The MPLS WG Archive[Date Prev][Date Next][Thread Prev][Thread Next] [Date Index][Thread Index][Author Index][Subject Index] cos tspec in RSVP-TE
Markus Jork wrote: > > There was a minor change from draft-ietf-mpls-rsvp-lsp-tunnel-04 > to draft-ietf-mpls-rsvp-lsp-tunnel-05. The section about the > CoS Tspec/Flowspec maximum packet size (M) parameter now says: > "When the object is contained in Path > messages, this parameter is updated at each hop with the lesser > of the received value and the MTU of the outgoing interface." > > This is clearly the wrong thing to do and unnecessary. It conflicts > with the traditional rsvp model where the sender_tspec characterizes > the sender's traffic and the adspec is used to gather information > along the path. Now requiring that a tspec has to be updated along > the path with link specific information is not a good idea, especially > when we don't gain any advantage by doing so. The adspec already > has the path mtu information and there is no point in collecting > the same information in the tspec where it doesn't belong. > And doing this because it is more convenient than parsing > an intserv adspec is a rather weak argument, too. Personally, I agree with you here. I think the current draft breaks with the spirit of RSVP. I think it would be best to simply require the ADSPEC (as straight RSVP does for router-to-router Path messages). The egress node, seeing the CoS TSPEC will then know to ignore all parameters other than "M". By making the TSPEC behave as an ADSPEC, you effectively eliminate the TSPEC. The egress node no longer knows what MTU the ingress node is requesting. It only knows the network's capabilities - which may be greater than the amount required. I do realize that MTU isn't bandwidth here, but I think the egress node should get both the requested value as well as the ADSPEC value. On some routers, resources may be wasted if an MTU greater than required is signalled. I would also add as a side comment that there doesn't seem to be any purpose to signalling the MTU in the CoS FLOWSPEC if the sender's requirements are not propagated all the way the recipient. If you only want to signal "whatever the network's capable of", you can do that without specifying anything. -- David
|
|