On Fri, Dec 6, 2019 at 9:31 PM Yasuyuki Tanaka <[email protected]> wrote:
> Hi Pascal, Tengfei, > > On 12/6/2019 6:48 PM, Tengfei Chang wrote: > > Yes, MSF indeed aware of the routing information such as RPL parent, I > > consider this is like an information stored at IPv6 layer that MSF can > > read it from without touching the frame L2 payload. > > In such sense, I could consider the DSCP value can be another > > information stored at upper layer that MSF have read access to it.. > > I think these two are different things... > I agree, I am only referring Pascal's reply before. What you discussed with Malisa is another thing. > > Handling DSCP value will be a per-packet process. Can we pass DCSP value > to the TSCH layer using the interface for transmission defined by > IEEE802.15.4? I don't think so. > TC: Not sure this is a standard way to do so. For implementing, tut this value or a flag could have a default value. TC: If this value is not given, i.e. frame from IEEE802.15.4 layer, just use the default value. > > I don't see any problem in allocating additional cells for "acceptable" > amount of traffic including relayed join requests. To prevent > application packet drops, such allocations could be a good thing. > I get your point from your previous comments. I think the key point is the during the join phase, the JP can't distinguish the traffic from: -pledge or - an attacker Another way of thinking of security, if we didn't adapt the unauthorized traffic, the attacker can still send lots of "fake" join request, Instead of occupying the bandwidth, it delays or drops the join request from pledge, which is an security issue as well. I think the compromise solution is allowing MSF adapt the traffic but report if there is unusually mount of join traffic detected. > Rather than giving some L3 information to L2, the L3 may need L2 > information, like available bandwidth (cells) for a certain link. For > the MSF case, if the IPv6 layer on an intermediate node limits outgoing > traffic of relayed join requests below 20% of available bandwidth, > undesired cell allocations could be avoided, assuming > LIM_NUMCELLSUSED_HIGH is 75% and the link has almost the perfect link PDR.. > > Best, > Yatch > -- —————————————————————————————————————— Dr. Tengfei, Chang Postdoctoral Research Engineer, Inria www.tchang.org/ ——————————————————————————————————————
_______________________________________________ 6tisch mailing list [email protected] https://www.ietf.org/mailman/listinfo/6tisch
