Hello, You cannot use an IP address based on the short address and be *assured* you have mapping between the L2 & IPv6 address.
You can only be assured of this with EUI-64 (e.g.: long address) based devices. Does anything stop a device from using an address of e.g. fe80::ff:fe00:7463 with a different short address? The long address is 'protected' against this because of the U/L bit, a node couldn't use an address that looks EUI-64 based because it can't guarantee universal uniqueness unless it is actually EUI-64 based. For this reason 6lowpan-nd-15 seems to specify address resolution is only skipped for EUI-64 based IPv6 addresses. Peter's other problem is that if you wish to use a global address, you do always need to send to a 6LR, even if the destination is actually on-link to you. I think it is just 'not possible' based solely on the current draft. Regards, -Colin -----Original Message----- From: Z Z [mailto:[email protected]] Sent: March 7, 2011 11:23 PM To: Peter van derStok Cc: Robert Assimiti; Colin O'Flynn; '6lowpan 6lowpan' Subject: Re: [6lowpan] nd-15 for isolated network Peter, I think there is some confusion. If you are sending packets from one node to another within the PAN you do not necessarily have to send the packet to the 6lbr. If the routing protocol has a mechanism to route point to point or if you are using some sort of layer 2.5 (mesh under) forwarding protocol then packets should be deliverable directly between the nodes based on the MAC address which is the IID. Only if the 6lbr is the only node with knowledge of network topology would you have to send packets to the border router. Certainly for next hop neighbors you can use the IID (which is device's short or long address mac address) as the dest address in the 15.4 header and deliver the packet directly. For nodes two hops or more away you would use the routing protocol to deliver the packet to the next IP hop toward the destination node. With mesh under you would use the mesh forwarding mechanism to send the packet to the destination via layer 2.5 forwarding (only one ip hop away). On Fri, 2011-03-04 at 10:23 +0100, Stok, Peter van der wrote: Dear Robert, Colin, > > > > Thanks for your comments. > > Let me motivate my interest in looking at the routing issues. > > > > In the applications I am involved, the bulk (say >90%) will be messages to > one hop neighbors and possibly two-hop neighbors. Actually, I expect that the > quality of the link between source and destination can be better than the > link quality between source and 6LBR. (people may argue that this is bad > network engineering, but given costs and knowledge today it looks a very > probable situation) > > What worries me is that for a communication between a source and a > destination with the same prefix (same LOWPAN) the packets first need to be > sent to a 6LBR (or 6LR) before they are routed to the destination, although > source and destination are only one hop away. > > In my opinion removing this technically superfluous overhead is important. > > > > Greetings, > > > > peter > > > > From: Robert Assimiti [mailto:[email protected]] > Sent: Thursday 3 March 2011 19:43 > To: Colin O'Flynn; Stok, Peter van der; '6lowpan 6lowpan' > Subject: RE: [6lowpan] nd-15 for isolated network > > > > > It is good to finally see efforts (following lengthy discussions) being made > for the inexorable extinction (following a long logical effort) of the > mesh-under paradigm, a vestige of proprietary routing blunders…… > > > > Robert Assimiti > > Office: [678]-202-6859 > > Mobile: [404]-578-0205 > > [email protected] > > > > > From: [email protected] [mailto:[email protected]] On Behalf Of > Colin O'Flynn > Sent: Thursday, March 03, 2011 1:35 PM > To: 'Stok, Peter van der'; '6lowpan 6lowpan' > Subject: Re: [6lowpan] nd-15 for isolated network > > > > > Hi Peter, > > > > As a disclaimer: I’m not an author of the document, they would probably be > better to give a more complete answer. > > > > I believe your conclusions are entirely correct, as I had reached the same > myself. Basically for mesh-under you are limited to using the link-local > addresses based on EUI-64, as otherwise a 6LN cannot perform address > resolution on another 6LN. > > > > Regards, > > > > -Colin O’Flynn > > > > > > From: Stok, Peter van der [mailto:[email protected]] > Sent: March 3, 2011 2:26 PM > To: Colin O'Flynn; '6lowpan 6lowpan' > Subject: RE: [6lowpan] nd-15 for isolated network > > > > > Hi Colin, > > > > Thanks for the clarifications. > > > > Sending packets to destinations using mesh-under still remains a bit vague to > me. > > Given a LOWPAN with a 6LBR connected to DHCP, DNS, etc, nodes will use the > prefix of the LOWPAN in the IP address. > > According to 5.6 a packet with a non link-local IP address is assumed to be > off-link and sent to 6LBR. > > However, when the prefix of the destination is the same as the prefix of the > source, source and destination are hosts in the same LOWPAN. In this case the > packet can be sent over the link with mesh-under routing to the destination. > > In my view sending the packet to 6LBR contradicts the mesh-under concept, > because the packet is first routed to 6LBR. > > To send the packet without passing via the 6LBR, the source needs the Link > address of the destination, but this link address is only available to the > 6LBR. > > A solution is configuring every host as a 6LR, but that defeats the purpose > of the 6LR. > > > > What have I missed? > > > > If the above reasoning is correct, a mechanism is lacking in which a host can > ask the 6LBR the link address of a given IP address with the LOWPAN prefix to > allow proper mesh-under routing. > > > > Looking forward to your reaction. > > > > Peter > > > > > > From: Colin O'Flynn [mailto:[email protected]] > Sent: Tuesday 1 March 2011 14:03 > To: Stok, Peter van der; '6lowpan 6lowpan' > Subject: RE: [6lowpan] nd-15 for isolated network > > > > > Hi Peter, > > > > I think you are basically correct, with some additional constraints or > clarifications: > > > > The 6LN NC will have entries for routers it has registered with, so it’s not > always empty. > > > > Section 5.6 allows you to use the MAC extraction only if the link-local > address is EUI-64 based. This basically means if the U/L bit is set, > indicating the address in question was generated from a known-unique MAC > address. 802.15.4 for example has 16-bit addresses you could be using instead. > > > > I think most networks would have the 6LBR, although obviously if you have a > very specific situation as you outlined you could skip it. The 6LBR at > minimum would manage the compression context (if required) and serve as an > alternative way to reach a node by a default route. From a > maintenance/deployment/management perspective the 6LBR is an easy way to see > what nodes are alive on your network too by checking its tables. > > > > Regards, > > > > -Colin > > > > From: [email protected] [mailto:[email protected]] On Behalf Of > Stok, Peter van der > Sent: March 1, 2011 1:21 PM > To: 6lowpan 6lowpan > Subject: [6lowpan] nd-15 for isolated network > > > > > Dear authors, > > > > The document looks rather complete and comprehensive. > > > > There are a few questions: > > Do I understand correctly that contrary to RFC 4861, the neighbor cache is > always empty in 6LN. If true, this remark may be added to Registration term > of section 2. > > From that do I deduce correctly that access to the link for link-local > addresses (LLA) involves extracting the MAC address from the LLA. > > > > MUST a 6LBR be present in an isolated LOWPAN? (6LBR text in section 2 seems > to imply this) > > Assuming an isolated LOWPAN without 6LR or 6LBR, then there will be no answer > to the RS message, but the node can continue sending messages to LLA, where > the MAC address is again extracted from the LLA. Is that correct? > > > > Peter > > > > Peter van der Stok > > Philips Research Laboratories Eindhoven > > High Tech Campus HTC > 34 (WB) 1-067 > > 5656 AA Eindhoven The > Netherlands > > phone +31 40 2749657 Fax: + 31 > 40 2746321 > > mailto: [email protected] > > > > > > > The information contained in this message may be confidential and legally > protected under applicable law. The message is intended solely for the > addressee(s). If you are not the intended recipient, you are hereby notified > that any use, forwarding, dissemination, or reproduction of this message is > strictly prohibited and may be unlawful. If you are not the intended > recipient, please contact the sender by return e-mail and destroy all copies > of the original message. > > > > > This e-mail (including any attachments to it) is confidential, proprietary, > legally privileged, subject to copyright and is sent for the personal > attention of the intended recipient only. If you have received this e-mail in > error, please reply to advise us immediately, delete it and destroy any > printed copies of it. You are notified that reading, disclosing, copying, > distributing or taking any action in reliance on the contents of this > information is strictly prohibited. No employee is authorized to conclude any > binding agreement on behalf of NIVIS LLC with another party by e-mail without > express written confirmation by an officer of the company. Although we have > taken reasonable precautions to ensure no viruses are present in this e-mail, > we cannot accept responsibility for any loss or damage arising from the > viruses in this e-mail or attachments. > > > _______________________________________________ > 6lowpan mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/6lowpan > _______________________________________________ 6lowpan mailing list [email protected] https://www.ietf.org/mailman/listinfo/6lowpan
