Hi Dino,
> On 26 Jan 2018, at 19:31, Dino Farinacci <[email protected]> wrote: >> >> The fact that (P)ITRs use this mechanism does make it a data-plane >> mechanism. is the LISP control plane running on (P)ITRs that does it, using >> control plane messages. >>> RLOC-Probing is a method that an ITR or PITR can use to determine the >>> reachability status of one or more Locators that it has cached in a >>> Map-Cache entry. The probe-bit of the Map-Request and Map-Reply >>> messages is used for RLOC-Probing. > > No one said PITRs are special or different than ITRs in this regard. For > both, they use the control plane messages to determine reachabiltity for > RLOCs that THE DATA-PLANE encapsulates to. May be I did not express clearly. I did not suggest that PITR and ITR are different in RLOC probing. > The data-plane operation decides based on packet counter, RTT, TTL and other > mechanisms if the RLOC is of quality for use. > > Any other entity that doesn’t do packet forwarding that does mapping system > lookups doesn’t consider these things. Hence, why this issue of the > control-plane needs to be described in the document related to packet > forwarding. > Two quick comments: 1. The first sentence of the second paragraph of the RLOC Probing section clearly states that RLOC probing is a control plane mechanism based on a time. 2. RLOC Probing can be done by anybody, not only ITRs and PITRs. Any other entity can check RLOCs to make sure they are alive. As such better put the text in 6833bis. Ciao L. _______________________________________________ lisp mailing list [email protected] https://www.ietf.org/mailman/listinfo/lisp
