Re: few comments on the prot-06 draft

[Andy Bierman <ietf@andybierman.com>]

Steven Berl (sberl) wrote:

>  
>
> > -----Original Message-----
> > From: owner-netconf@ops.ietf.org 
> > [mailto:owner-netconf@ops.ietf.org] On Behalf Of Andy Bierman
> > Sent: Thursday, May 05, 2005 9:36 AM
> > To: netconf
> > Subject: few comments on the prot-06 draft
> >
> > Hi,
> >
> > Throughout the text, whenever a parameter value is mentioned, 
> > such as test-then-set, it should appear in single quotes, 
> > e.g., 'test-then-set'.  Some text like this could be 
> > confusing if you don't have all the tag names memorized yet ;-)
> >
> > The 'confirm-commit' parameter in the XSD is incorrect.
> > There is no 'type' attribute defined so an empty element is 
> > the only allowed construct.  The following text is needed
> > in the 'confirm-timeout' element:   type="xs:positiveInteger"
> >    
>
> I would have thought it should be type="xs:duration".
>  
Yikes!  This is way more complicated than we need!
(Look at the example in the Primer Part 0 spec:
P1Y2M3DT10H30M12.3S is 1 year, 2 months, 3 days, 10 hours, 30 min. and 
12.3 sec

I don't think we need a minimum interval less than one second or
a maximum interval over billions of seconds.  I prefer to keep the
encoding as simple as possible -- which is a positiveInteger.

Andy

>  
>
> > It appears an issue (which was brought up on the mailing 
> > list) slipped through the cracks, which would cause a semantic change:
> >
> >  - PROT, sec. 8.4.5.1  (confirmed commit parameters)
> >    units of confirm-timeout not granular enough
> >    
>
> Xs:duration gives all the granularity you could ever want from years
> down to milliseconds.
>
> -steve
>
>  
>
> > I strongly agree with the proposal to change the units of the 
> > confirm-timeout parameter from minutes to seconds.  There are 
> > already use cases where the minimum rollback trigger of one 
> > minute is too slow.  Networks get faster not slower, so we 
> > should fix this now before it's too late.  The default can 
> > still be 10 minutes (value == 600 instead of 10).
> >
> > Please comment on the mailing list regarding this issue, so 
> > we can decide if there is consensus to make this change.
> >
> > thanks,
> > 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/>
> >
> >    
>
> --
> 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/>
>
>
>  


--
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/>