[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: draft-shafer-netconf-syslog-00.txt

Martin Bjorklund wrote:

Phil Shafer <phil@juniper.net> wrote:

Vincent Cridlig writes:
I don't really understand the design choice of the get-syslog-streams operation because I think it breaks the idea of the get-config operation. 
Syslog stream definitions are not limited to configuration, since
the device may define default streams.


So a manager can do <get> to retreive this info.

Suppose an agent actually did make this configurable.  Without a
standard data model, the vendor will have to invent his own model for
this data.  That means that different vendors will use different
models.  So even if there's a standard way to view the data
(get-syslog-streams), there's no standard way to write it.
Furthermore, this data will be available through two different rpcs,
get-syslog-streams and with a vendor-specific namespace through
get-config.

If on the other hand a standard data model is used, there would be a
standard way to read (<get>) and write (for the agents that support
it).



IMO this is a serious issue.
I completely agree with Martin.
What is the excuse for NETCONF to ignore configuration
and do read-only data models instead?  Should we tell
the ADs and rest of the IESG "That network configuration
problem is really hard -- you should get a WG to do something
about it! Good think we only have to worry about monitoring
in the NETWORK CONFIGURATION PROTOCOL WG."

For years vendors would not do read-write because SNMP was
insecure (but passing cleartext telnet passwords around
didn't bother them too much at the time.)  Then the excuse
was "SNMP is too hard, XML is easier."  Now what's the excuse?
It's time to put up or shut up wrt/ standards based configuration.







/martin


Andy



--
to unsubscribe send a message to netconf-request@ops.ietf.org with
the word 'unsubscribe' in a single line as the message text body.
archive: <http://ops.ietf.org/lists/netconf/>