> > > There is one problem with allowing ping in access phase only. > > > > > > Suppose both a PaC and a PAA are in access phase, and the PaC sends a > > > ping request while the PAA sends a PAR to start re-authentication at > > > the same time. Then PAA will not accept the ping request because it > > > has entered re-authentication phase. If ping is allowed in any > > > state, then PAA can answer to the ping request. > > > > I think the issue is about having two outstanding requests (one PANA- > Ping > > and one other) at the same time. The PAA (or PaC) having to respond to a > > PANA-Ping-Request while it has an outstanding request shall not have a > > problem. What do you think? > > I think having two outstanding requests sent in different directions > is acceptable. > > > > > > > > BTW, I've found another issue here. In the above example, The PaC > > > will accept the PAR sent from the PAA and can return a PAN, but it > > > cannot send a new PAR (this can happen if the PAN does not carry EAP), > > > until the ping request is answered. This is actually breaking > > > lock-step behavior of PANA (queuing requests should not be required > > > for a lock-step protocol). > > > > What would be the issue of having request queuing? > > Queueing itself may not be an issue. But the actual issues is the > sender would need to distinguish two cases (i.e., if there is an > outstanding ping request or not) for each state. > > For example, supposed that the PaC is in a state waiting for a ping > answer. If another event that needs to send a PAR is generated, then > the PaC processes the event, generates and queues a PAR and transits > to a state that is waiting for a PAN, and it is still waiting for a
It should not transit to such state until the PAR is actually transmitted. I'd say PAR (or the event that leads to PAR generation) shall be queued until the PANA-Ping is completed (ping answer is received). > ping answer. This would require N*2 states (N is number of states > except for ping, multiplied by 2 to consider whether there is a ping > request or not. > > > > > > > I'd like to see if we can avoid creating special case for PANA-Ping > messages > > (e.g., putting them under a separate treatment). > > I think the only way to avoid it is: > > - Allow sending ping request in access phase only. > Makes sense. > - Allow receiving ping request in any phase. Makes sense. > > - When there is an outstanding ping request, discard all incoming > messages except for ping answer or error request sent in resopnse to > the ping request. If the PaC receives a PAR while waiting for a PANA-Ping-Answer, why should it discard the PAR? Alper > > This may be an acceptable operation. > > Yoshihiro Ohba > > > > > > Alper > > > > > > _______________________________________________ Pana mailing list [email protected] https://www1.ietf.org/mailman/listinfo/pana
