The MPLS WG Archive

["Thomas D. Nadeau" <tnadeau@cisco.com>]

         Hi,

> > If all you are interested in is Juniper switches I understand your
> > viewpoint. However, the point of putting together a standard MIB
> > is to make it general enough to implement on other switches as well.
>
>That I do not argue with.  In fact, we would very much like to have this
>on all LSRs, such that we can switch out the front ends without having to
>change the backends out.  What we would like is that the kireeti MIB be
>standardized with the specical JNX bits generalized. The MIB is a good
>match to how networks are being operated today, and as such, should form a
>good basis for a standard MIB implemented across all platforms.

         So why not just take the salient features out of that MIB and
put them into the MPLS-TE-MIB? All of these features could easily be
included in the TE-MIB, as well as other features for the
others who were interested in them. If you have looked at the
MPLS-TE-MIB lately (which I unfortunately do not believe many
of the folks commenting for keeping this other MIB today have not),
it is not that much larger than Kireeti's MIB in terms of tables to
manage. The change in your operational software should not be that significant
considering the structure of things. We currently have 4 tables, one of which
is optional. Two of these tables are basically for the tunnel Hop information
which Kireeti has crammed into a single OCTET-STRING (very bad design).
If you don't care to look at LSP information, then don't follow the
XCIndex down into the MPLS-LSR-MIB and be done with it. We can easily
make conforming with the MIB easier by making some other tables
optional.

         --Tom