Re: NHS Record extensions seem to be unnecessary

[Andrew Smith <[e-mail]>]

> From [e-mail] Wed Aug 23 15:07:34 1995
> To: "Andrew G. Malis" <[e-mail]>
> Cc: [e-mail]
> Subject: Re: NHS Record extensions seem to be unnecessary  
> Date: Wed, 23 Aug 95 14:49:44 PDT
> From: Yakov Rekhter <[e-mail]>
> 
> > >   - NHRP Forward NHS Record Extension
> > >   - NHRP Reverse NHS Record Extension
> > > 
> > > do not serve any useful purposes in NHRP. Therefore I would suggest
> > > to delete them.
> > 
> > I strongly disagree.  I expect these to be invaluable for debugging an
> > implementation, and when tracking down the inevitable operational
> > problems once NHRP has been deployed.
> 
> Fine. Just say that these two options are for debugging.

Isn't this another case of complicating the client-server protocol in 
order to support the server-server interactions? If the NHS is giving
a client the wrong answer then it is that NHS that needs the debugging
help, not the poor client. I can't imagine anyone writing a NHRP-debugging 
application to be run by clients which would include these options.

I suggest mandating that these options not be used by clients but may
be used by NHSs.

I also disagree with including protocol elements to aid debugging an
implementation. Add stuff for diagnosing operational problems, yes but not
to detect software bugs, please.

> Yakov.
> 

Andrew 


********************************************************************************
Andrew Smith					TEL:	[phone]
Technology Synergy Unit				FAX:	[phone]
Bay Networks, Inc.				E-m:	[e-mail]
Santa Clara, CA
********************************************************************************