Hi Michael,
when you say minimal, do you refer to the minimal configuration or the minimal security? I understand it in the following way. 1) a mesh network is formed and all nodes that already joined the network run minimal + 6P + some SF (e.g SF0). This guarantees that any node already joined can relay traffic. Nodes react allocating more cells if they detect and increment of cell usage. 2) a joining node learns the minimal schedule from the EB and then starts the joining process through that minimal slot. Until it has not been allocated the security keys it cannot start 6P to allocate more cells. 3) Your point is that a malicious node can start to do join requests and cause 6P to allocate more cells in the schedule in the intermediate nodes in the network, introducing churn. Is this your concern? Could't we protect this from the JP? X 2017-02-03 17:52 GMT+01:00 Michael Richardson <[email protected]>: > > Malisa and I, when editing the documents this week came across a concern > that we had. It involves the traffic between the Join Proxy (JP), and the > Join Registrar/Coordinator (JRC, formerly known as JCE). > > This is regular traffic, and is would flow through the mesh as would any > other traffic. Only the JP and JRC would know it was join traffic, and > if the JRC is not-within the LLN (beyond the DODAG root), then the DODAG > root wouldn't be able to identify the join traffic. In a > smarter-than-minimal network a specific track could be allocated for this > traffic. > > The specific concern I have with minimal is that there could be surges in > join traffic (from newly activated pledges, some of which could be > malicious). That minimal might react by allocating new cells for the extra > traffic, and this might be a bad thing. > > I can't say that I know what the right behaviour is. Clearly the network > should not be allocating cells that it needs for more important things, > and I think that 6p can do deny things appropriately. But the intermediate > nodes might not recognize the traffic as join traffic, and might allocate > anyway. > Is there a way we can mark join traffic so that this doesn't happen? > Either explicitely (DSCP perhaps?), or implicitely (this cell is join > traffic). > > -- > Michael Richardson <[email protected]>, Sandelman Software Works > -= IPv6 IoT consulting =- > > > > > _______________________________________________ > 6tisch mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/6tisch > > -- Dr. Xavier Vilajosana Wireless Networks Lab *Internet Interdisciplinary Institute (IN3)Professor* (+34) 646 633 681 [email protected] <[email protected]> http://xvilajosana.org http://wine.rdi.uoc.edu Parc Mediterrani de la Tecnologia Av Carl Friedrich Gauss 5, B3 Building 08860 Castelldefels (Barcelona). Catalonia. Spain [image: Universitat Oberta de Catalunya]
_______________________________________________ 6tisch mailing list [email protected] https://www.ietf.org/mailman/listinfo/6tisch
