dgriffi | if the deb file has nothing to do with systemd, can I just drop in a regular Debian deb into a devuan system and expect things to work ok? | 01:51 |
---|---|---|
mason | dgriffi: You'll get regular .deb files through the Devuan mirrors as it is. | 01:51 |
mason | Unless you're talking about something third-party, in which case it's really anyone's guess what they did. | 01:51 |
dgriffi | there's an ongoing problem in MATE that makes the aspect ratio of the pager too narrow no matter what you do. the solution, so far, is to downgrade three libraries and peg them to not be upgraded | 01:52 |
dgriffi | I installed debuan just now and am facing applying that fix. I looked for the downgraded libs in deb.devuan.org's pool directories and didn't find them. | 01:53 |
fsmithred | that's because most of "our" packages are in the debian repos. | 01:54 |
fsmithred | and we pull from there. | 01:55 |
dgriffi | no repackaging or rebuilding? | 01:55 |
fsmithred | Only a couple hundred packages are forked - those are the ones in our repos. | 01:55 |
dgriffi | I see. | 01:55 |
fsmithred | and the server merge them so it appears to be one repo | 01:55 |
fsmithred | so in your case, pulling the debs from packages.debian.org is the correct thing to do (unless you can get apt to do it for you) | 01:56 |
fsmithred | brb | 01:56 |
dgriffi | I manually pulled the debs to apply the fix in the first place. | 01:58 |
dgriffi | see https://lists.debian.org/debian-mate/2022/11/msg00014.html for what I'm up to. | 01:58 |
dgriffi | Now I'm having a weird problem with the SSD in my laptop. it's stuck in some sort of sleep-yet-not-sleep such that I have to set SATA mode to "compatible" rather than "AHCI". | 01:59 |
dgriffi | not really relevant here, but it did give me a nasty time of trying to figure out wtf was going on. | 02:00 |
* dgriffi is pleased with how peppier this devuan install feels than does debian. | 02:00 | |
dgriffi | now.. for shits and giggles... what would it take to add to devuan the option to use systemd? | 02:01 |
fsmithred | can't be done | 02:02 |
fsmithred | devuan with systemd has a special name | 02:02 |
fsmithred | Debian | 02:02 |
fsmithred | packages that require systemd are necessarily banned from the repo | 02:03 |
dgriffi | fsmithred: my point is to work at the problem of offering choice in the reverse direction | 02:03 |
fsmithred | right now the installer includes the choices of sysvinit, openrc or runit | 02:04 |
dgriffi | and wasn't there some back and forth about getting debian to support choices like that? | 02:05 |
golinux | dgriffi: If Debian had supported and offered choice, there would have been no need for Devuan to fork. | 02:05 |
golinux | Yeah back in 2014 | 02:05 |
dgriffi | golinux: I know. | 02:05 |
fsmithred | I don't think their installer offers a choice of init system, but they do have the same ones we have. | 02:06 |
golinux | Hold on a sec I have the canonical link explaining that . . . | 02:06 |
dgriffi | golinux: my point is that if somehow systemd could be made an option in devuan, then we could say "hey, debian! you said this couldn't be done! lookie here!" | 02:06 |
golinux | https://forums.debian.net/viewtopic.php?t=120652 | 02:07 |
golinux | Doesn't MX Linux already do that? | 02:07 |
fsmithred | We're going off-topic here. They already know what can and can't be done. | 02:08 |
golinux | But we are now off topic | 02:08 |
golinux | :) | 02:08 |
fsmithred | will or won't is more the problem | 02:08 |
golinux | Why would we want to offer an inferior init option? | 02:10 |
dgriffi | golinux: to stick it to the systemd true-believers | 02:11 |
dgriffi | golinux: to show them that it doesn't have to be systemd today tomorrow and forever. | 02:12 |
rustyaxe | No thanks. | 02:12 |
rustyaxe | We dont need no stinkin systemdos | 02:12 |
gnarface | they literally designed it to make that impossible on purpose | 02:12 |
dgriffi | gnarface: I can understand and appreciate a system that once it's systemd, it cannot be switched to sysvinit or whatever. | 02:13 |
fsmithred | ? | 02:14 |
fsmithred | we have directions for migration from debian to devuan for every release | 02:14 |
dgriffi | gnarface: what I'm specualating is that you could possibly have the choice of many inits on install, one of them is systemd and the whole shebang works | 02:15 |
golinux | We have no interest in systemd | 02:15 |
dgriffi | fsmithred: I guess I misinterpreted the "designed it to make that impossible on purpose" | 02:15 |
fsmithred | we have to fork some packages to remove their dependence on systemd | 02:16 |
fsmithred | and we have to supply init scripts that disappear | 02:16 |
dgriffi | golinux: I have no interest in it either... for doing actual work. My interest in very narrowly-defined with the intent of pissing in poeterring's cheerios | 02:16 |
fsmithred | although there is now a package called orphan-sysvinit-scripts that has a lot of them. | 02:16 |
golinux | That us extremely childish and not productive. | 02:16 |
dgriffi | fsmithred: I found that one and asked why it's not a prereq for de-systemd'ed debian machines. I got a lot of "but it's not" nonanswers. | 02:18 |
golinux | us > is | 02:18 |
dgriffi | golinux: why not? | 02:18 |
fsmithred | I think our continued existence does that. | 02:18 |
golinux | Because it is a waste of time | 02:19 |
fsmithred | put energy into continuing our existence | 02:19 |
dgriffi | golinux: how much effort do you suppose would be required to do this? | 02:19 |
golinux | And this discussion belongs in #devuan-offtopic | 02:19 |
golinux | Feel free to make that your pet project. | 02:20 |
* golinux needs a bath . . . later . . . | 02:20 | |
fsmithred | I'm going outside for a bit. | 02:21 |
fsmithred | bbl | 02:21 |
markizano | omg, I been searching at this for a few hours and am baffled by this .... | 05:34 |
markizano | casper is an ubuntu app, not an Debian app ?? | 05:34 |
markizano | trying to build a live-iso and seems casper is the one thing I am missing for this .... | 05:35 |
markizano | target OS in live-iso is a custom debootstrap'd devuan I've compacted into a squashfs | 05:35 |
gnarface | seems that way | 05:39 |
gnarface | it's not on the banned packages list so you can probably make it work | 05:40 |
gnarface | no idea why it's not in debian | 05:40 |
al1r4d | What is casper? markizano | 06:01 |
markizano | al1r4d re: https://launchpad.net/ubuntu/+source/casper/1.445.3 | 06:04 |
markizano | "Run a "live" preinstalled system from read-only media" | 06:04 |
Unit193 | Ubuntu's version of live-config, or a ghost. | 06:04 |
markizano | you install it in the target system and the kernel is able to boot into a special mode that runs from ISO images rather than connecting to hardware directly. | 06:04 |
markizano | I pondered if it might be a fork of live-usb or some package similar. Was going to peek into the source and author | 06:05 |
brocashelm | dgriffi: unfortunately, as systemd continues to encroach much of the gnu/linux ecosystem, there have to be boundaries to ensure no traces of it are present on a distro that uses it by default (e.g. debian, arch, fedora) | 06:22 |
brocashelm | debian 13 trixie will be even more systemd-rotted, so devuan, antix, mx, exist to allow for full functionality of other init systems | 06:22 |
brocashelm | and also, if sysvinit is too old/slow, one could ideally install runit and runit-services, and see how the individual likes it (i run it on all my devuan systems and it's getting a lot better support as of daedalus) | 06:23 |
mason | brocashelm: sysvinit is pretty snappy | 06:31 |
markizano | just made some videos about why I use Devuan. Need to edit and post. Will probably end up doing tomorrow since it's ~23:40 here, LoL | 06:42 |
dgriffi | for a stock install of Devuan using MATE, why would the terminal bell for xterm and MATE terminal not sound? | 21:27 |
djph | dgriffi: pcspkr turned off? | 21:56 |
mason | dgriffi: Given that it's been years since xset b parameters actually functioned, that might be a blessing in disguise. | 22:00 |
golinux | xset b off still works for me. I have it in session and startup | 22:17 |
mason | Yeah, but the tone, duration, and volume haven't worked for a dog's age. | 22:22 |
mason | It'd be neat if they did. My one dive into the code was frightening. | 22:23 |
gnarface | dgriffi: the window manager, the user's shell, or the terminal's own default command-line parameters may suppress it | 22:39 |
gnarface | (may suppress or actually replace with "visual" bell) | 22:40 |
gnarface | as for getting the old-school pc speaker functionality, it still can work if you have the right type of hardware, but there's two implementations in the kernel and you may have to force it to use the right one | 22:40 |
gnarface | the primary distinction is whether the internal speaker control is piggy-backing off the main soundcard or not, but there's some other hardware variations between those major two camps that i've seen | 22:41 |
gnarface | despite that, afaik several years ago the kernel developers just picked the "new" one and forced it on everyone by default even if it's not the best one | 22:42 |
gnarface | i used to be able to recall more configuration details about this, but they've faded as most of my use cases were just based on figuring out how to disable both in every case | 22:43 |
gnarface | first step to restoring legacy behavior i believe is blacklisting one kernel module so it can load the other | 22:45 |
gnarface | but i think there may be certain hardware where you have to pick one or the other | 22:46 |
gnarface | as in, you may have to disable the onboard soundcard to get linux to let you load the old pc speaker | 22:46 |
gnarface | not sure about that though, been a while like i said | 22:47 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!