[netmod] Re: RE I-D Action: draft-ietf-netmod-node-tags-11.txt

[Andy Bierman <[e-mail]>]

On Sat, May 11, 2024 at 11:01 PM Qin Wu <[e-mail]> wrote:

> *发件人:* Andy Bierman [mailto:[e-mail]]
> *发送时间:* 2024年5月11日 1:10
> *收件人:* Qin Wu <[e-mail]>
> *抄送:* Kent Watsen <[e-mail]>; [e-mail]
> *主题:* Re: [netmod] RE I-D Action: draft-ietf-netmod-node-tags-11.txt
>
>
>
>
>
>
>
> On Thu, May 9, 2024 at 7:19 AM Qin Wu <[e-mail]> wrote:
>
> Thank Andy for valuable comments, I have update introduction in the github
> to clarify the motivation and provide more guidance
>
> See diff:
>
>
> https://author-tools.ietf.org/diff?url_1=https://raw.githubusercontent.com/billwuqin/draft-ietf-netmod-node-tags/main/draft-ietf-netmod-node-tags-12.txt
>
>
>
>
>
>
>
> I cannot find any evidence that the IETF module-tags are being used
> anywhere.
>
>
> https://www.rfc-editor.org/rfc/rfc8819.html#name-ietf-yang-module-tags-regis
>
>
>
> Why do we need to go through every leaf in every YANG module and decide if
> it is a metric, log, trace, or info?
>
> Wouldn't every leaf be tagged as 'info'?
>
>
>
> [Qin]:I think you don’t need to go through all leafs and in most cases,
> you just pick the leaf you are interested or you might just pick parent
> leaf (e.g., list, container, etc).
>
> Take user configured tag as an example,
>
> You as user or operators can use edit-data to label all instances of data node and establish the mapping between the label and instance of data node using ietf-node-tags,
>
> Which can be get accessed to other users for data analytics, or decide to where to subscribe the data. These data can be seen as user intent in some sense.
>
>
>
> IMO it makes sense (in some cases) to improve the granularity of the
> module-tag with per-node tagging.
>
> That means using the same tag registry defined in RFC 8819, not inventing
> a new classification system.
>
>
>
> [Qin]:note that what we proposed in this document is to extend module tag module to establish the mapping between tag and leaf.
>
> Or you think we should add one parameter under module list of ietf-module-tags to distinguish module name from node name such as:
>
>     module: ietf-module-tags
>
>       +--rw module-tags
>
>          +--rw module* [name]
>
>             +--rw name          yang:yang-identifier
>
>             +--rw type             //indicate whether it is module or node.
>
>             +--rw tag*          tag
>
>             +--rw masked-tag*   tag
>
> Or you just think we should not introduce yang extension for node-tag, just reuse YANG extension for module-tag, I am open to this.
>
>
>


I cannot find any evidence that the module-tag extension is being used.
There are no standard protocol operations that use module tags.
YANG Doctors are not telling WGs to add module-tag extensions.

The tag mechanism has some value, although variables (i.e. RFC 7952
annotations)
would be much better than tags.

The IETF classifications should not be expanded since they are not even
being used now.
There is still some value in the vendor and user tags.

We have proprietary augments to get* operations and to NACM rules to use a
module-tag as a filter.
Node tags could be more useful. They could provide a simple way to use YANG
Push.
But maybe this makes more sense as a vendor or operator-configured feature.


-Qin
>
>
>
>
>
> Andy
>


Andy


>
>
> *发件人:* netmod [mailto:[e-mail]] *代表 *Andy Bierman
> *发送时间:* 2024年2月21日 8:18
> *收件人:* Kent Watsen <[e-mail]>
> *抄送:* [e-mail]
> *主题:* Re: [netmod] RE I-D Action: draft-ietf-netmod-node-tags-11.txt
>
>
>
>
>
>
>
> On Tue, Feb 20, 2024 at 8:03 AM Kent Watsen <[e-mail]> wrote:
>
> Juergen, Tom, Andy,
>
> Gentle reminder.
>
>
>
> I read draft-11.  It is an improvement. No objections.
>
> There are 4 IETF tags defined:  metrics, logs, traces, info
>
> I do not see much value in these standard tags, but more guidance and
> explanation would help.
>
>
>
> Since the IETF does not define any protocol usage of tags, there is little
>
> impact caused by this draft, except for the additional administrative
> overhead.
>
>
>
>
>
> Kent // shepherd
>
>
>
> Andy
>
>
>
> > On Nov 14, 2023, at 4:49 PM, Kent Watsen <[e-mail]> wrote:
> >
> > Juergen, Tom, Andy,
> >
> > The previous WGLC for this draft didn’t succeed due to your comments.
> > Qin’s update (1) below removes all the (metric) specific node-tags.
> > All that is left now is the generic mechanism for tagging nodes.
> > Can you confirm that this update (-11) addresses your concerns?
> >
> > Thanks,
> > Kent
> >
> >
> >> On Oct 23, 2023, at 6:28 AM, Qin Wu <bill.wu=
> [e-mail]> wrote:
> >>
> >> v-11 addresses comments during WGLC, the main changes include:
> >> 1. Remove all specific metrics from both terminology section and
> section 9.2 on IETF YANG Data Node Tags Registry based on WGLC discussion.
> >> 2. Align with Open Telemetry and Open Metrics open source
> implementation specification, introduce traces, log for data nodes
> classification.
> >> 3.Fix normative reference issues in section 9.2.
> >>
> >> -Qin
> >> -----邮件原件-----
> >> 发件人: I-D-Announce [mailto:[e-mail]] 代表
> [e-mail]
> >> 发送时间: 2023年10月21日 17:56
> >> 收件人: [e-mail]
> >> 抄送: [e-mail]
> >> 主题: I-D Action: draft-ietf-netmod-node-tags-11.txt
> >>
> >> Internet-Draft draft-ietf-netmod-node-tags-11.txt is now available. It
> is a work item of the Network Modeling (NETMOD) WG of the IETF.
> >>
> >>  Title:   Node Tags in YANG Modules
> >>  Authors:  Qin Wu
> >>           Benoit Claise
> >>           Mohamed Boucadair
> >>           Peng Liu
> >>           Zongpeng Du
> >>  Name:    draft-ietf-netmod-node-tags-11.txt
> >>  Pages:   30
> >>  Dates:   2023-10-21
> >>
> >> Abstract:
> >>
> >>  This document defines a method to tag nodes that are associated with
> >>  the operation and management data in YANG modules.  This method for
> >>  tagging YANG nodes is meant to be used for classifying either data
> >>  nodes or instances of data nodes from different YANG modules and
> >>  identifying their characteristic data.  Tags may be registered as
> >>  well as assigned during the definition of the module, assigned by
> >>  implementations, or dynamically defined and set by users.
> >>
> >>  This document also provides guidance to future YANG data model
> >>  writers; as such, this document updates RFC 8407.
> >>
> >> The IETF datatracker status page for this Internet-Draft is:
> >> https://datatracker.ietf.org/doc/draft-ietf-netmod-node-tags/
> >>
> >> There is also an HTMLized version available at:
> >> https://datatracker.ietf.org/doc/html/draft-ietf-netmod-node-tags-11
> >>
> >> A diff from the previous version is available at:
> >>
> https://author-tools.ietf.org/iddiff?url2=draft-ietf-netmod-node-tags-11
> >>
> >> Internet-Drafts are also available by rsync at:
> >> rsync.ietf.org::internet-drafts
> >>
> >>
> >> _______________________________________________
> >> I-D-Announce mailing list
> >> [e-mail]
> >> https://www.ietf.org/mailman/listinfo/i-d-announce
> >>
> >> _______________________________________________
> >> netmod mailing list
> >> [e-mail]
> >> https://www.ietf.org/mailman/listinfo/netmod
> >
> > _______________________________________________
> > netmod mailing list
> > [e-mail]
> > https://www.ietf.org/mailman/listinfo/netmod
>
>