| TwistedFate | Hello | 01:33 |
|---|---|---|
| Xenguy | o/ | 02:05 |
| onefang | Just to remind everyone, exactly one week from now my server, sledjhamr.org, gets shut down coz the server company is moving a bunch of us to a new datacentre. They tell me it'll take eight hours. | 04:02 |
| onefang | They also assure me the latency and speed will stay the same, so you lot can keep recommending people use it as a Devuan mirror. | 04:03 |
| onefang | And for those of us using it for apt-panopticon, there's a bunch of others running it - | 04:05 |
| onefang | https://borta.devuan.dev/apt-panopticon/results/Report-web.html | 04:06 |
| onefang | http://veritas.devuan.org/apt-panopticon/results/Report-web.html | 04:06 |
| onefang | https://mishka.snork.ca/apt-panopticon/results/Report-web.html | 04:06 |
| onefang | https://ap.in.devuan.org/apt-panopticon/results/Report-web.html | 04:06 |
| unclouded | What does this line from /etc/console-setup/cached_setup_keyboard.sh do? "kbd_mode '-u' < '/dev/tty1'". I read that "kbd_mode -u" sets the keyboard mode to UTF-8, but what does the redirection do? | 04:09 |
| gnarface | not sure why but seems to be reading the string console name from stdin instead of passing it with -C | 04:16 |
| Guest9 | gnarface I'm a newb. What's a backports kernel? How would I do this? | 04:55 |
| gnarface | Guest9: there's a special optional set of packages for the stable release that are newer than the normal rules for stable would allow, they're called backports | 05:00 |
| gnarface | Guest9: the instructions for doing it on debian work if you substitute devuan repo and release names | 05:03 |
| gnarface | https://backports.debian.org/Instructions/ | 05:03 |
| gnarface | if that doesn't work though i got nothing | 05:05 |
| gnarface | so basically instead of the line they say to add in your sources.list on that page, you use this instead: deb http://deb.devuan.org/merged chimaera-backports main | 05:07 |
| gnarface | and in the apt-get/apt-cache commands you use "-t chimaera-backports" instead of bullseye-backports | 05:08 |
| gnarface | the kernels are all named starting with "linux-image-" so you can search for that | 05:09 |
| onefang | These backports are just versions of packages from later versions of Devuan. So you can get a more recent kernel, Chimaera-backports is on kernel 6 now. | 05:09 |
| gnarface | Guest9: assuming you're on amd64 hardware the one you'd want to get is this one: linux-image-6.0.0-5-amd64 | 05:11 |
| gnarface | ...i think | 05:11 |
| onefang | I use that to support my newish graphics card, plus it has a fix for that twenty year old "fix" that slows down recent AMD Zen CPUs that didn't exist twenty years ago. Twenty years ago it was a good thing for some CPUs. But now with more recent ones, that have lots of cores, like my 64 core Threadripper, it slows things down. | 05:12 |
| gnarface | oh no, that's not it | 05:12 |
| Guest9 | i have a librebooted thinkpad t400 | 05:12 |
| gnarface | i'm not sure what cpu is in that | 05:13 |
| gnarface | can you find out if it's 64-bit? | 05:13 |
| onefang | linux-image-6.0.0-0.deb11.2-amd64 for Chimaera. | 05:13 |
| gnarface | thanks onefang | 05:13 |
| gnarface | Guest9: this one^ | 05:13 |
| gnarface | onefang: t400 is a 64-bit cpu right? | 05:14 |
| onefang | No idea. | 05:14 |
| gnarface | ih | 05:14 |
| gnarface | Guest9: make sure it's 64-bit, it probably is unless it's ancient but the last thinkpad i saw was ancient | 05:15 |
| onefang | And the version will be 6.0.3-1~bpo11+1 (stable-backports). The bpo means backport. | 05:15 |
| gnarface | hmm, google says it's core 2 duo, they didn't make those in 32-bit though they sold a few with 32-bit windows and lied about it | 05:16 |
| gnarface | the amd64 one should work afaik | 05:16 |
| gnarface | Guest9: if you run "uname -a" it should be clear | 05:16 |
| onefang | In my case that says "6.0.0-0.deb11.2-amd64 #1 SMP PREEMPT_DYNAMIC Debian 6.0.3-1~bpo11+1 (2022-10-29) x86_64 GNU/Linux " | 05:17 |
| Guest9 | 5.10.0-9-amd64 #1 SMP Debian 5.10.70-1 (2021-09-30) x86_64 GNU/Linux | 05:17 |
| onefang | Amd64 it is then. B-) | 05:18 |
| onefang | Also if you are using synaptic, then the option to pick the backport (or any other version) is Package menu -> Force Version... -> then select the wanted version from the drop down in the new window. | 05:25 |
| Guest9 | So if I use Synaptic, I wouldn't have to use the terminal instructions in https://backports.debian.org/Instructions/ , correct? | 05:43 |
| onefang | Synaptic is a GUI wrapper around apt, so correct. | 05:44 |
| unclouded | Are there tools to generate /etc/initramfs-tools/conf.d/resume, /etc/kernel-img.conf and /var/lib/urandom/random-seed, or do I create them by hand? | 05:49 |
| Guest9 | I'm in Synaptic's "Kernel and modules" section and cannot find any "linux-image" greater than 5.10.18 | 05:52 |
| Guest9 | Is this a repository issue? | 05:52 |
| Guest9 | I looked through my repositories, and I only have "chimaera" selected. Is it safe/stable for me to select the "chimaera-updates", "chimaera-security", and "chimaera-backports" repositories? | 06:11 |
| onefang | You should have chimaera-updates, chimaera-security enabled for sure. chimaera-backports is mostly safe and stable. Consider that Devuan is based on Debian, and we both put great effort into being stable, selecting slightly unstable options is fine. | 06:14 |
| onefang | Ah and you wont see the kernel 6.0 until you have backports selected installed. | 06:15 |
| Guest9 | I'm sold. I've installed "linux-image-6.0.0-0.deb11.2-amd64" version "6.0.3-1~bpo11+1". | 06:41 |
| Guest9 | I still don't see my bluetooth mouse on my blueman-applet, but I probably have to restart my system since I updated the kernel, correct? | 06:42 |
| onefang | Correct, reboot if you changed kernels. | 06:45 |
| onefang | I tend to reboot if I change anything to do with the booting sequence. So basically anything that updates the initram file system. | 06:46 |
| Guest9 | In that case, I'll exit this chat to reboot. onefang , gnarface , thanks to both of you for the help. | 06:55 |
| Guest9 | Hopefully this works. | 06:55 |
| onefang | You are welcome, and good luck. | 06:55 |
| gnarface | unclouded: those are all packaged or generated files yes, though off the top of my head i don't know exactly by what, i know you don't create them by hand | 07:02 |
| gnarface | unclouded: er, maybe the resume file is manually created, not sure, that one i don't have here | 07:05 |
| gnarface | that one might only be needed if you have multiple swap partitions and it picks the wrong one | 07:06 |
| gnarface | hmm, i'm not sure if that thinkpad Guest9 had needs nvidia drivers from non-free or not, but if it does they need to come from backports too or they most likely won't work... someone needs to pass the word along | 07:29 |
| TwistedFate | Does anyone here use BIND as DNS on Devuan? | 09:46 |
| TwistedFate | How would one go about setting it up as a basic DNS server to use the root servers? | 09:46 |
| TwistedFate | I once had it setup but I forgot how | 09:46 |
| FatPhil | If you're not bound to bind there are much easier DNS servers to manage. I found MaraDNS trivial back in the days I wanted to run my own server. | 09:59 |
| bb|hcb | TwistedFate: The default setup does just that (I suppose you are going to use it as a caching resolver) | 10:01 |
| bb|hcb | The only part to configure is to allow recursion for the intended clients. | 10:02 |
| bb|hcb | For authoritative DNS bind has certain flaws, I'd recommend using nsd | 10:03 |
| TwistedFate | bb|hcb ah, how does recursion config goes? | 10:29 |
| ravehaver9000 | quit | 12:11 |
| ravehaver9000 | sorry | 12:11 |
| ravehaver9000 | kinda new to irc clients, was gonna say i reset dpms tho | 12:11 |
| ravehaver9000 | now to test if it werks | 12:11 |
| ravehaver9000 | i still get the modprobe error after booting but i dont think i should worry much about it anyways. decided not to fuck around with the persistence daemon either. | 12:15 |
| ravehaver9000 | anyways, im really enjoying using devuan these days. ive mostly used arch-based distros (including arch) before so moving to devuan was already kind of a change of pace from the beginning, but honestly i really like this distro the most out of all the ones i've used. | 12:17 |
| ravehaver9000 | i unironically feel like arch these days has lost all its "diy" reputation it had with archinstall and similar stuff (ironically, its diy apprach made it so popular it just works now), and i unironically feel like debian-based distros, most notably devuan in testing and unstable give that exact same do it yourself feeling arch was meant to give | 12:19 |
| ravehaver9000 | for one, yesterday was like the first time i ever had to downgrade my kernel. it wasnt really hard but it was still a pretty nice achievement for me | 12:19 |
| bb|hcb | TwistedFate: apt install bind9; it will do recursive lookups for all local ips by default; if you want to add more ranges, then add these to /etc/bind/named.conf.options: allow-recursion { localnets; localhost; 10.0.0.0/8; 172.16.0.0/12; 192.168.0.0/16; }; allow-query-cache { localnets; localhost; 10.0.0.0/8; 172.16.0.0/12; 192.168.0.0/16; }; ...and obviously change the perfixes | 12:56 |
| bb|hcb | Let me stress that it is REALLY a bad idea to have an open resolver answering to anyone. [OK, it is possible with complex rate limiting] | 12:58 |
| gnarface | damnit ravehaver9000 was back again someone please please tell him if he comes back i meant DKMS NOT DPMS | 15:46 |
| * gnarface signs | 15:46 | |
| * gnarface sighs even | 15:46 | |
| jjakob | steal-ctty segfault, something about can't open reopen-console | 16:59 |
| jjakob | in chimaera minimal iso | 16:59 |
| jjakob | https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=932149 yeah I have just 512mb | 17:03 |
| jjakob | the netinstall iso actually | 17:06 |
| gnarface | and no swap? | 17:06 |
| jjakob | it's booting from usb, how can it have swap | 17:08 |
| jjakob | it's the iso | 17:08 |
| gnarface | well, it can still get it, believe it or not | 17:10 |
| gnarface | but you're saying this happens before disk partitioning? | 17:10 |
| gnarface | or you're saying the usb is the system disk? | 17:11 |
| jjakob | yeah immediately as it starts to boot | 17:11 |
| gnarface | oh, bummer | 17:11 |
| gnarface | hmm | 17:11 |
| jjakob | no, just booting the installer | 17:11 |
| gnarface | and that's in text mode? | 17:11 |
| jjakob | I'm searching for a 1gb ram stick | 17:11 |
| gnarface | not graphical mode? | 17:11 |
| jjakob | expert mode | 17:12 |
| jjakob | isn't netinstall text mode always? | 17:12 |
| gnarface | uh, i didn't think so, i thought it also had a graphical mode these days but it's been so long since i've used it | 17:14 |
| gnarface | not graphical like the live image that boots to a full desktop, but a interface analogous to the text one but with X11 rendered buttons and mouse support etc | 17:15 |
| jjakob | should I try DEBIAN_FRONTEND=newt? or text? | 17:17 |
| jjakob | now that I found the 1gb stick, lol | 17:17 |
| gnarface | i'm not sure honestly, i'd try the 1GB stick | 17:18 |
| jjakob | 1gb stick works, frontend didn't | 17:25 |
| jjakob | it's text anyway | 17:25 |
| gnarface | noted | 17:25 |
| gnarface | this is a bad bug but i don't know any other tricks except to just do a debootstrap install instead | 17:26 |
| gnarface | sorry :( | 17:26 |
| xisop | hello. as of late, i've been having issues with i3 and possibly firefox. the issue is that the window manager stops responding to right mouse clicks. I was just wondering if anyone has reported anything similar? i've only noticed it within the last 24-48 hours | 18:49 |
| xisop | i've submitted a bug report to i3's github issues page and will be waiting on their team to respond | 18:49 |
| gnarface | nothing i've seen so far, i'd try a different WM for comparison, and if it's happening in both i'd start to suspect the mouse itself | 18:49 |
| xisop | the thing is that if i restart i3, it's fine | 18:50 |
| xisop | what's a recommended way to look into issues with the mouse? | 18:50 |
| mdt | i had that lately and it was indeed the mouse ;) | 18:50 |
| xisop | mdt: what wm did you use? | 18:51 |
| gnarface | xisop: interesting, can you isolate a series of specific steps that cause it to start misbehaving? it could be some program grabbing focus and then sticking on it | 18:51 |
| gnarface | xisop: usually errors related to x11 go into the xorg log | 18:51 |
| xisop | gnarface: i wish it was more specific, but all i have to do is use firefox for about an hour or two | 18:51 |
| xisop | gnarface: i will look at the xorg log | 18:52 |
| gnarface | firefox is one program i've previously noted that's very bad about stealing focus especially when changing between adjacent windows too fast | 18:52 |
| xisop | i upgraded to firefox 107.0.1 thinking it would fix it, but it didn't. i think it might also be related to using youtube | 18:53 |
| xisop | i will try to get more details | 18:54 |
| gnarface | another thing you could try is seeing if you can recreate it without firefox | 18:54 |
| xisop | youtube made me think of stealing focus because it usually has a fullscreen interface | 18:54 |
| xisop | gnarface: ok, i'll try that too | 18:54 |
| gnarface | yea, i have problems with focus stealing if i'm using two fullscreen windows on adjacent desktops (not using i3 but since it's tiling, the way it tiles may make tile borders act similarly) | 18:55 |
| xisop | fwiw, i was thinking it was related to this which is why i upgraded firefox https://www.mozilla.org/en-US/security/advisories/mfsa2022-48/#CVE-2022-45404 | 18:59 |
| mdt | xisop: same setup ;) | 19:00 |
| mdt | but is was the switch in the mouse, i replaced it and it was gone | 19:01 |
| xisop | mdt: the switch? | 19:02 |
| xisop | i'm going to build i3 from source and then go from there | 19:02 |
| mdt | behind the left mouse button, its a micro switch that sometimes did not fire | 19:02 |
| xisop | (latest version, i mean) | 19:02 |
| mdt | give a mouse-change a try ;) | 19:03 |
| xisop | mdt: but then how does that explain it working fine after logging out and back in again? | 19:04 |
| gnarface | maybe you're subconsciously pressing it harder? | 19:05 |
| gnarface | maybe it's cooling down or something? i had an optical sensor in an old mouse go out in a way where if it were unplugged for half an hour it'd work again for 5 minutes | 19:06 |
| xisop | let's say i am pushing harder. would that be fixed by restarting i3? | 19:07 |
| gnarface | i'm saying, it seems very unlikely but slightly possible | 19:08 |
| xisop | i see | 19:09 |
| gnarface | try a different WM to be sure | 19:09 |
| _ds_ | You could confirm whether it's the mouse by watching input events. | 19:09 |
| xisop | i'm not trying to be a jerk, i'm just trying to follow the line of questioning | 19:09 |
| gnarface | i've seen weird shit happen it's pretty rare though | 19:09 |
| gnarface | keep in mind when the driver re-inits there's gonna be a slight surge in power | 19:10 |
| gnarface | very slight, but possibly enough to clear some sort of hysteresis in a switch that's only barely damaged | 19:10 |
| gnarface | i think it's more likely you're running into an issue specifically with i3 and certain programs | 19:10 |
| xisop | k | 19:11 |
| gnarface | the only way to REALLY test that would be deductively, with a different WM | 19:11 |
| gnarface | but, you can always try out xev to see if you can catch the mouse signals there when firefox isn't seeing them | 19:11 |
| gnarface | that could at least isolate the issue to firefox | 19:11 |
| gnarface | you should be able to have multiple WMs installed concurrently and change between them | 19:12 |
| gnarface | if you're using "startx" it's easy but it's still possible if you're using a graphical login | 19:13 |
| _ds_ | hexdump -C $(sudo lsinput|grep -i mouse -B5|head -n1) | 19:13 |
| _ds_ | (note: lift the mouse to remove interference from motion events) | 19:13 |
| xisop | i will keep an eye on it. thank you immensely for the help gnarface , mdt, _ds_ | 19:15 |
| xisop | technically, if i'm using i3, the mouse is the enemy :P | 19:16 |
| _ds_ | Enemy? Hmm. i3 must be a cat equivalent, then. | 19:17 |
| xisop | you can do a lot without the mouse in i3 | 19:17 |
| xisop | it's the only reason i use it: i can avoid using the mouse to manage workspaces | 19:17 |
| gnarface | xisop: it tends ot have problems with opengl fullscreen contexts when they are tiled, and firefox has been using opengl rendering by default for a while... it occurs to me you might be able to also test by disabling that in firefox (hardware accelerated page rendering, or whatever firefox calls it) | 19:18 |
| xisop | i believe i disabled hw rendering, but let me double check | 19:18 |
| gnarface | if it stops happening entirely after that, you pretty much know | 19:18 |
| xisop | confirmed: hw accel is disabled | 19:18 |
| gnarface | hmm, so much for that theory then | 19:18 |
| gnarface | well, xfce is pretty well known around here, i like enlightenment though | 19:19 |
| gnarface | enlightenment does also have a tiling mode | 19:19 |
| gnarface | (though i suspect there's a possibility it's the inherent behavior of tiling that could be causing this) | 19:19 |
| xisop | enlightenment.. not sure i've used that, though i do beleive i used a terminal emulator for it and it was quite beautiful | 19:19 |
| xisop | i'd like to see what building 4.21.1 from source could do for me. | 19:20 |
| xisop | ninja is such a fast build system. that i3 compile was over in no time | 19:32 |
| unclouded | Thanks gnarface. I'm making a comparison between debootrap and an installation at https://pub.rui.nz/debootstrap_vs_installation.html and these are the last three differences. It's good to know that /etc/initramfs-tools/conf.d/resume is not essential | 19:58 |
| gnarface | unclouded: i think that if you run update-initramfs it will tell you which partition is currently designated as the resume partition and the variable you need to define to override it | 19:59 |
| gnarface | i believe you'll also see the same info whenever installing a new kernel | 20:00 |
| gnarface | my assumption is that the kernel-img.conf file comes from the kernel package pre/postinst scripts and the urandom offset is generated by the kernel itself or maybe udev, but those are just guesses | 20:01 |
| unclouded | Hmm. I already installed a kernel in the bootstrapped system and it didn't generate these files. I'll keep looking | 20:01 |
| gnarface | the system i'm looking at also has had a number of custom kernel packages created with the kernel-package tool so that might be where i got them | 20:02 |
| unclouded | Ah, so you have those files in a tree created by debootstrap? Interesting | 20:02 |
| gnarface | ah, no i have those files in a tree created by multiple updates of debian unstable | 20:03 |
| gnarface | i'm just now checking another system which was created by debootstrap but does not have a packaged kernel is also missing the /etc/kernel-img.conf file, but does have the /var/lib/urandom/random-seed file | 20:04 |
| gnarface | so it looks like kernel-img.conf is also optional | 20:04 |
| unclouded | OK. I am wondering if they're created by the installer rather than a dpkg-reconfigure | 20:04 |
| gnarface | possible based only on the examples i have here, but i note that kernel-img.conf man pages show up in the linux-base package and kernel-img.conf examples show up both in kernel-package as well as a custom packaged kernel i made with it | 20:06 |
| gnarface | so that's why i thought kernel-img.conf might have something to do with packaged kernels | 20:06 |
| gnarface | the random-seed file being present on a install i made with debootstrap and no packaged kernel suggests to me it's generated on the fly after boot | 20:07 |
| gnarface | i don't really know for sure at this point | 20:07 |
| unclouded | Nice. I plan to boot this system, so will find out if random-seed is generated at boot | 20:08 |
| gnarface | the system without the packaged kernel is also not using an initrd.img, and on that one the entire /etc/initramfs-tools/ directory is missing | 20:09 |
| gnarface | so you don't need any of that if you're not using an initrd.img apparently | 20:09 |
| unclouded | I guess that makes sense. I plan to use a regular packaged kernel, which is already installed in the bootstrapped tree | 20:10 |
| unclouded | it might not be critical. The man page says do_symlinks has a default value and the packaged kernel has created the /initrd.img symlink and others | 20:13 |
| xisop | looks like i3 won't support bug reports for older versions. i'll update here if i run into any issues with v4.21.1 | 20:45 |
| Udo | 21:06 | |
| smpl | xisop; I've experienced broken left-clicking in the most recent update of FF on Mint. | 22:27 |
| deuxexmachina | http://dev1galaxy.org/viewtopic.php?id=4191 | 22:33 |
| deuxexmachina | I realize this is 4yo... But do many devuan users use unattended upgrades? | 22:34 |
| deuxexmachina | Wondering if I should just use debian if I want unattended upgrading reliably long term on production | 22:34 |
| deuxexmachina | Would prefer non-sysd ofc | 22:34 |
| deuxexmachina | nevermind, not 4yo, only about 1yo | 22:35 |
| deuxexmachina | also curious how many issues devuan has not experienced since the fork that debian has... I'm guessing there's a thread showing off its simplicity with examples like that somewhere? | 22:36 |
| * gnarface shrugs | 22:39 | |
| gnarface | i don't think we do a lot of gloating around here | 22:39 |
| gnarface | i could be wrong | 22:39 |
| deuxexmachina | :( | 22:40 |
| deuxexmachina | *seeks gloat* | 22:40 |
| gnarface | it would be easy to find that info though | 22:40 |
| deuxexmachina | could be phrased as incidents | 22:40 |
| deuxexmachina | as far as i can tell so far devuan should benefit from all of debian's security work | 22:40 |
| deuxexmachina | but i could be wrong | 22:40 |
| gnarface | it seems to | 22:44 |
| gnarface | and it's immune to any systemd-specific problems | 22:44 |
| gnarface | deuxexmachina: try asking in #devuan-offtopic, i think golinux might have been keeping score somewhere. this channel is more about support issues (as you can see there aren't many but we like to keep it that way) | 23:02 |
| deuxexmachina | why golinux - how's that related to devuan? | 23:06 |
| deuxexmachina | gnarface: sounds good... So long as a bunch of other devuan users are depending on unattended upgrades | 23:07 |
| deuxexmachina | which I guess could be partially debian-specific, based on that thread I pasted | 23:07 |
| gnarface | i meant about the "issues" score vs systemd | 23:08 |
| gnarface | i don't actually know how many users are depending on unattended upgrades, i know there's at least 2 who asked about it but it's actually not something we advise using just because of the inherent risks involved | 23:09 |
| deuxexmachina | But debian doesn't advice against it, do they? | 23:18 |
| deuxexmachina | some people have relied on that across many production servers for years without issue... | 23:19 |
| gnarface | really? it's a surprise to me if debian doesn't advise against it, it's quite reckless | 23:22 |
| gnarface | nothing devuan specific about that, you should always know what is being installed on your system, over-automation leads to problems | 23:22 |
| gnarface | anyway it'll work if you wanna use it, but i personally would advise you always check your updates | 23:23 |
| gnarface | there's a simpsons episode about a plastic water-bird that seems relevant here | 23:23 |
| xisop | i've been locked out of servers because of failed upgrades that happened without my approval | 23:23 |
| xisop | it's usually debian or ubuntu servers that just kinda do whatever they want | 23:24 |
| xisop | then you have to launch recovery console in digital ocean or physically get on the box | 23:24 |
| xisop | that's just one man's use case though, but i've found it's easier for me to be in control of upgrades | 23:25 |
| gnarface | just seems like an unnecessary risk to me too | 23:25 |
| unclouded | same | 23:25 |
| gnarface | like i said though, it's been asked about at least twice before, and i know those two people used it successfully | 23:26 |
| unclouded | I can cause enough trouble for myself on remote servers even without unattended upgrades ;) | 23:28 |
| deuxexmachina | gnarface: but as i understand it unattended upgrades is only for mission critical fixes, stuff that should break - like mandatory security patches | 23:28 |
| xisop | i think the problem is when you get those upgrades that force you to make a decision about the config file powering that package | 23:29 |
| deuxexmachina | well if i can figure out how to do very clean live migrations of a VPSs networking and its docker containers - then I'll just deploy fresh images each time and migrate... | 23:29 |
| deuxexmachina | people say not to use rolling releases on servers too and I've had way fewer problems on those than the popular choices | 23:30 |
| xisop | btw, TCL might be of use to you if you want to control those types of things | 23:30 |
| deuxexmachina | everything on the servers in question is run in docker too, except some backup init scripts | 23:30 |
| xisop | tcl is made for this sorta thing... granted you are aware of the output these tools give you | 23:31 |
| deuxexmachina | the security work, especially reproducible builds, is a main reason I'm considering debian for these servers | 23:31 |
| xisop | debians solid. i just dont know how to audit systemd | 23:31 |
| deuxexmachina | who does :P | 23:32 |
| xisop | or if it's been attempted, or what the status is | 23:32 |
| xisop | well, i guess that's why we're all here eh :) | 23:35 |
| deuxexmachina | a devuan forum post says use sysvinit if you're priority is keeping the server running | 23:35 |
| deuxexmachina | but id choose runit due to familiarity | 23:35 |
| deuxexmachina | I'd want to* | 23:35 |
| deuxexmachina | what's the sentiment here about the reliability of the various init choices? | 23:36 |
| gnarface | runit is fine, sysvinit is just the one that more people around here can actually help you with | 23:36 |
| xisop | i've heard good things about openrc, though i don't know much about it | 23:36 |
| deuxexmachina | I'd be happy with openrc as well, would just prefer not to use sysvinit since i havent used it in forever | 23:36 |
| deuxexmachina | mostly use runit these days because my workstation is on void | 23:37 |
| gnarface | sysvinit is also the one with the most populated set of startup files, though debian sometimes deletes them without cause | 23:37 |
| xisop | well, regarding init choices... for me, it's a security issue. | 23:37 |
| xisop | i don't look at sysvinit and wonder what the motivation is. i look at systemd and can't help to think that it wants to *become* linux | 23:37 |
| xisop | when you have a firm grasp on the security issues that matter to you, you don't want to have to just assume a program does what it says it does | 23:38 |
| xisop | the same sort of sentiment applies to people with windows gaming rigs. they hate windows but can't run their high end games on linux | 23:39 |
| xisop | and how do you audit windows if it's closed source. you just have to assume it works | 23:40 |
| xisop | i'm forced to assume the bad choices (in my opinion) of systemd's issues (like scope creep) are leading to something that's not terrible | 23:41 |
| xisop | less cognitive load, less to think about when you have one thing that does one thing | 23:41 |
| deuxexmachina | xisop: ya same here | 23:42 |
| deuxexmachina | and same with wayland | 23:42 |
| gnarface | i just don't like shit changing out from underneath me and causing more workload | 23:42 |
| deuxexmachina | tho i dont care about gui because i wont use debian-based stuff for desktop and workstation anyway | 23:42 |
| deuxexmachina | gnarface: indeed | 23:43 |
| deuxexmachina | Would like to do security-only unattanded upgrades tho, since these boxen only run docker, ssh, and some backups scripts. Then can deploy new images for major upgrades. | 23:44 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!