If you are anywhere near KnoxBUG’s meeting place (mid-Tennessee, US), Joe Maloney will be presenting on OpenRC and TrueOS, tomorrow night. See the link for address and times.
10 Replies to “Joe Maloney: Using OpenRC with TrueOS @ KnoxBUG”
Comments are closed.
Are there any talks or indications of replacing the DragonFly boot process with something like OpenRC, nosh, or runit?
Not that I’ve seen. There also hasn’t been anyone having problems with the existing system.
if I post this here, people might not see it. Maybe Justin has a new comment alert in place. Since the changes of late, in this field, regarding trueos, it is now basically possible to use trueos init in freebsd. Now, there is that paper from openbsd folks regarding their “new” rc.d and their portable tool for service management, rcctl, paper that considers many many “solutions” but all of them get dropped off discussion, including this: rc+rcorder, the first modular init that appears in netbsd1.5 and from there in dfly1, later in freebsd5+. What would dfly do? Stay behind, holding hands with netbsd, the last mohican, since there is no one having problems with the existing system?
You sound like you really want to have a new init system. Why?
Ignorance isn’t just bliss anymore, things changed and it happened of course because of a controversed init system that crated a rift into the linux world. It is true that there was a beautiful era that moved slowly in which one could learn so much between two openbsd releases. Now, there is the latest clang in openbsd and not yet in.. dragonfly. Since trueos had to embrace a service manager or a supervision suite to go beyond what netbsd folks think is enough, since openbsd isn’t considering what is nobody-has-problems-with-it for others, aren’t there enough reasons to believe that, suddenly, what we use isn’t on par (anymore)? The init used by the os I use was discarded by THE folks that keep a way of doing it. I use dfly because of a way of doing it. See the reasons Dillon invokes in the rejections of the latest versions of NFS. See Dragonfly Mail Agent, et cetera. See Dfly’s HAMMER. Openrc is basically made by a netbsd developer. Isn’t netbsd living in an ignorance is bliss state? What is netbsd’s way of doing it if there is a netbsd dev that actually starts a linux distro with an init that works without service management (void linux, runit)? Back then there weren’t choices. See the compilers we have today. I follow a small european (I am an eastern european) community called “suckless dot org” and they had a list of software that suck more. That list used to point the GCC. Now it points both the GCC and the CLANG/LLVM.
About marino, is ada the language working in dragonfly?
Did you read that freebsd forum thread that was closed because marino had to explain to someone about the “alternative” compilers?
I am indiferent to openrc, but I don’t like Gentoo’s William H. activity. But, I like Laurent Bercot’- (s6 init and service management approach, he had a lightning talk at a recent linux conf) and Olivier Brunel’- (jjacky, he wrote anopa, a service manager that works on top of s6) -‘s work. s6 is portable to any posix system and I am saying this because I’d like to also point to something, good in my opinion, and not just be one of those, who say we need to scrap what seems old just because the new is new and also smells like raspberries. I am not saying necessarily that we need a new init. I am saying that people are more aware today than they were back then and that we live some really fast times and I am just curious of what will dfly will choose. Maybe dfly will stay as it is, maybe it will change by doing something more similar to openbsd’s change, next year or next decade. All I know is that dragonfly’s move will be a move I’ll be proud of. So I am curious what will that move be.
Init systems have been such a divisive headache in Linux-land that I’m certainly not looking forward to any changes here. That being said, I don’t think it’s on anyone’s radar, simply because it causes more problems than it solves… at least right now.
Justin, you just read the last few phrases, didn’t you? There is svc(8). That’s a starting point for THE dragonflybsd modern init, but since there’s nothing on the sonars..
I read the whole thing.
https://2016.eurobsdcon.org/PresentationSlides/NanakoMomiyama_TowardsFastIPForwarding.pdf
https://sourceforge.net/projects/e1000/files/ixgbe%20stable/
https://events.linuxfoundation.org/sites/events/files/slides/pushing-kernel-networking.pdf
https://people.netfilter.org/hawk/presentations/nfws2016/nfws2016_next_steps_for_linux.pdf
https://people.netfilter.org/hawk/presentations/xdp2016/xdp_intro_and_use_cases_sep2016.pdf
https://www.iovisor.org/technology/xdp
bsd, did you mean to post that here or on the most recent story about network benchmarking?