The MPLS WG Archive

Cell Relay Retreat>MPLS WG Archive>month:2002-Dec> msg00352



[Date Prev][Date Next][Thread Prev][Thread Next]  
  [Date Index][Thread Index][Author Index][Subject Index]

Last call on LSP Ping

  • From: Curtis Villamizar <curtis@fictitious.org>
  • Date: Wed, 18 Dec 2002 17:30:15 -0500
  • cc: "'curtis@fictitious.org'" <curtis@fictitious.org>, Kireeti Kompella <kireeti@juniper.net>, "Gray, Eric" <egray@celoxnetworks.com>, mpls@UU.NET


In message <4B6D09F3B826D411A67300D0B706EFDEB03BBA@nt-exch-yow.pmc-sierra.bc.ca
>, Shahram Davari writes:
> > > 1, 2, 4, 5 haven't been answered.
> > 
> > 4 and 5 were deferred.
> 
> Thanks got the answers.
> 
> > 
> > > > > 1) In L2VPN or L3VPN do we need to push both inner and 
> > outer label?
> > > > > if not then do people agree to add the inner label to 
> > the FEC TLV?
> > > > 
> > > > Depends on which LSP you are testing.  Is the inner label 
> > semirandom
> > > > payload to exercise load split or is it the LSP actually 
> > under test.
> > > 
> > > I want to test the inner LSP. Do I need to push both 
> > labels? or only the
> > > transport label? In case I push both labels there is no way 
> > to detect
> > > the ping packet at the egress. So my suggestion was to only push
> > > the top label and add the inner label to the echo-request.
> > > (This point was also raised by Eric Rosen)
> > 
> > My interpretation is push both labels.  Both start at TTL=1.
> > Increment top, then next label.  If they both terminate at the same
> > egress LSR (normal for BGP VPN) then bottom TTL never gets past 1.
> 
> What do you do for pinging the inner LSP?

For ping mode, set all the labels that you want tested to TTL=255 and
send.  If there are any hops after the top, the packet gets delivered
to the egress of the label you want tested.

Did you mean traceroute?

> > > > > 2) How do we return the label mapping TLV in case 
> > RSVP-TE is used
> > > > > for the return path? (no such object is defined for RSVP-TE)
> > > > 
> > > > According to the instructions in section 5.  Use the 
> > RESV.  Add the
> > > > object specified in the doc.  What's not clear?
> > > 
> > > Section 3.2 describes something called Downstream Mapping TLV. This
> > > is defined for IP packets only. When the return path is the RSVP-TE,
> > > how does the downstream mapping gets encoded in RESV 
> > message? May be the
> > > answer is trace-route has no value when the return path is 
> > RSVP-TE ! or
> > > may be a Downstream mapping Object needs to be defined.
> > 
> > Downstream is in the forward direction so this question doesn't make
> > sense.
> > 
> > The Downstream Mapping TLV never refers to the return path.
> 
> Check the last paragraph of section 4.3:
> 
> 
>    If the echo request contains a Downstream Mapping TLV, the replier
>    SHOULD compute its downstream routers and corresponding labels for
>    the incoming label, and add Downstream Mapping TLVs for each one to
>    the echo reply it sends back.
> 
> -Shahram


At TTL=4, the packet goes 3 hops, then at the fourth has label=X and
get a TTL Expire.  The incoming label is X and the outgoing mapping is
the mapping in the forward direction, ie: the direction of the LSR
which is going to get the next packet when TTL=5.  The router at 4
should confirm that the packet arrived with label X and that the last
Downstream Mapping TLV in the request (that was originally copied from
the LSR at TTL=3) is X.  If the LSR at TTL=4 swaps label X for Y it
puts Y in the Downstream Mapping TLV in the echo reply.  The ingress
copies this into the next echo request.  The LSR at TTL=5 should see
label Y, etc.

That is the forward direction.

Curtis