Re: [Call-home] draft minutes

Thank you, Bert.  I'll fix the nits and use the help.  As to the comment
you made, your recollection matches mine, I think.

Wijnen, Bert (Bert) wrote:
> Eliot,
> 
> nits:
> - pls check for spelling errors.
> - You also made a few "duplicate words or sets of words"
> 
> help:
> - the gentleman from a cable company was Jean-Francois Mule
>   from cablelabs
> 
> My comment you recorded:
> 
>    Bert Wijnen:
>    SNMPv3 can send a coldstart trap, but this does not mean it addresses
>    the call home problem.
> 
> I was reacting to Keith's comment on "cold start" and the question of
> "who would initiate the session". 
> First, I think keith also brought up that an agent can send a "cold
> start" today and then the manager can start communicating with the agent.
> And with SSH such might not be possible/easy.
> 
> And so my comment was to make clear that if in the current SNMPv3 world,
> if an agent is behind a NAT, then it can send a (UDP) "cold start".
> But the Manager can still not setup a "connection" or "manage" the
> device behind the NAT, because it may still (certainly not after a
> while) be reachable from the manager. So my point is: it is not a
> new problem specific to SSH. I agree that SSH may add additional
> aspects to the problem.

Or subtract depending on how we handle session initiation.

I'll amend the minutes to include your comment, assuming no violent
objections.

Eliot

_______________________________________________
Call-home mailing list
Call-home@ietf.org
https://www1.ietf.org/mailman/listinfo/call-home