Re: [netmod] Question draft-wwx-netmod-event-yang

Good point for consideration, Igor.  Sorry for late follow up.
I can add several consideration points into objective section
And generalized other objectives in the objective section.

-Qin
发件人: Igor Bryskin [mailto:i_bryskin@yahoo.com]
发送时间: 2019年11月19日 23:59
收件人: Qin Wu <bill.wu@huawei.com>; draft-wwx-netmod-event-yang <draft-wwx-netmod-event-yang@ietf.org>; NETMOD Group <netmod@ietf.org>; Lou Berger <lberger@labn.net>
主题: Re: [netmod] Question draft-wwx-netmod-event-yang

Lou, Qin, All.
I am glad you asked this. I thought about the same thing quite a bit myself. I wish you asked this 2 years back ;)
There are several  things to consider.  To name some:
1. How the client configures the ECAs. Currently  we suggest using NETCONF/RESTCONF,, but any configuration protocol could be, of course used just as well.
2. How do we express ECA components,  such as Action? Today we suggest in the form of NETCONF RPCs (either native, such as edit-config for re-configurations, or YANG model defined, such as TE path computations). But other platforms/modeling languages, such as, most obviously,  gNMI/gRPC/protonufs could be used just as well.
3.How an ECA execution thread talks to the client? Today we suggest using PUSH/PUSH extensions, but this could be expanded to communication channels of other types.
4. Etc.
In short, I like the suggestion very much. My only concern is this: we were told that we are setting already bar too high. It would be better to limit the work's scope to relatively simple PUSH extension to enable some basic RMON, elementary smart filters, etc. Your suggestion pushes the work in quite different direction - generic network ECA style automation.
I'd be happy to work on that, but I suggest to keep the generic architecture in mind as the end game and focus on delivering NETCONF/YANG solution first. This done quickly and properly, the generic extensions would be relatively straightforward IMHO.
Igor
Get Outlook for Android<https://aka.ms/ghei36>

On Tue, Nov 19, 2019 at 5:23 AM -0500, "Lou Berger" <lberger@labn.net<mailto:lberger@labn.net>> wrote:

Qin,

     if you want to  be agnostic of protocol, the document will need to

be updated from its current netconf specific form.

Thanks,

Lou

On 11/19/2019 2:45 AM, Qin Wu wrote:

> Talking with our coauthors, we believe should not limit usage of this

> model only to netconf. Other protocols could be applied. Let us know

> if any update is needed.

>

> ------------------------------------------------------------------------

> 吴钦 Qin

> Mobile：+86-13914734360 (Mobile Number)

> Email：bill.wu@huawei.com<mailto:bill.wu@huawei.com>

>

>

>

> *发件人： *Lou Berger>

> *收件人：

> *draft-wwx-netmod-event-yang >;NETMOD

> Group>

> *主题： *[netmod] Question draft-wwx-netmod-event-yang

> *时间： *2019-11-19 16:23:41

>

> Hi,

>

>      Authors, is there a reason tat the body of the text is only focused

> on netconf?  Why would you limit this work to any specific protocol?

>

> Thanks,

>

> Lou

>

> _______________________________________________

> netmod mailing list

> netmod@ietf.org<mailto:netmod@ietf.org>

> https://www.ietf.org/mailman/listinfo/netmod

_______________________________________________

netmod mailing list

netmod@ietf.org<mailto:netmod@ietf.org>

https://www.ietf.org/mailman/listinfo/netmod