Quoth Michael Shapiro on Fri, Dec 08, 2006 at 04:07:27PM -0800: > > > Dan Mick wrote: > > >> It seems to me tha the problem is there are two states and three cases: > > >> > > >> 1) administrative disablement because of administrator preference > > >> 2) "maintenance" because of a correctable error (s/w or h/w config, > > >> dependency problems) > > >> 3) "can't run" because a required h/w or condition is not present (not > > >> right platform, absence of device, absence of particular bus > > >> technology, etc.) ... > I agree with Dan's list, and I agree with Richard about (3) above: > the proper implementation of (3) is offline because of a missing > dependency, whether it be h/w or s/w. And of course since our > dependencies are expressed as FMRIs they can be any kind of thing.
The problem with offline is that we consider it to be an erroneous state. That could be changed, but what if SMF allowed developers to associate (Enhanced) profiles with hardware? A sort of NWAM for hardware. Presence of the BMC device could be associated with a profile which enables ipmievd. The default policy would be "ipmievd is only enabled when the BMC is present", which is presumably what people really want, but the current meaning of "svcadm enable ipmievd" (I want this service to run all the time) would be preserved (in the absence of the device the service would either fall into maintenance or go to offline if we have device dependencies), and the default policy could be restored by deleting the general/enabled property from the customization profile (the "local" profile in my design). (I imagine I'll provide an easy way to do this with svcadm.) David
