It seems to be a bug to me and not a RFE. -- Renaud
Christine Tran wrote: > David Bustos wrote: >> Quoth Christine Tran on Wed, Feb 20, 2008 at 10:53:45AM -0500: >>> It's because of some error. I can, of course, pkill -c <ctid> everyone >>> (it's in svcs -pv, that makes it easy for me) but I thought what's the >>> harm in providing the :kill feature in the SMF framework. >> If you're saying that we should kill all processes in the contract if >> the start method fails, then that sounds like a reasonable change to me. >> Please file a change request. >> > > Well, I hesitate to file RFEs for something that is easily done by some > other administrative means, and I'm not sure how much work would have to > be done by the SMF team to fix this. Especially when there's so many > other useful things that could be added to SMF. So, open call to the > SMF community: pile-on to this thread if you think SMF should be able > to kill all processes in a transient service if the start method fails. > Right now, there's no :kill for transient services. > > Just because I think it's useful doesn't mean most people find it useful. :) > > CT > _______________________________________________ > smf-discuss mailing list > smf-discuss at opensolaris.org
