Not sure what happened with the messed'd up formatting. Resending with some
more <CR><RET>'s in between lines...
Sorry for the confusion.
Robert
This is the intended post:
***********************
"Recipients of a merge patch document
RP> i.e., the function inside the device responsible for generating
RP> the config data nodes for a particular instance
determine the exact set of changes being requested by comparing the content of
the provided patch
RP> i.e., the few data nodes specific to a particular instance (cf "dark
blue dots" in slide 12)
against the current content of the target document.
RP> i.e., the copy of the template (cf "green dots" in slide 12)
If the provided merge patch contains members that do not appear within the
target, those members are added. If the target does contain the member, the
value is replaced."
NB: the "null" value in the patch further mentioned in RFC7396 is something
that does not fit in the proposed template mechanism, though, because the
template (green) and instance (blue) use the same data nodes and in a YANG
model, there is no data node value that means "remove the data node itself")
*********************
-----Original Message-----
From: Robert Peschi (Nokia)
Sent: Friday, July 26, 2024 10:13 AM
To: Carsten Bormann <[email protected]>
Cc: Jürgen Schönwälder <[email protected]>; Italo Busi
<[email protected]>; [email protected]
Subject: RE: [netmod] Yang Scalability
Hi Carsten,
> So a mental model of “default-value + merge-patch where needed” (RFC 7396)
> would work?
I like the term "mental model" that you coined !
I am not sure which environment RFC7396 is intended for, but indeed this quote
from RFC 7396 is exactly what I meant:
"Recipients of a merge patch document
RP> i.e., the function inside the device responsible for generating
RP> the config data nodes for a particular instance determine the exact
set of changes being requested by comparing the content of the provided patch
RP> i.e., the few data nodes specific to a particular instance (cf "dark
blue dots" in slide 12) against the current content of the target document.
RP> i.e., the copy of the template (cf "green dots" in slide 12) if the
provided merge patch contains members that do not appear within the target,
those members are added. If the target does contain the member, the value is
replaced."
NB: the "null" value in the patch further mentioned in RFC7396 is something
that does not fit in the proposed template mechanism, though, because the
template (green) and instance (blue) use the same data nodes and in a YANG
model, there is no data node value that means "remove the data node itself")
Best regards,
Robert
-----Original Message-----
From: Carsten Bormann <[email protected]>
Sent: Thursday, July 25, 2024 11:24 PM
To: Robert Peschi (Nokia) <[email protected]>
Cc: Jürgen Schönwälder <[email protected]>; Italo Busi
<[email protected]>; [email protected]
Subject: Re: [netmod] Yang Scalability
[You don't often get email from [email protected]. Learn why this is important at
https://aka.ms/LearnAboutSenderIdentification ]
CAUTION: This is an external email. Please be very careful when clicking links
or opening attachments. See the URL nok.it/ext for additional information.
On 25. Jul 2024, at 16:59, Robert Peschi (Nokia) <[email protected]>
wrote:
>
> The actual data nodes of an instance to be configured are internally
> generated from a “photocopy” of the data nodes of an original configuration
> (template, green dots on slide 12) that is customized here and there (cf blue
> dots on slide 12) for this specific instance. The gain in the running data
> store size is that
> • only one copy of the many green dots of one (or say, a few) templates
> need to be persistently configured and stored on the device
> • only a few blue dots need to be persistently configured and
> stored on the device, different for each instance (many ones)
So a mental model of “default-value + merge-patch where needed” (RFC 7396)
would work?
Grüße, Carsten
_______________________________________________
netmod mailing list -- [email protected]
To unsubscribe send an email to [email protected]
