Re: Why are we doing netconf?

[Chris Lonvick <clonvick@cisco.com>]

Hi,

I, too, have a concern about sending notification messages through 
netconf.

syslog is frequently the oldest piece of running code on many boxes. 
Since it works and is well understood by automated parsers, there is 
little impetus to change the log messages or the underlying protocol. 
That's a challenge that we knew about when we started the WG.

The BoFs and email threads from before the WG was formed showed that many 
people have strong opinions about how the format of a syslog message, 
including the payload, should be structured.  We figured that a WG would 
never come to consensus on that problem.  From that, the syslog WG was 
charterd in the security area to provide security mechanisms for the 
syslog protocol.  We tried working with the rather undefined header 
information but came to the conclusion that we couldn't provide a good set 
of documents without some work on standardizing the header fields.  That 
allowed us to add structured data (Rainer has described that).

When we started doing that, we also separatied the transports from the 
protocol.  The netconf WG could use that format within netconf if desired. 
This will allow the message format to be utilized within new secure 
transports as they are developed (e.g., dtls) without having to make 
changes to the message body.

If the netconf WG decides to provide notifications, we hope that the WG 
will consider the use of The syslog Protocol within whatever transports 
you decide upon.  I believe that the framework provided within The syslog 
Protocol will allow you to define message bodies as you see fit.


We are making progress on the current set of IDs within our new charter. 
We have:

For the basic protocol and transports

The syslog Protocol
http://www.ietf.org/internet-drafts/draft-ietf-syslog-protocol-16.txt

Transmission of syslog messages over UDP
http://www.ietf.org/internet-drafts/draft-ietf-syslog-transport-udp-06.txt
(this provides compatability with existing syslog/udp)

TLS Transport Mapping for SYSLOG
http://www.ietf.org/internet-drafts/draft-ietf-syslog-transport-tls-00.txt


We are still going to provide Signed syslog Messages
http://www.ietf.org/internet-drafts/draft-ietf-syslog-sign-17.txt
as a way to ensure end-to-end security of the messages.  We are also going 
to provide a revision to RFC 3195 that supports The syslog Protocol 
(standardized headers, larger payload size, etc.).  The driver for 
revising RFC 3195 is that it is being implemented in the medical health 
industry and they do require larger payloads.


As far as the interest level goes, the IESG felt that there was enough 
support and interest to develop code to allow us to recharter.  I agree 
with that.

Thanks,
Chris





On Tue, 28 Mar 2006, Andy Bierman wrote:

> David Harrington wrote:
---remainder deleted for brevity---

--
to unsubscribe send a message to netconf-request@ops.ietf.org with
the word 'unsubscribe' in a single line as the message text body.
archive: <http://ops.ietf.org/lists/netconf/>