rpc.kstatd: Remote Kstat Access Service [PSARC/2008/502 FastTrack timeout 08/13/2008]

[James Carlson]

Andrew Gabriel writes:
> James Carlson wrote:
> > Whether someone can convince the kstat suppliers to hold enough stable
> > to make this useful for some application is perhaps an issue for some
> > other project.
> 
> I think this is a really bad architecture for a network protocol.

Indeed.  And I doubt this case really makes it as obvious enough to be
a fast-track.

> I 
> would like PSARC to make it clear that kstat suppliers should in no way 
> be constrained to avoid breaking this protocol or any users of this 
> protocol.

I think that's already the case.  Unless the users of this interface
can either convince those kstat suppliers to assert higher stability
or to establish contracts for use in particular applications, the
kstat suppliers are under no obligation whatever to avoid breaking
applications built using this new feature.

(Anything else would be unfair to the implementors of those kstats --
how could they have known that someone would camp out on their
undocumented debug interfaces?)

-- 
James Carlson, Solaris Networking              <james.d.carlson at sun.com>
Sun Microsystems / 35 Network Drive        71.232W   Vox +1 781 442 2084
MS UBUR02-212 / Burlington MA 01803-2757   42.496N   Fax +1 781 442 1677