openrc vs sysvinit reboot time on Slackware Virtual Machines
SlackwareThis Forum is for the discussion of Slackware Linux.
Notices
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
If you have any problems with the registration process or your account login, please contact us. If you need to reset your password, click here.
Having a problem logging in? Please visit this page to clear all LQ-related cookies.
Get a virtual cloud desktop with the Linux distro that you want in less than five minutes with Shells! With over 10 pre-installed distros to choose from, the worry-free installation life is here! Whether you are a digital nomad or just looking for flexibility, Shells can put your Linux machine on the device that you want to use.
Exclusive for LQ members, get up to 45% off per month. Click here for more info.
A mixture of sysvinit, OpenRC, and runit would be a major conglomeration of effort but it would institute a major step forward for an alternative init and service supervision system for Slackware, though doing the hand off from OpenRC to Runit service startup would be quite the interesting part.
One effort that would be interesting would be to implement s6, but s6 is currently one of the most difficult init and service supervision systems to set up correctly for any distribution due to complexity. I know of none that use it. However it is one of the most advanced init and service supervision systems of any out there for GNU/Linux.
A mixture of sysvinit, OpenRC, and runit would be a major conglomeration of effort but it would institute a major step forward for an alternative init and service supervision system for Slackware, though doing the hand off from OpenRC to Runit service startup would be quite the interesting part.
One effort that would be interesting would be to implement s6, but s6 is currently one of the most difficult init and service supervision systems to set up correctly for any distribution due to complexity. I know of none that use it. However it is one of the most advanced init and service supervision systems of any out there for GNU/Linux.
I like the cleanliness and simplicity of runit services, very unixy.
There is one distro I know of that uses S6, Obarun.
The Runit package is a mixed bag because there are no runit script sets available officially for Slackware or the much needed extra system utilities. Plus, runit has been updated since the last SBO release to 2.1.2 which has a few bug fixes and updates to the script handling system, the SBo package is far from a viable kit.
Runit for Slackware would effectively require a new pause binary (required for delaying script execution phases), stage 1~3 boot and shutdown/reboot scripts (along with any various staged boot scripts also I have a working model in Runit-for-LFS), init mode changing scripts, a new halt and shutdown binaries (also which I do have as part of Runit-for-LFS to replace sysvinit), new scripts to enable/disable services, as well as a collection of service scripts and log collection scripts (most of which have to be written specifically to the distribution). In short Runit would be a lot of work, if not completely too much work. You're more than welcome to use Runit-for-LFS's init-shim kit to get any missing tools from sysvinit replaced, but the rest will be on you.
VoidLinux uses Runit if you want to have a look into it.
If anything, Aaditya's OpenRC and Bartgymnist's systemd are as close to viable alternatives as you will get.
Why do we need a new pause binary? The run scripts are written in shell, so I thought that sleep would work.
I have successfully reused rc.M and rc.S for stage 1, and I copied the tty scripts from Void Linux for stage 2. For stage 3, I reused rc.6. I wish I had more time to proceed with more scripts, but I lead a very busy life.
Scripts to replace shutdown with runit-init are available on Void Linux but need some tweaking to work on Slackware. I have not done that yet.
May I ask for more information on Runit-for-LFS's init-shim kit? Where can I find it?
I like the cleanliness and simplicity of runit services, very unixy.
There is one distro I know of that uses S6, Obarun.
That's interesting. Thanks for the info; I've been messing around with that (slightly) in a VM. When I get some spare time, I'll see how difficult it is to s6 a Slackware64 installation.
ChrisAbela, thanks for your runit Slackbuild, been running it for a few days now. :-) Very simple to switch init to runit as described in the readme.
Several runit services are available here, have forked it and added some more.
If you are interested we can add your updated runit Slackbuild as well as runit-services to SBo for easy access and consumption. This would give Slackers another init system to use besides sysvinit (default), OpenRC (SBo), and systemd (Dlackware).
ChrisAbela, thanks for your runit Slackbuild, been running it for a few days now. :-) Very simple to switch init to runit as described in the readme.
Several runit services are available here, have forked it and added some more.
If you are interested we can add your updated runit Slackbuild as well as runit-services to SBo for easy access and consumption. This would give Slackers another init system to use besides sysvinit (default), OpenRC (SBo), and systemd (Dlackware).
Thank You for the encouraging remarks.
Yesterday I wrote an email to the current SBo runit SlackBuild maintainer for permission to upload my work to SBo and he has not replied yet.
Gentoo people may say otherwise, but I'm not sold on Gentoo by any means. I will say Aaditya's work with the OpenRC packages has been phenomenal to offer what has been available.
Gentoo is a bit weird. I liked the idea of Gentoo before I tried it, but Gentoo is not what it sold itself to be. I thought the point of Gentoo was maximum control and configuration over your system from the very start of the build to the final system. However, this is not the case at all. Gentoo pretty much tells you what do to from the start and you have to base your experience around the package manager portage that does everything for you after the basic system install. You are pretty much an "emerge" autopilot from that point on, leaving you far less control of your system than expected and mostly automated build processes.
I don't mind Gentoo, I just don't think it is what it sells itself to be. It is not an alternative to Slackware!
Aside from that, it seems Gentoo is slowly moving to SystemD, against it's users wishes. Leaving their own faith in openRC exposed in the process. Anyways, more and more systemD stuff is making it into Gentoo, and now it is installed by default. The question is when does it become the default option?
Anyways, Gentoo was an option for me as a distro because I believed it was a bit like Slackware, and furthermore because I believed they were against SystemD. However, it seems to be incorrect. They are forsaking themselves to move to SystemD and the distro is surely not anything but a package based distro that automates most things, and not at all an alternative to Slackware.
Anyways, I think what I mean to say is that there is no point following up on OpenRC unless some distroes plan on moving from SystemD to OpenRC.
Ps. in the back of my mind was actually how arch linux betrayed their users and themselves by switching to SystemD. Seemingly Gentoo is slowly taking the same route. There are only a few Islands left without SystemD.
Pff, Arch was one of the few sensible distroes, the only option would have been to not switch to SystemD. Arch is selling themselves as a back to basic "raw" style distribution, KISS principle and all. SystemD made no sense, it's against the whole core purpose of Arch.
Even Debian had to kill off Ian Murdoch (their founder) to make the unwanted changed to SystemD, which basically played out like a coup d'etat and a forced dictate to switch Debian to SystemD. That was the high watermark, the change of the tide. Had Debian stayed away from SystemD, there would still be hope for a GNU/Linux world where all end products does not end up relying on a single spiderweb to tie all of their parts together and force them to stay the same.
Quote:
Originally Posted by Lysander666
EDIT: I'm not going there.
Ok, I get it.
Sorry! It was not my intention to go off track here and start an anti-systemd discussion. I was merely trying to point out in context... Bla bla..
Anyways.. I think the point I was originally trying to make is that OpenRC is not worth continuing unless something changes. It seems even Gentoo is giving up on OpenRC..
Quote:
Originally Posted by wikipedia
OpenRC is the default init system of TrueOS[3], Gentoo, Alpine Linux, Parabola GNU/Linux-libre, Artix Linux
That's not exactly a strong fortress to build this on, and when the main proponent seem to be giving it up themselves and slowly moving to SystemD, the hope seems to be slowly dying.
Even Debian had to kill off Ian Murdoch (their founder) to make the unwanted changed to SystemD.
OK, now you're just being silly. I like a good conspiracy theory, but Ian Murdoch being killed so that the distro could switch to systemd is another level of bonkers. Even I don't believe that, and I have a lot of tin foil at home.
OK, now you're just being silly. I like a good conspiracy theory, but Ian Murdoch being killed so that the distro could switch to systemd is another level of bonkers. Even I don't believe that, and I have a lot of tin foil at home.
Haha, sure. It was half a joke, half not. Debian never changed to SystemD under Murdoch, and as far as I remember never had an intention to do so, the opposition was very strong and traditions stood strong. But then he dies and they suddenly change to SystemD?
I am not saying it's a conspiracy, I'm just stating with bravado, what happened, in a very outrageous way.
Haha, sure. It was half a joke, half not. Debian never changed to SystemD under Murdoch, and as far as I remember never had an intention to do so, the opposition was very strong and traditions stood strong. But then he dies and they suddenly change to SystemD?
I am not saying it's a conspiracy, I'm just stating with bravado, what happened, in a very outrageous way.
Your claims are unsubstantiated. Debian 8 "Jessie" was released on April 25 2015 with systemd as the default init; Ian Murdoch died later that year on December 28.
Your claims are unsubstantiated. Debian 8 "Jessie" was released on April 25 2015 with systemd as the default init; Ian Murdoch died later that year on December 28.
LinuxQuestions.org is looking for people interested in writing
Editorials, Articles, Reviews, and more. If you'd like to contribute
content, let us know.