On Thu, May 1, 2014 at 2:01 AM, t.petch <[email protected]> wrote: > ----- Original Message ----- > From: "Jeffrey Haas" <[email protected]> > To: "t.petch" <[email protected]> > Cc: <[email protected]> > Sent: Thursday, May 01, 2014 1:34 AM > > On Fri, Mar 28, 2014 at 10:48:13AM +0000, t.petch wrote: > > > This tri-partite split has been the norm for Ops Area for many years > > > now. > > > It may or may not be appropriate for I2RS but I do think that I2RS > > > should > > > at least take cognizance of this in deciding what terms to use > > > > I think this is one of the clearest description of one of our > problems - and > > thus requirements - that I've seen thus far. :-) > > Yeah, problems are my speciality - I then look to Andy, Juergen, Lada > and Martin for solutions:-) > > But, Jeff, you then say, in a parallel post > > "The issues then comes back to the ones noted in > draft-bjorklund-netmod-operational-00: How do we distinguish > operational, > config or ephemeral config states? " > > whereas I see operational state, config and read-only statistics, with > state (unqualified) referring to the first and last collectively! > 'ephemeral' does not appear in Martin's I-D, nor would I expect it to - > I don't see it as a concept in YANG/NETCONF. > > There is a separate issue of persistent and ephemeral in YANG, or, > arguably, in NETCONF, which is also not documented AFAICS. This is > probably of less interest to I2RS at this time. > > N/Y treats config=true data nodes as special. Collectively these data nodes form a configuration datastore. This is only important for validation purposes -- determining what is a "valid" running configuration.
If there is one running datastore, then, presumably, it is persistent > (across reboots) - the documents appear not to say. If there is running > and startup, presumably startup is persistent and running is not. But > if you have running and acme-special datastores, then which is > persistent? This is one of several issues, like operational state, that > have surfaced from time to time and, for me, have not got nailed down as > well as I would like, and so - surface from time to time. > > There is no "acme-special" sort of datastore. There are only the standard datastores. If the server supports the "startup" datastore, then all configuration changes are saved manually to NV-storage. If the client does not save changes, they are lost at the next reboot. If the server does not support the "startup" datastore then the NV-storage must mirror the running configuration. All config changes are saved automatically and immediately. Operational state is different than config because only config is validated according the the YANG constraints. The operational datastore may need some validation (e.g. field syntax check) but it is not in anyway considered when validating the running datastore. IMO the difference between ephemeral config and operational state is not the NV-storage, but the validation procedure that accepts or rejects the write request. Tom Petch > > > -- Jeff > > Andy > _______________________________________________ > i2rs mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/i2rs >
_______________________________________________ i2rs mailing list [email protected] https://www.ietf.org/mailman/listinfo/i2rs
