Nicolas Williams wrote: > On Tue, Jun 03, 2008 at 07:11:21PM -0700, Jordan Brown wrote: >> Nicolas Williams wrote: >>> Well, yes; that implies strict shutdown ordering, at least for the last >>> one or two services. >> Well, sure. Isn't sequencing startup and shutdown SMF's job? > > I was alluding to the fact that for a long time SMF didn't (does it > now?) get shutdown ordering right.
Can you be specific, rather than alluding? SMF does shutdown in reverse dependency order as of the fix for 6207705, which was 3 years ago in onnv and 2 years ago in s10. (Which means that I think the initial problem could be solved today by having a service with :true as the start method, a stop method to chat to the UPS, and a *dependent* (not a dependency) set for network/loopback. I think inittab-based solutions will also continue to work, but admit I'd need to re-check the code again to be sure. Perhaps not ideal, but not any more challenging than creating a service with Jim's posited special flag either. [1]) liane [1] I have a concern about a 'run-last' mechanism, given that there were two posited consumers for it within 24 hours of it being proposed, which is the same thing that happens every time such a mechanism is proposed. The specific request in this thread may or may not have been sound, but the time-to-second-consumer is standard, which means that I worry that there will never really be a single run-last service on any system, and that those run-last services will prove to be buggy in the face of their expectation being violated.
