The MPLS WG Archive[Date Prev][Date Next][Thread Prev][Thread Next] [Date Index][Thread Index][Author Index][Subject Index] [Fwd: I-D ACTION:draft-pan-rsvp-te-restart-01.txt]
Ping Pan replied 15 August 2001 14:34 > > Can you please explain (i) what causes such RSVP nodal > re-starts in the 1st > > place > The same cause why routing engine may fail in the network, while data > forwarding is fine. The graceful restart technique is not > new, as being > define for IGP and BGP already. > > > and (ii) how frequently you see them per reason (I assume > they must be > > of some concern to have written this ID)? > > > > If you believe the data collected at > http://www.merit.edu/ipma/instability/, network does run into instable > conditions. Our draft is to localize the problem when that happens. > Thanks Ping.....but when I looked at the site the topic is route-flaps. Which I can understand for LDP stability, since this is dependent on IGP/BGP stability. But I failed to understand why you consider this the cause of RSVP-TE restarts. I would have thought other mechanisms would have been the key cause, which are (RSVP) protocol specific....maybe error events taking out 3 hello messages perhaps? Can you clarify what I am missing here? regards, Neil
|
|