Re: Comments on draft-manral-ospf-router-lsa-unknown-type-02

[Vishwas Manral <[e-mail]>]

Hi Alex,

> Seems it's time to put the cards on the table :) What kind of 
> information do you think we would put in link records?
I think we can have something equivalent to what ISIS has (TLV for
experimental use). 
http://www.ietf.org/internet-drafts/draft-ietf-isis-experimental-tlv-04.
txt . We could also use it for sending messages for informational
purposes.

Thanks,
Vishwas

-----Original Message-----
From: Mailing List [mailto:[e-mail]] On Behalf Of Alex
Zinin
Sent: Wednesday, March 16, 2005 12:27 PM
To: [e-mail]
Subject: Re: Comments on draft-manral-ospf-router-lsa-unknown-type-02

Vishwas,

>>> In this case too as long as we are sure no routing loops are
>>> occurring it will not cause any problems.
>> How would you propose we do that?
> The way to impose it is that we do not allow any standard link type
> extensions which can cause routing changes and hence loops.

Seems it's time to put the cards on the table :) What kind of
information
do you think we would put in link records?

We need to keep in mind that we only have one router LSA per router and
its size
needs to be well-controlled. I personally would be reluctant to put
there
anything that isn't related to the basic topology abstraction.

> I think adding the option bit as you suggest will help for sure.
However
> the point is do we want to extend OSPF that way (we already have
opaque
> LSA's). If there is demand to add information which can cause routing
> changes, we could go in for the heavier extension as you suggest.

This is what was done for stub and NSSA areas. I would use the same
method for
other extensions that affect route calculations. If they don't, then the
question is whether this info belongs in the router-LSA at all.

Alex