> From: Jeremy Huntwork <[email protected]> > Date: Fri, 28 Oct 2016 17:39:33 +0000 > Subject: Re: [lfs-dev] s6 > . . > [...] . I've actually implemented s6 in > another system
(lightcube os, or roundcube os ?). > and while certainly there are some new concepts and ideas to > learn, I didn't really find it headache. A lot of the design just makes > sense to me. One part that did require some re-thinking was execline and > the way it chains one command to another through a series of execs. Also, > I'd say the design and implementation of s6 is _very_ transparent. The > author's site has both documentation about run-time use of all of its > components, but also documentation about the reasoning behind the design > choices made. > > Anyway, I'm much more interested in technical comments or opinions about > s6's design. As I said, a lot of it makes sense to me, and feels right, but > there are some concepts which are a departure from some long-held ideas and > implementations. > In your experience: if, say, a server daemon crashes, then how well, if at all, does s6: -- * clean-up the debris; * restart the daemon cleanly; and handle properly and dependent-daemon stuff; * avoid getting into a (runaway) loop of persistent crashes; * log info on the above stages; * notify root/&c of the above. -- rgds, akh -- -- http://lists.linuxfromscratch.org/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
