The MPLS-OPS Archive

["Mark Lewis" <markl11@hotmail.com>]

Hi,

The penultimate P router does only swap the outermost label, but with PHP of 
course the outermost label is popped.

As the RFC states, this (PHP) saves one of two lookups that would otherwise 
take place. If the packet were to arrive at the egress PE router with an 
intact label stack (ie no PHP), then the PE router would do a label lookup 
(in the LFIB/TFIB) based on the outermost label (1st lookup). This lookup 
would result only in the router 'discovering' that there was no 
corresponding outbound label, and therefore that a layer 3 lookup was 
required. The PE router would then do a lookup in the routing table in order 
to forward the packet (2nd lookup).
However, if PHP is performed, the outermost label is popped at the 
penultimate router, and the 1st lookup (as mentioned above) is avoided, and 
only the layer 3 (routing table) lookup is performed - hay presto, a saving 
of one lookup.

Hope that helps,

Mark Lewis
Frontrunner Solutions


>From: "Alexander Marhold" <alexander@marhold.at>
>To: "Bhupesh Bajaj" <mpls@indiatimes.com>, <prabhu_b@in.huawei.com>,   
><asheeshj@in.huawei.com>, <srinivasch@in.huawei.com>
>CC: <mpls-ops@mplsrc.com>
>Subject: RE: doubt in RFC 3031
>Date: Mon, 30 Apr 2001 12:49:05 +0200
>
>IMHO you are right with your statement, that normally a P-router (the
>penultimate hop) does only label swapping based on the outermost label and
>this a single lookup.
>
>the sentence maybe could be better phrased in that way:
>" This technique allows the egress to do a single lookup, without requiring
>any additional lookup by the penultimate node "
>
>But I think, that by reading the whole chapter the meaning of PHP is clear
>and that this single sentence in his original wording put into the context
>does not do any harm.
>
>regards
>
>Alexander Marhold
>Senior Consultant
>PRO IN
>
>-----Original Message-----
>From: owner-mpls@UU.NET [mailto:owner-mpls@UU.NET]On Behalf Of Bhupesh
>Bajaj
>Sent: Monday, April 30, 2001 12:12 PM
>To: prabhu_b@in.huawei.com; asheeshj@in.huawei.com;
>srinivasch@in.huawei.com
>Cc: mpls-ops@mplsrc.com
>Subject: doubt in RFC 3031
>
>
>hi all.
>while going thru RFC 3031 , in Page 18 , second paragraph says...
>
>" This technique allows the egress to do a single lookup, and also requires
>only a single lookup by the penultimate node "
>
>
>isnt the penultimate node going to do a single lookup anyway...? or i have
>got it all wrong..?
>
>according to what i perceived...the penultimate node is going to look into
>the label which it received and then sends the packet to the next hop
>according to the information in the label it received with packet...which
>means ...single lookup....
>
>when would it do a second lookup..?
>
>is it referring to the second lookup that the penultimate node will have to
>do when it wouldnt find the label in the stack which it is expecting ...and
>it will go for network layer information...?
>
>kindly let me know.
>thanks and regards,
>
>bhupesh.
>Get Your Private, Free E-mail from Indiatimes at 
>http://email.indiatimes.com
>
>-------
>The MPLS-OPS Mailing List
>Subscribe/Unsubscribe:  http://www.mplsrc.com/mplsops.shtml
>Archive: http://www.mplsrc.com/mpls-ops_archive.shtml

_________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.

-------
The MPLS-OPS Mailing List
Subscribe/Unsubscribe:  http://www.mplsrc.com/mplsops.shtml
Archive: http://www.mplsrc.com/mpls-ops_archive.shtml