Hi Pushpasis,
On 8/26/15 16:44 , Pushpasis Sarkar wrote:
Hi Les,
On 8/26/15, 7:45 PM, "Les Ginsberg (ginsberg)" <[email protected]> wrote:
Stephane -
Implementations based on the drafts that currently exist advertise a topology
independent SRGB. A SID which is advertised in a specific MT Prefix
Reachability advertisement is interpreted as an index into the topology
independent SRGB. This is NOT compatible with an implementation which is
written assuming that a SID is an index into a topology specific SRGB. So the
introduction of topology specific SRGBs would have to be supported network-wide
before it could be deployed. Sub-TLVs cannot resolve this incompatibility.
[Pushpasis] What if we use the current SR-capability sub-TLVs only for single
topology deployments? And use a new MT-SR-Capability SubTLV for MT deployments?
Please note, I am not saying MT cannot be supported with current SR-Cap SubTLV.
It can be, but with the limitation (as I like to see this cuurently) is that we
MUST use separate SID-index for the same prefix in separate topologies. If
operator does not want to live with the limitation then all the vendor
implementations must implement the new MT-SR-Cap SubTLV and make it happen. If
the operator can live with the implementation they continue with per-topology
SID-index and single SRGB for all topology.
above would require vendors to implement both options. For operators
managing transitions between one option to the other would be difficult.
Interoperability with the implementations that only support one option
would become problematic.
Do we really want to create all this? Does the gain we would get with
per topo/algo SRGB justify all this, especially given that the gain is
not functional, but rather operational and fairly limited?
thanks,
Peter
Thanks
-Pushpasis
_______________________________________________
spring mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/spring
