> -----Original Message----- > From: Templin, Fred L [mailto:fred.l.temp...@boeing.com] > Sent: Thursday, June 02, 2016 6:51 AM > To: Xuxiaohu; Joe Touch; Brian E Carpenter > Cc: int-area@ietf.org > Subject: RE: [Int-area] Call for adoption of draft-xu-intarea-ip-in-udp-03 > > Hi, > > > -----Original Message----- > > From: Xuxiaohu [mailto:xuxia...@huawei.com] > > Sent: Tuesday, May 31, 2016 7:32 PM > > To: Templin, Fred L <fred.l.temp...@boeing.com>; Joe Touch > > <to...@isi.edu>; Brian E Carpenter <brian.e.carpen...@gmail.com> > > Cc: int-area@ietf.org > > Subject: RE: [Int-area] Call for adoption of > > draft-xu-intarea-ip-in-udp-03 > > > > > > > > > -----Original Message----- > > > From: Int-area [mailto:int-area-boun...@ietf.org] On Behalf Of > > > Templin, Fred L > > > Sent: Wednesday, June 01, 2016 12:30 AM > > > To: Joe Touch; Brian E Carpenter > > > Cc: int-area@ietf.org > > > Subject: Re: [Int-area] Call for adoption of > > > draft-xu-intarea-ip-in-udp-03 > > > > > > Brian and Joe are correct - tunnels within tunnels means that > > > fragmentation and reassembly are inevitable. Wish it weren't so, but that > > > is > the reality. > > > > If so, could you give a concrete example in real network environment > > where outer fragmentation is widely enabled and works very well? It would be > better if the reassembly buffer size needed in the tunnel egress could be > given as > well. > > Any tunnel that traverses a 1280 link has to fragment, but instead of outer > fragmentation it should use tunnel fragmentation which is something I have > been advocating for a very long time. The tunnel egress should configure at > least > a 2KB reassembly buffer size.
Hi Fred, The major concern about tunnel fragmentation and reassembly is whether reassembly needs to be performed by tunnel egress routers (Note that routers here mean hardware routers rather than software routers). When performing reassembly on tunnel egress routers, the throughput would be significantly decreased (e.g., by 50% to 90%). If SPs are not willing to afford that cost, the only feasible way for them to deal with this MTU issue is to configure the MTU of the core large enough so as to accommodate the added encapsulation headers. With the 2K reassembly buffer size, what's the maximum throughput that tunnel egress routers are assumed to support? especially in the case where the traffic over the tunnel is IPv6 traffic and the tunnel needs to traverse a 1280 link (in other words, all traffic would have to be fragmented after encapsulation). BTW, did you know any router vendors who had implemented the tunnel fragmentation approach that you had been advocating for a very long time? Best regards, Xiaohu > Thanks - Fred > fred.l.temp...@boeing.com > > > Best regards, > > Xiaohu > > > > > Thanks - Fred > > > > > > > -----Original Message----- > > > > From: Int-area [mailto:int-area-boun...@ietf.org] On Behalf Of Joe > > > > Touch > > > > Sent: Tuesday, May 31, 2016 8:34 AM > > > > To: Brian E Carpenter <brian.e.carpen...@gmail.com> > > > > Cc: int-area@ietf.org > > > > Subject: Re: [Int-area] Call for adoption of > > > > draft-xu-intarea-ip-in-udp-03 > > > > > > > > > > > > > > > > On 5/30/2016 1:45 PM, Brian E Carpenter wrote: > > > > > If you intend to support recursive datagram tunneling and > > > > > believe that any path has a minimum MTU, then you have to accept > reassembly. > > > > > > > > Agreed- at least one of the layers between the message tunnel and > > > > when it recurses must support fragmentation and reassembly. > > > > > > > > Those who deploy or sell systems otherwise are in denial, not a > > > > counter-proof. Again, this is where compliance validation would be > > > > useful. > > > > > > > > Joe > > > > > > > > _______________________________________________ > > > > Int-area mailing list > > > > Int-area@ietf.org > > > > https://www.ietf.org/mailman/listinfo/int-area > > > > > > > > > _______________________________________________ > > > Int-area mailing list > > > Int-area@ietf.org > > > https://www.ietf.org/mailman/listinfo/int-area > _______________________________________________ Int-area mailing list Int-area@ietf.org https://www.ietf.org/mailman/listinfo/int-area
