Peter Tribble wrote: > On Nov 9, 2007 5:28 AM, Rainer Heilke <rheilke at dragonhearth.com> wrote: >> Mark Martin wrote: >>> Gang, >>> >>> I believe there's minimal consensus that start/stop as an alias for >>> enable -t does have some value, if only for historical connotations. >>> After re-reading the discussions here many times, I believe here's how >>> I'd like to proceed. I'm going to assume that this is the correct >>> approach unless someone tells me I will get veto'ed at RTI time. >>> >>> (continuing the list thinking meme) >>> I will create a new svcadm start command which will an alias for enable -st >>> I will create a new svcadm stop command which will be an alias for >>> disable -st >> Thank you. If I understand the -s switch correctly from the man page, I >> think this does capture my preferences. Others will have to confirm/deny >> based upon their own wants. > > I'm still unsure as to the -s bit. The reasoning is based on the observation > that the wrappers around smf present in /etc/init.d generally just use > the plain -t. I think I would expect that svcadm stop/start would behave > the same way as running the old rc scripts with stop and start. >
As my later post indicates, if start/stop are just aliases, then I feel the change is incomplete. I agree with you; start/stop should behave the same way as for the init scripts (and I must not have understood the implications of -s, or they aren't expressed well in the man page). This, I believe, would be the behaviour expected by admins. Rainer -- Mind the gap.
