Re: OSPF TLVs

[Acee Lindem <acee at REDBACK.COM>]

Hi Kireeti,

The original question is draft-ietf-ccamp-ospf-gmpls-extensions-12.txt.
It was whether or not the fields labeled "Padding" should be included
in the sub-TLV.  The answer is that they are part of the sub-TLVs
and should be included in the length (from Adrian). I think it would
be clearer if you replaced "Padding" with "Reserved" since the padding will
be there whether it is implicit or explicit.

Thanks,
Acee
----- Original Message -----
From: "Kireeti Kompella" <kireeti at JUNIPER.NET>
To: <OSPF at PEACH.EASE.LSOFT.COM>
Sent: Wednesday, May 05, 2004 12:23 PM
Subject: Re: OSPF TLVs


> On Mon, 26 Apr 2004, Adrian Farrel wrote:
>
> > I don't see why there is any change required.
>
> Agree with Adrian.
>
> > You appear to be precluding TLVs from having any length that is not a
> > multiple of four bytes. This seems unnecessary.
>
> (I don't think Venkata says that.  But then again, he can chime in.)
>
> > What we currently have is a rule that tells you how to encode a TLV into an
> > OSPF LSA if it is not a multiple of four bytes.
> > You have a separate rule that says that a specific TLV has some reserved
> > bytes.
>
> If a TLV has reserved bits/bytes that it explicitly defines, that's
> just part of the Value field as far as the TLV is concerned.
>
> > Why do we need to make any changes?
>
> Why indeed?  :-)
>
> > The only change that might be of use would be to change the name of any
> > field that is included in a TLV from "padding" to "reserved" to avoid
> > confusion.
>
> Good idea.
>
> > Perhaps the draft's authors would like to contribute to this thread.
>
> Don't know which draft you're referring to, unless it's draft 3630 :-)
>
> Kireeti.
> -------