Re: [mpls] Concerns about ISD

[Robert Raszuk <[e-mail]>]

> If those 9 are not MNA capable, then they would see a label (bSPL) that
> they don’t recognize and should drop the packet.
>

The whole idea is to avoid having them to even look beyond their own
transport label.

No solutions have proposed using non-standard top labels and the MNA
> special label is not supposed to make it to the top of stack.
>

Neither do I. Top label is a standard label. The lookup of a standard
label may yield three switching vectors:

* swap with another label and fwd

* forward (no swap in case of domain wide label)

* swap or fwd but look higher as label stack contains something more you
may recognize or not. Here your list of options comes in.

Thx,
R.






>
>
> Yet for some packets we may want to have such actions executed and those
> will be using different top label.
>
> This does double the amount of labels (if one is to use
> uncoordinated allocations) but the benefit seems very sweet. And it works
> well with domain wide and local labels.
>
> Some may just perhaps use a special label range (ex: set top bit) to
> indicate special action trigger.
>
> Thoughts ?
>
>
>
> I don’t understand what you’re proposing.  Maybe if you wrote it down...
>
> Tony
>
>
>
>