Re: [Pppext] LCP echo request/reply support over multilink interface (RFC 1990)

[Y Prasad <[e-mail]>]

James,

I understand all the options we have :)
Just wanted make effort to convey the reason for this interoperability
and prevent such of these at the next draft if possible.

Regards
yp

-----Original Message-----
From: James Carlson [mailto:[e-mail]] 
Sent: Monday, March 22, 2010 12:54 AM
To: Y Prasad
Cc: Karl Fox; Vernon Schryver; [e-mail];
[e-mail]
Subject: Re: [Pppext] LCP echo request/reply support over multilink
interface (RFC 1990)

On 03/21/10 12:03, Y Prasad wrote:
> Thanks karl. Yes this would work if we are at the sending side.
> Unfortunately we are at the receiving. We have got no other option
> except implementing bundle keep alive reply though its optional.

I'm confused.

I believe the entire discussion we've had so far was with the assumption
that you wanted to send these messages, and you weren't sure whether the
peer would respond properly or how to find out whether the peer was
willing.

If you're the recipient of these LCP Echo-Request messages rather than
the sender, though, then things change quite dramatically.  You're in no
position whatsoever to make any demands on your peer; he'll do what he
wants.

I believe your choices are:

  - Be kind, and respond to his messages, regardless of what you
    or anyone else may think of them.

  - Refuse to respond, and suffer the consequences, which may
    include a lack of interoperability with that specific peer.

  - Ask the person who operates that peer system whether it can
    be configured, upgraded, or otherwise modified such that it
    doesn't rely on this useless feature.  (If not, then opt for
    one of the other two above.)

None of those choices involve protocol-level or documentation changes.

-- 
James Carlson         42.703N 71.076W         <[e-mail]>