The MPLS WG Archive

[Curtis Villamizar <curtis@workhorse.fictitious.org>]

In message <4B6D09F3B826D411A67300D0B706EFDE84A481@nt-exch-yow.pmc-sierra.bc.ca
>, Shahram Davari writes:
> Curtis,
>  
> > ICMP is sent to detect failure.  UDP based LSP Ping is sent after an
> > failure is detected using ICMP.  Why we need the latter is somewhat of
> > a mystery to me, apparently to either distinguish a data plane from
> > control plane failure or to to circumvent anti-ICMP religion.
> > 
> 
> May be that is why you advocate that LSP-Ping could be easily extended to oth
> er types
> of MPLS control-planes. 
> 
> The reason that you need LSP-ping after the ICMP-ping failure is that, the 
> ICMP failure could have been due to reverse IP (hop-by-hop) path failure. So
> to determine whether the LSP is down or the reverse IP path, you need a disjo
> int
> reverse path => RSVP reverse path :)

Both reverse path for ICMP and LSP are IP paths.  If they went down
control traffic wouldn't flow so they can be assumed to be up.

> -Shahram

Nice try.  

Curtis