Re: [netmod] comments on draft-ietf-netmod-snmp-cfg--05

Juergen Schoenwaelder <j.schoenwaelder@jacobs-university.de> Fri, 13 June 2014 21:03 UTC

Date: Fri, 13 Jun 2014 23:03:47 +0200
From: Juergen Schoenwaelder <j.schoenwaelder@jacobs-university.de>
To: Andy Bierman <andy@yumaworks.com>
Message-ID: <20140613210347.GA38910@elstar.local>
Mail-Followup-To: Andy Bierman <andy@yumaworks.com>, "netmod@ietf.org" <netmod@ietf.org>
References: <CABCOCHQnzdvOGGLDSOrNsRrW19aDvRNs+o9D-sqgqjtx21WN+g@mail.gmail.com> <20140613062605.GA37482@elstar.local> <CABCOCHS2FL2_8pjTWttQ2BTLD6DPFrKgv1dg0-DB8W=s3FyWeA@mail.gmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Disposition: inline
In-Reply-To: <CABCOCHS2FL2_8pjTWttQ2BTLD6DPFrKgv1dg0-DB8W=s3FyWeA@mail.gmail.com>
User-Agent: Mutt/1.4.2.3i
Archived-At: http://mailarchive.ietf.org/arch/msg/netmod/5Z6BuHn1K8aKPf2MACRytZTz8LE
Cc: "netmod@ietf.org" <netmod@ietf.org>
Subject: Re: [netmod] comments on draft-ietf-netmod-snmp-cfg--05
Precedence: list
Reply-To: Juergen Schoenwaelder <j.schoenwaelder@jacobs-university.de>

On Fri, Jun 13, 2014 at 12:01:37PM -0700, Andy Bierman wrote:
> On Thu, Jun 12, 2014 at 11:26 PM, Juergen Schoenwaelder <
> j.schoenwaelder@jacobs-university.de> wrote:
> 
> > On Thu, Jun 12, 2014 at 02:49:27PM -0700, Andy Bierman wrote:
> > > Hi,
> > >
> > > It's probably past WG Last Call, but I still do not understand
> > > the value of splitting the ietf-snmp configuration module into
> > > 12 files (1 main + 11 sub-modules) instead of just 1 file.
> > >
> > > IMO it makes it harder to read, not easier, and wrt/ YANG library
> > > management,
> > > there are 12 files to update and keep in sync, instead of 1.
> > >
> > > Is there some expectation that individual sub-modules for SNMP
> > > configuration will be updated regularly in new RFCs?
> >
> > The organization essentially reflects the modular structure of the
> > SNMPv3 MIB modules. The SNMPv3 MIB modules are roughly the same
> > number.
> >
> > If updates will ever will be necessary, we might benefit from the
> > modular structure. If updates will never be needed, there is not much
> > work with synchronizing the sub-modules either. Whether modularity is
> > a good thing of not can likely be discussed endlessly. In this case,
> > we followed the modular structure of the SNMPv3 MIB modules.
> >
> >
> 
> Actually, you did not really follow the SMIv2 structure, or you
> would have used modules instead of submodules.

I wrote "reflects the modular structure of the SNMPv3 MIB modules". I
think my statement was precise.

> The main module (ietf-snmp) is just a list of include-by-revision
> statements.

Correct.

> All submodules have to be present or the tool will exit with a fatal
> error.

Correct.

> I suppose the main module could be altered by vendors (e.g., comment
> out the include-stmts they don't want to support), but that would be
> non-compliant for standard modules.

Yes, it is possible to implement a standard incorrectly.

It does not matter whether a vendor comments out an include statement
or comments out complete containers. The result is equally broken.

/js

-- 
Juergen Schoenwaelder           Jacobs University Bremen gGmbH
Phone: +49 421 200 3587         Campus Ring 1, 28759 Bremen, Germany
Fax:   +49 421 200 3103         <http://www.jacobs-university.de/>

[netmod] comments on draft-ietf-netmod-snmp-cfg--… Andy Bierman
Re: [netmod] comments on draft-ietf-netmod-snmp-c… Juergen Schoenwaelder
Re: [netmod] comments on draft-ietf-netmod-snmp-c… Andy Bierman
Re: [netmod] comments on draft-ietf-netmod-snmp-c… Andy Bierman
Re: [netmod] comments on draft-ietf-netmod-snmp-c… Juergen Schoenwaelder