Re: [sip-clf] WGLC: SIPCLF Problem Statement (draft-gurbani-sipclf-problem-statement-01)

["Vijay K. Gurbani" <[e-mail]>]

On 03/26/2010 11:10 PM, Hadriel Kaplan wrote:
> Sorry for dredging up an old email, but I was scanning old ones and
> came upon this... [I removed lots of text from the original]
[...]
>> I don't think this information model need to be extensible. In fact
>> the one thing I am going to continue to strongly argue is that it
>> should not be extensible. Note this does not mean we can not have
>> new sip headers or new body types, we can, they just go in the list
>> of sip headers or body type.
>
> I don't understand that statement.  You don't want it to be
> extensible, but you're ok with adding more headers and bodies?

If you look at the indexed-ASCII approach, it defines a limited
set of fixed headers (Timestamp, To, From, etc.)  Then it has a
TLV block where you can add as many headers as you want.
The advantage is that this scheme is extensible -- you add
anything you want in the TLV block without perturbing the
fixed block.  You can be creative and come up with a format
string that tell a SIP entity what to put in the TLV block
(a la Apache) at run-time.

> 1) We won't get the list right.  We'll try, but odds are some admin
> will want header Foo which we thought she wouldn't want.
> 2) Some products should/need to record things others don't.  SBC's for
> example need to record some stuff that make no sense for a proxy, and
> likewise a forking proxy might need to record stuff a simple UAC
> wouldn't. We won't want to clutter the list of fields up to
> accommodate the different product types.
> 3) We haven't really talked about events or processing results,
> but I can imagine someday we might want to record more than just
> headers as part of the message record.  For example, if we ever
> make a deployable version of rfc4474, then a verifier generating
> a clf entry for a received Invite might want to add "verified"
> vs. not, or some such. ;)

I believe all of the above use cases will be supported by the
TLV block.

Thanks,

- vijay
-- 
Vijay K. Gurbani, Bell Laboratories, Alcatel-Lucent
1960 Lucent Lane, Rm. 9C-533, Naperville, Illinois 60566 (USA)
Email: vkg@{alcatel-lucent.com,bell-labs.com,acm.org}
Web:   http://ect.bell-labs.com/who/vkg/