Re: [Isis-wg] draft-amante-isis-reverse-metric-01

[Mikael Abrahamsson <[e-mail]>]

On Mon, 6 Dec 2010, Les Ginsberg (ginsberg) wrote:

> What we don't agree on is that the convenience of being able to go to 
> Node A and issue command(s) which make the necessary temporary config 
> changes (as opposed to going to Node A and B) justifies the introduction 
> of the Reverse Metric TLV and its associated complexities.

I'd imagine anyone working daily using CLI to administer their large 
network would greatly appreciate this "convenience" as you call it.

I pitched the idea to our ops guys and they were really enthusiastic about 
it, same thing with our NOC guys.

I'd say it probably is the same convenience as having an automatic gearbox 
in your car. It's perfectly fine to shift manually (I personally only buy 
manual gearbox cars), but a lot of people like the convenience of having 
an automatic.

Personally I think the work being done here the past 1-2 months when it 
comes to bringing more functionality into ISIS when it comes to more 
granularity to LAN (having bidir metric to the pseudo node) and the 
reverse metric functionality will make ISIS more usable long term, and 
it's important that we keep doing this. Just because this isn't the 
perfect solution (as "perfect" in this case would be "don't administer the 
network using CLI") I see nothing better on the horizon, so the proposed 
changes would be very welcome.

Considering a lot of other "complexities" being implemented by people 
(RSVP-TE schemes with autobw re-route, MPLS-TP, 1588v2 etc), this seems as 
complex as a walk in the park.

Put simply, the proposed changes to ISIS should have been done a long time 
ago to help the people who work daily to administrate these kinds of 
networks.

-- 
Mikael Abrahamsson    email: [e-mail]