Kent:
I am pleased to see your progress on draft-dsdt-netconf-restconf-nmda-00txt.
You asked me to suggest comments on what would be necessary in the following
cases:
1) Normal,
2) Dynamic datastore
3) Ephemeral dynamic datastore
#1) Normal
Since the current description does not allow dynamic and ephemeral dynamic,
I think this is fine.
#2) Dynamic I2RS datastore which is not ephemeral
These changes to restconf for the nmda-00.txt did not include the following
comments from draft-hares-netconf-i2rs-restconf-02.txt:
1. I2RS Normal datastore must redefine be able to redefine a mask to
the entity tag to encode <client-id> <priority-id>
Section 3.5.2 of RFC8040 states:
"The "ETag" header field can be used by a RESTCONF client in
subsequent requests, within the "If-Match" and "If-None-Match" header
fields." - This is great, because the I2RS requirements need a match on
this
field on the client id, the priority-id or both.
Section 3.5.3 of RFC8040 states:
"This entity-tag is only affected by configuration data resources and
MUST NOT be updated for changes to non-configuration data."
This must be revised to indicate that for dynamic datastores.
any data written from any origin must be updated. Does this belong in
the main document
or in the I2RS ephemeral datastore definition.
2. Dynamic datastores are excluded from this draf
If I2RS creates a dynamic datastore draft, should it discuss how
section
* "with defaults" - How RFC8040 section 3.5.4 is adapted for dynamic
work,
* "with validation" - How validation is normally handled,
* "how client edit collisions are handled - RFC 8040 3.4 or different
method
3. Ephemeral dynamic datastores are excluded from this draft
If I2RS creates an dynamic ephemeral datastore draft, then it should
discuss the following:
. "with defaults" setting - that section 3.5.4 to not keep data
after a reboot, but to provide some default setting upon being downloaded.
. "entity tag" (RFC 3.5.2) - adaptation to provided client and
priority,
. Restriction of ephemeral dynamic datastores to exclude the
"rollback
I hope this helps.
Sue Hares
_______________________________________________
i2rs mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/i2rs
