Indeed.
When we start trying to implement some of these features based on the
infrastructure I've added, we should keep an open mind to revisiting the
infrastructure assumptions if it looks like it is not quite what is
needed ... but implementing a "delta(v) on delta(v)" metric was so easy
(after being un-doable from day one), that I'm quietly confident.
On Wed, 2010-11-03 at 17:15 +1100, Mark Goodwin wrote:
> Looks good Ken - in addition to the client context for event tracing,
> I think this also helps pave the way for configurable authentication
> and access control mechanisms on metric subtrees, or maybe per-PMDA.
> .. I'm thinking about it.
>
> Cheers
> -- Mark
>
> _______________________________________________
> pcp mailing list
> pcp@xxxxxxxxxxx
> http://oss.sgi.com/mailman/listinfo/pcp
|