Re: [netmod] draft-ietf-netmod-syslog-model-12

[t.petch <[e-mail]>]

---- Original Message -----
From: "Clyde Wildes (cwildes)" <[e-mail]>
Sent: Thursday, March 16, 2017 5:58 PM

> Tom,
>
> Inline…
>
> On 3/16/17, 10:04 AM, "t.petch" <[e-mail]> wrote:
>
>     ----- Original Message -----
>     From: "Clyde Wildes (cwildes)" <[e-mail]>
>     Sent: Monday, March 13, 2017 6:11 PM
>
>     > Tom,
>     >
>     > The next revision of the draft-ietf-netmod-syslog-model includes
a
>     change to the Yang tree diagram explanation to include the text
from
>     RFC6087bis.
>     >
>     > RFC5426 is referenced in the model where "Transmission of Syslog
>     Messages over UDP” occurs:
>     >
>     >               case udp {
>     >                  container udp {
>     >                    description
>     >                      "This container describes the UDP transport
>     >                       options.";
>     >                    reference
>     >                      "RFC 5426: Transmission of Syslog Messages
over
>     UDP";
>     >
>     > This is a correct reference.
>
>     Indeed but I don't think that [RFC5426]  is going to work as part
of
>     leaf sig-number-resends
>     while
>
>            leaf structured-data {
>             ...
>            as per RFC5424
>
>     is not the usual style.
>
> [clyde] I was urged to use the descriptions directly from RFC5848 – in
this case from section 6.1.2. Configuration Parameters for Signature
Blocks (page 25)
>
>     sigNumberResends = number of times a Signature Block is resent.
>       (It is recommended to select a value of greater than "0" in
>       particular when the UDP transport [RFC5426] is used.)

Clyde

The semantics are fine but it is the [ syntax ] I think incorrect.

> [clyde] Please suggest a new name for the selector that you think
should be changed. I could not find a reference to this in my notes.
>
>     On another tack, I thought that you agreed to change one of the
>     identifiers in
>
>          grouping selector {
>            description
>
>            container selector {

I made  a note of a comment by AC (but which?) that selector selector
was confusing, which I agreed to, and I thought you agreed too, but I
discarded that thread and cannot now find it (it could be several
versions ago).

Anyhow, I find it confusing and would suggest 'filter' (which RFC5424
uses) not 'selector' .  Since the container appears in the diagram, I
would be inclined to use 'filter' there but it is removing the apparent
ambiguity e.g. of  a usage such as

uses selector {
refine " selector" {

that concerns me more.

Tom Petch

> Thanks,
>
> Clyde
>
>
>     Tom Petch
>
>     > Thanks,
>     >
>     > Clyde
>     >
>     > On 3/1/17, 4:20 AM, "netmod on behalf of t.petch"
>     <[e-mail] on behalf of [e-mail]> wrote:
>     >
>     >     The explanation of the YANG tree diagram is not that in
RFC6087; I
>     think
>     >     that it should be (or else explain why not)
>     >
>     >     I am confused by the variation in the references to RFC in
the
>     modules.
>     >
>     >     I see
>     >     RFC 5424
>     >     [RFC5426]
>     >     RFC5424
>     >
>     >     I think the first correct
>     >
>     >     Tom Petch
>     >
>     >     ----- Original Message -----
>     >     From: "Alexander Clemm" <[e-mail]>
>     >     To: "Kent Watsen" <[e-mail]>;
>     >     <[e-mail]>
>     >     Cc: <[e-mail]>
>     >     Sent: Friday, February 24, 2017 12:25 AM
>     >     Subject: Re: [netmod] WG Last Call for
>     draft-ietf-netmod-syslog-model-11
>     >
>     >     _______________________________________________
>     >     netmod mailing list
>     >     [e-mail]
>     >     https://www.ietf.org/mailman/listinfo/netmod
>     >
>     >
>     >
>
>
>
>