The MPLS WG Archive

[Ravi Shekhar <ravi_shekhar007@yahoo.com>]

Ping,
> 
> In RSVP-TE, this is the RRO object. For RSVP, you can even skip this two
> steps.

  Thats true, but I am not assuming that an LSP is signalled just using RSVP.
  It will work for CR-LDP, LDP which might distribute same label to multiple
  upstreams or any LSP setup using mix of these protocols. (I think one of 
  your mail asked to suggest something which will work for LDP). Infact it 
  will work for multicast(point-to-multipoint) LSPs too.

> 
> Yep! This is all nice, but it is *not* backward compatible. In a network
> that has many routers, if one of them does not support this, the test
> will fail.
> 
> Also this is a self supply-demand condition: with one implementation bug
> on one of the LSR's, a whole bunch of LSP's may get torn down. So your
> approach is best to debug own bugs. :-)

  Isnt that goodness..;-).. LSP and LSP ping assist each other in 
  finding bugs. 
  Anyhow, you could say the same thing for traceroute. If ICMP is
  not implemented correctly on one of the intermediate hops, it wont work.
  Infact, this argument will completely shoot down multicast traceroute
  where state is appended on each intermedidate router. And I bet 
  implementations of LSP ping as you suggest wont be bug immune either..;)..

> 
> See, the LSP-ping proposal works, because it is based on the assumption:
> the only working path from egress to ingress is the control path itself.
> It is simple (no extra signaling other than adding one object) and
> backward compatible (no change to the network LSR's).

  If by backward compatible you mean that intermediate LSRs do not have 
  to implement LSP-ping, then I agree. 
  But is that really such a big win in the long run? If we are tracing LSP to an
  LSR(which could be one of these inetwork LSRs) and if we dont recieve a
  response from it, it could either be because LSP is down or the LSR does not
  implement LSP-ping. How do you know to distinguish between these two?
  And if you cant, then indirectly it implies that LSP-ping be implemented
  on all such routers and I think practically thats what will end up happening.
  Whichever ping mechanism is used, we will end up having to support it on all 
  the routers.  

  - Ravi Shekhar.

__________________________________________________
Do You Yahoo!?
Get personalized email addresses from Yahoo! Mail
http://personal.mail.yahoo.com/