Nurit Sprecher
> > Sent: 21 September 2006 12:38
> > To: Don Fedyk; Adrian Farrel
> > Cc: gels@rtg.ietf.org; Wijnen, Bert (Bert);
> ccamp@ops.ietf.org; Nurit
> > Sprecher
> > Subject: RE: Proposed liaison response to IEEE
> >
> >
> >
> > Hi,
> >
> >
> >
> > I think it is a good idea to send the liaison to the
> IEEE and get
> > once and for all a clear statement that the s-vid translation is
> > supported on each provider bridge port.
> >
> >
> >
> > In addition, the IEEE has recognized that MAC learning is not
> > required for point-to-point VLANs in a bridge. The 802.1Q rev-5
> > already includes this extension in section 8.7.
> >
> >
> >
> > " The Learning Process receives the source MAC
> Addresses and VIDs of
> > received frames from the
> >
> > Forwarding Process, following the application of the
> ingress rules
> > (8.6.2). It shall create or update a
> >
> > Dynamic Filtering Entry (8.8.3) that specifies the
> reception Port for
> > the frame's source address and VID, if
> >
> > and only if the source address is an Individual
> Address, i.e., is not
> > a Group Address, the resulting number of
> >
> > entries would not exceed the capacity of the Filtering
> Database, and
> > the filtering utility criteria for the
> >
> > receiving Bridge Port are met."
> >
> > The enhanced filtering utility criteria "are satisfied
> if at least
> > one VLAN that uses the FID includes the reception Port and at least
> > one other Port with a Port State of Learning or Forwarding in its
> > member set, and: a) The operPointToPointMAC parameter is
> false for the
> > reception Port; or b) Ingress to the VLAN is permitted
> through a third
> > Port. NOTE -The third port can, but is not required to be in the
> > member set.".
> >
> >
> >
> > I think it is clearly stated, but if it is required to
> ask the IEEE
> > about the context of forwarding based on a destination MAC,
> I suggest
> > being more precise and specify that we are talking on
> point-to-point
> > connections.
> >
> >
> >
> > Regards,
> >
> > Nurit.
> >
> >
> >
> > -----Original Message-----
> > From: gels-bounces@rtg.ietf.org
> [mailto:gels-bounces@rtg.ietf.org] On
> > Behalf Of Don Fedyk
> > Sent: Wednesday, September 20, 2006 23:16
> > To: Adrian Farrel; ccamp@ops.ietf.org
> > Cc: gels@rtg.ietf.org; Wijnen, Bert (Bert)
> > Subject: RE: Proposed liaison response to IEEE
> >
> >
> >
> > Hi Adrian
> >
> >
> >
> > The Liaison looks good however it does not capture a comment I
> > made.
> >
> >
> >
> > My understanding of the S-VLAN translation in IEEE
> documents and the
> >
> > liaison is that the translation is valid in the context
> of forwarding
> >
> > based on a destination MAC. My interpretation is that V-LAN is
> > optional
> >
> > MAC is not. If I am correct VLAN translation as
> specified is not
> > equal
> >
> > to Label switching. The Liaison is clear about not just
> the format of
> >
> > 'packets on the wire' but the relationship to the
> entities such as
> > MAC
> >
> > relay as well.
> >
> >
> >
> > I think we should also ask in our liaison for
> clarification if the
> >
> > S-VLAN translation can be valid by itself i.e. as a
> label agnostic to
> >
> > the MAC so there is no confusion.
> >
> >
> >
> > Thanks,
> >
> > Don
> >
> >
> >
> >
> >
> > > -----Original Message-----
> >
> > > From: owner-ccamp@ops.ietf.org
> >
> > > All,
> >
> > >
> >
> > > The IETF received an unsolicited liaison on the subject of
> >
> > > 802.1 Ethernet
> >
> > >
> >
> > > You can see the liaison at
> >
> > > https://datatracker.ietf.org/documents/LIAISON/file358.pdf
> >
> > >
> >
> > > The participants on the GELS mailing list have requested that
> >
> > > we respond to this to clarify an issue.
> >
> > >
> >
> > > Here is the draft of a liaison that we proposed to send.
> >
> > > Please shout at once if you have any concerns with this
> >
> > > liaison. I intend to ask Bert to send it on Monday of
> next week.
> >
> > >
> >
> > > Thanks,
> >
> > > Adrian
> >
> > > ===
> >
> > > Subject: Response to your liaison "Use of IEEE 802.1Q
> VLAN tags"
> >
> > >
> >
> > > Date: Sep 2006
> >
> > > To: IEEE802.1
> >
> > > Tony Jeffree, IEEE 802.1 WG Chair, tony@jeffree.co.uk
> >
> > >
> >
> > > From: Bert Wijnen, IETF liaison to IETF liaison to IEEE 802.1
> >
> > > Adrian Farrel, IETF CCAMP Working Group co-chair
> >
> > > Deborah Brungard, IETF CCAMP Working Group co-chair
> >
> > >
> >
> > > Cc: Bernard Aboba bernard_aboba@hotmail.com
> >
> > > Ross Callon rcallon@juniper.net
> >
> > > Bill Fenner fenner@research.att.com
> >
> > > CCAMP mailing list ccamp@ops.ietf.org
> >
> > > GELS mailing list gels@rtg.ietf.org
> >
> > >
> >
> > > Thank you for your very informative and clarifying liaison on
> >
> > > "Use of IEEE 802.1Q VLAN tags".
> >
> > >
> >
> > > We have a question arising from this liaison and our reading
> >
> > > of the relevant standards documents as follows.
> >
> > >
> >
> > > The liaison says: "Translation of 802.1Q S-VLAN tags (with 12
> >
> > > bit VIDs) is supported at S-tagged service interfaces, as an
> >
> > > option, by the IEEE Std 802.1ad Provider Bridging amendment
> >
> > > to IEEE Std 802.1Q."
> >
> > >
> >
> > > While this text in itself is clear it leaves open the
> >
> > > question of whether "S-tagged Service interfaces" is the only
> >
> > > type of interface where Translation of 802.1Q S-VLAN tags is
> >
> > > supported.
> >
> > >
> >
> > > It is our understanding that IEEE Std 802.1ad does not
> >
> > > preclude the translation of 802.1Q S-VLAN tags at any
> >
> > > S-tagged interface. Thus, this translation would be allowed
> >
> > > at Provider Network Ports, not just at Customer Network Ports.
> >
> > >
> >
> > > Can you please confirm whether this is the correct
> >
> > > interpretation of the IEEE Std 802.1ad amendment to the IEEE
> >
> > > Std 802.1Q?
> >
> > >
> >
> > > Many thanks,
> >
> > >
> >
> > > Bert Wijnen, IETF liaison to IEEE 802.1
> >
> > > Adrian Farrel and Deborah Brungard, CCAMP Working
> Group co-chairs
> >
> > >
> >
> > >
> >
> > >
> >
> > >
> >
> >
> >
> > _______________________________________________
> >
> > gels mailing list
> >
> > gels@rtg.ietf.org
> >
> > https://rtg.ietf.org/mailman/listinfo/gels
> >
> >
> >
> >
>
>
> --
> Loa Andersson
>
> Principal Networking Architect
> Acreo AB phone: +46 8 632 77 14
> Isafjordsgatan 22 mobile: +46 739 81 21 64
> Kista, Sweden email: loa.andersson@acreo.se
> loa@pi.se
>
>