> -----邮件原件----- > 发件人: nvo3 [mailto:[email protected]] 代表 Thomas Narten > 发送时间: 2013年11月27日 0:58 > 收件人: [email protected] > 主题: [nvo3] TTL handling in an L3 service > > Hi. > > In precisely defining L3 service, one question that comes up is how is the TTL > treated. That is, say NVO3 provides L3 VN service to a TS. When TSes on the VN > communicate with each other, they are always using IP. What happens to the > TTL in such packets? > > I see several choices: > > a) do not decrement the TTL at all. Treat the TSes as if they were directly > attached to each other (i.e., on the same link) > > b) Decrement by 1... > > c) Decrement by some random amount.. :-) > > Some protocols may care about TTL handling. IPv6 Neighbor Discovery, for > example, requires that ND packets be dropped if they are received with a TTL > other than 255 (i.e., they'd require choice a). I think some other routing > protocols do the same (as a way to ignore packets from offlink "attackers"). > > What do tenants of an L3 service expect? Do they care (other than in cases > like > ND)? > > Can we just define L3 service as saying the TTL of tenant packets are not > modified by NVO3? > > Any advice from L3 service providers that already provide such services today?
Hi Thomas, By the way, you may also ask feedbacks from some vendors who have implemented the feature of forwarding intra-subnet traffic at L3 in their released data center network products. From the following public information which can be googled, it seems at least the following three vendors have implemented it: www.juniper.net/us/en/local/pdf/whitepapers/2000535-en.pdf (quoted text: Fallback Switching Contrail supports a hybrid mode where a virtual network is both a L2 and a L3 overlay simultaneously. In this case the routing instances on the vRouters have both an IP FIB and a MAC FIB. For every packet, the vRouter first does a lookup in the IP FIB. If the IP FIB contains a matching route, it is used for forwarding the packet. If the IP FIB does not contain a matching route, the vRouter does a lookup in the MAC FIB―hence the name fallback switching. Note that the “route first and then bridge” behavior of fallback switching is the opposite of the “bridge first and then route” behavior of integrated routing and bridging (IRB).) http://blogs.enterasys.com/dci-made-simple-with-onefabric/ (quoted text: Fabric Routing with IP mobility uses host routing techniques to dynamically distribute and inject host routes from the data center switch (that has fabric routing enabled) that a VM is closest connected to �C and remove them from the previous closest fabric routing switch) http://www.cisco.com/en/US/solutions/collateral/ns224/ns945/white_paper_c11-728337.pdf (quoted text: Cisco DFA advancements include enhanced forwarding, in which IP addresses are used regardless of whether the communication is within or between traditional Layer 2 subnets. This feature introduces several optimizations and simplifications, including the elimination of a first-hop redundancy protocol, the use of small MAC address tables, and optimal forwarding for all unicast frames.) http://www.valleytalk.org/wp-content/uploads/2013/08/ciscoDFA.pdf Best regards, Xiaohu > Thomas > > _______________________________________________ > nvo3 mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/nvo3
_______________________________________________ nvo3 mailing list [email protected] https://www.ietf.org/mailman/listinfo/nvo3
