Hi Kireeti,

I would like to challenge a bit your fundamental assumption which is to
state that while loopbacks are very important and locally significant and
warrant manual/nms provisioning SIDs are not.

Well let me clarify - SIDs are NOT MPLS labels (regardless if we think
local MPLS labels or global/domain wide MPLS labels). So I am of the
opinion that provisioning at least node SIDs should be an NMS function.

If I am building my path segment chain my abstraction to uniquely identify
nodes it traverses may just use SID abstraction without need of mapping
loopbacks to SIDs. Mapping is good, but too much mapping is not a feature.

Perhaps for other SID types some way of mapping is OK .. and such mapping
has been already implemented in SR controllers so why do we need a new
mechanism here for that?

Thx,
R.

On Wed, Jul 24, 2019 at 4:55 PM Kireeti Kompella <[email protected]>
wrote:

> Hi All,
>
> Two questions:
> a) should the SPRING WG work on ways to manage global SIDs of various
> types?
> b) if yes, what is the best-suited protocol to do this?  I suggested DHCP;
> others have suggested PCEP, SRMS, LISP mapping, etc.
>
> --
> Kireeti
> _______________________________________________
> spring mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/spring
>
_______________________________________________
spring mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/spring

Reply via email to