Re: In Service Software Upgrade Draft

So you're saying that some of the internal state
(counters, statistics) may not be maintained across 
the up-grade?

That's something to note in the report, IMO.
Al

> -----Original Message-----
> From: Fernando Calabria (fcalabri) [mailto:[email protected]]
> Sent: Tuesday, July 08, 2014 8:28 AM
> To: MORTON, ALFRED C (AL); Banks, Sarah; [email protected]
> Subject: Re: [bmwg] In Service Software Upgrade Draft
> 
> The point we were discussing was the fact that some implementations will
> reset the session timers / statistics. While the ISSU process may / should
> not affect the control and forwarding plane, in  some implementations the
> timers for the session up-time and other statistics   are reset . Should
> we address this as part of the Exit criteria for an ISSU process ?
> 
> 
> Rgds
> 
> 
> Fernando
> 
> 
> 
> On 7/8/14, 8:16 AM, "MORTON, ALFRED C (AL)" <[email protected]> wrote:
> 
> >> -----Original Message-----
> >> From: bmwg [mailto:[email protected]] On Behalf Of Banks, Sarah
> >...
> >> 	The ISSU draft is in a pretty solid state, from the authors point of
> >> view. We wanted to solicit a bit of feedback, and make sure we've
> >>covered
> >> all bases. There was some discussion around the authors as to adding
> >>some
> >> text around timers and counters in the draft - do you trust the DUT?
> >> Should you confirm these outside of the DUT? Etc. Thoughts?
> >>
> >> Kind regards,
> >> Sarah & Fernando & Gery
> >>
> >
> >I gave this topic some thought last week:
> >
> >http://tools.ietf.org/html/draft-morton-bmwg-virtual-net-01#section-4.2
> >
> >   ... external observations remain essential as the
> >   basis for Benchmarks.  Internal observations with fixed specification
> >   and interpretation may be provided in parallel, to assist the
> >   development of operations procedures when the technology is deployed,
> >   for example.  Internal metrics and measurements from Open Source
> >   implementations may be the only direct source of performance results
> >   in a desired dimension, but corroborating external observations are
> >   still required to assure the integrity of measurement discipline was
> >   maintained for all reported results.
> >
> >This text is new and isn't agreed yet, it's my thoughts on the topic.
> >Al
> >(as a participant)
> >
> >_______________________________________________
> >bmwg mailing list
> >[email protected]
> >https://www.ietf.org/mailman/listinfo/bmwg