[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Question on LMP Adjacency



Hi,
 
I *think* what you are saying is that in the control plane the physical connectivity is
 
node#1----node#2----node#3
 
 
but in the data plane the TE connectivity is
 
node#1----node#2      node#3
  |                     |
   ---------------------
 
Note that these are TE links and there may be other underlying physical connectivity.
 
 
Note that an LMP control channel is formed simply when the two LMP speakers are in communication. Since they communicate using LMP over UDP, this is not a big problem in your figure so long as:
 
1. node#1 knows about the existence of node#3
   and can send UDP packets to it.
2. node#2 is capable of forwarding IP packets
 
Cheers,
Adrian
 
 
----- Original Message -----
From: "Harish M" <harishm@huawei.com>
To: "±e¿μE­" <yhwkim@etri.re.kr>; <ccamp@ops.ietf.org>
Sent: Friday, December 10, 2004 7:29 AM
Subject: RE: Question on LMP Adjacency

> Question is not clear.
>
>  -----Original Message-----
> From:
owner-ccamp@ops.ietf.org [mailto:owner-ccamp@ops.ietf.org]On Behalf Of
> 김영화
> Sent: Friday, December 10, 2004 12:37 PM
> To:
ccamp@ops.ietf.org
> Subject: Question on LMP Adjacency
>
>
>   Hi, ccampers
>
>   I've a question on the LMP adjacency.
>   In the LMP draft, the definition is as follows:
>
>      An "LMP adjacency" is formed between two nodes when at least one bi-
>      directional control channel is established between them. Multiple
>      control channels may be active simultaneously for each adjacency;
>      control channel parameters, however, MUST be individually negotiated
>      for each control channel.
>
>   Then, my question is this.
>
>   For an example,
>   we assume a sample network consisting of Node#1, Node#2, and Node#3.
> Node#1 is directly connected to Node#2, and Node#2 is directly connected to
> Node#3.
>   This means there is no direct connection between Node#1 and Node#3 in the
> sample network.
>   In this situation, we assume that an EoS channel(EoS#1) between Node#1 and
> Node#2 is terminated, but another EoS channel(EoS#2) between Node#1 and
> Node#3 is terminated through Node#2.
>   Here, I think that in case of the EoS#1, LMP adjacent nodes are Node#1 and
> Node#2, but in case of the EoS#2, LMP adjacent nodes are Node#1 and Node#3.
> In addition, I think that the type of data link for EoS#1 and EoS#2 is a
> component link and the link connection verification of EoS#2 between Node#1
> and Node#3 could be performed. The only condition is, of course,  is to
> observe the definition of the LMP adjacency above.
>
>      Node#1         Node#2         Node#3
>         |-----EoS#1----|              |
>         |------------EoS#2------------|
>
>   Could you help me if my understanding is short ?
>
>   Thanks,
>
>   Young.
>
>
>
>