talismanick | Do the repos not support https? >_> | 00:04 |
---|---|---|
rwp | Most of the mirrors are impossible to mirror using https because the names will never match. | 00:04 |
rwp | The security of the system was designed without requiring https. | 00:04 |
talismanick | uuuuuh | 00:05 |
talismanick | As in, checksums? | 00:05 |
rwp | Yes. SHA hashes that are GPG signed by the keyring. | 00:05 |
talismanick | "the names will never match" | 00:06 |
talismanick | What does that mean? | 00:06 |
rwp | deb.devuan.org is a round-robin list of other file server systems. They all have names that are different. | 00:06 |
onefang | But you can access them directly via their own name using HTTPS. | 00:07 |
rwp | Check out the mirror list at https://www.devuan.org/get-devuan | 00:07 |
onefang | Beat me to it. lol | 00:07 |
talismanick | Well, if it's good enough for Debian too... | 00:07 |
* onefang goes back to waking up. | 00:08 | |
rwp | Good morning onefang! :-) | 00:08 |
rwp | talismanick, Are you installing a new system or upgrading an existing system? | 00:09 |
onefang | It's summer and I just turned the heater on. "Good" is relative. lol | 00:09 |
rwp | onefang, You need to be in offtopic so I can comment that it was -15C here this morning. | 00:10 |
rwp | talismanick, If you are installing a fresh system then you should use an https mirror to download the install image. | 00:10 |
rwp | If you are upgrading then using an http://deb.devuan.org round-robin mirror is best. It's secure due to the keyring. | 00:11 |
rwp | Using https for deb repositories has several subtle problems. Like the system clock must be set. That snags Raspberry Pi users quite often. | 00:11 |
rwp | When we talk about "repos" that name is most often for the package database at deb.devuan.org and not the install images from the other mirror sites. | 00:14 |
onefang | We got package mirrors, which is what we call repos, and listed at https://pkgmaster.devuan.org/mirror_list.txt. This is used for apt sources.list. Then we have the ISO / filo mirrors, for the install images, listod at https://www.devuan.org/get-devuan. | 00:20 |
onefang | deb.devuan.org is a round robin DNS for the former. | 00:20 |
talismanick | rwp: migrating a Debian system | 00:24 |
golinux | talismanick: Have you seen the migration instructions here https://www.devuan.org/os/install ? | 00:27 |
rwp | talismanick, Welcome to the community! | 00:27 |
talismanick | thanks | 00:35 |
talismanick | golinux: yeah | 00:35 |
talismanick | speaking of which, what do I do differently from the Bullseye->Chimaera update guide for runit? | 00:36 |
talismanick | aside from changing sysvinit-core to runit-init (right?) | 00:39 |
rwp | I have *installed* selecting runit but I haven't *converted* a system to runit. But I think you just need to apt-get install the desired runit* packages and then reboot. But don't trust me on that since I haven't done it. | 00:39 |
talismanick | rwp: `apt -f install` is what changes out the init regardless, right? | 00:41 |
rwp | Not that specific command. That's to resolve and repair a broken package dependency. | 00:42 |
Jesse39 | Hello, I'm having some trouble getting my wifi working | 03:08 |
Jesse39 | I installled via ethernet and during install it said that I would need firmware "regulatory.db" and "iwlwifi-6000-4.ucode" | 03:10 |
Jesse39 | And I'm not sure how to find or add them now | 03:10 |
debdog | according to https://packages.debian.org "iwlwifi-6000-4.ucode" is provided by the firmware-iwlwifi package | 03:12 |
Jesse39 | so I only need to install that package? | 03:13 |
debdog | I think so, yes | 03:13 |
debdog | but I am not a wifi expert | 03:14 |
debdog | at least it is a step in probably the right direction | 03:16 |
Jesse39 | thanks. I'll look in to it a bit more | 03:16 |
Jesse39 | I've installed it on this computer before but can't remember what I had to do for wifi | 03:17 |
debdog | well, if installing said package doesn't improve your situation, you just can purge it afterwards | 03:19 |
Jesse39 | I tink I need to add some repo | 03:23 |
Xenguy | Jesse39, paste your /etc/apt/sources.list to paste.debian.net if need be | 03:25 |
Jesse39 | was able to install it by adding non-free repo to sources.list | 03:28 |
Xenguy | Jesse32, Great, for more info: https://www.devuan.org/os/packages | 03:30 |
Jesse32 | it seems that I am still unable to use wifi | 03:31 |
debdog | but did it load the firmware? does "lspci -k" list a kernel driver for your wifi device? | 03:33 |
debdog | "Kernel drive in use" or simililar | 03:33 |
debdog | hehe *similar | 03:33 |
Jesse32 | I'm not seeing the device there | 03:36 |
Jesse32 | maybe I'm just overlooking it since I don't understand a lot of this | 03:37 |
debdog | could you paste the output, please? preferably on https://paste.debian.net/ | 03:38 |
Jesse32 | Network Controller? | 03:38 |
debdog | sounds about right. if uncertain the entire output | 03:39 |
Jesse32 | https://paste.debian.net/1262451/ | 03:39 |
Jesse32 | I gave the whole output but I think this is it | 03:39 |
Jesse32 | it sayss "Kernel modules: iwlwifi" but no driver in us3e | 03:40 |
debdog | 02:00.0 Network controller: Intel Corporation Centrino Advanced-N 6200 (rev 35) | 03:40 |
debdog | Subsystem: Intel Corporation Centrino Advanced-N 6200 2x2 AGN | 03:40 |
debdog | Kernel modules: iwlwifi | 03:40 |
Jesse32 | Ye | 03:40 |
debdog | this is after a reboot, right? | 03:41 |
Jesse32 | oof no I should probably try that | 03:41 |
Jesse32 | I'll be right back | 03:41 |
Jesse17 | yeah it seems to be working now | 03:44 |
Jesse17 | Thanks for the help | 03:44 |
debdog | no worries. but celebrate when you have an actual connection | 03:44 |
Jesse17 | I am connected | 03:45 |
debdog | well then... CELEBRATION | 03:45 |
debdog | hehe | 03:45 |
Jesse17 | Thanks again, I don't know why I wasn't able to find this solution | 03:46 |
Jesse17 | This could have been a lot more frustrating lol | 03:46 |
debdog | your lspci output reminds me of an elderly laptop I recently installed win7 on... friggin PITA | 03:47 |
talismanick | https://0x0.st/o0In.txt | 05:05 |
talismanick | Is it supposed to do this? | 05:05 |
talismanick | same trouble with sysvinit-core instead of runit-init | 05:05 |
gnarface | use paste.debian.net and i'll actually look | 05:12 |
talismanick | gnarface: https://paste.debian.net/1262460/ | 05:32 |
talismanick | (0x0 is what I have set up for pasting killed regions in Emacs) | 05:33 |
gnarface | talismanick: uh... yes that's expected behavior but you shouldn't have had systemd installed in the first place, are you attempting a migration from debian? if not, you have accidentally mixed repos at some point | 05:43 |
talismanick | gnarface: yes, migrating | 05:43 |
gnarface | there is detailed instructions on the safest path around here somewhere, but i'd recommend you have a live image you can boot from in case something goes wrong | 05:45 |
rrq | this one still works well: https://git.devuan.org/rrq/debian-to-devuan | 05:45 |
gnarface | talismanick: was this install bullseye or buster? | 05:46 |
talismanick | gnarface: I have everything backed up anyways, so I can roll back to the previous state and try again | 05:46 |
gnarface | talismanick: so the risky part is it's gonna have to remove systemd and all the dependencies, including the running kernel which will then have to be reinstalled along with all the devuan startup stuff before you reboot | 05:47 |
gnarface | the script might be a better choice if you're not sure of yourself | 05:48 |
gnarface | we've seen various pitfalls, especially in situations where your grub setup is weird | 05:49 |
gnarface | or if you get unlucky and your rootfs gets flagged for fsck on that critical reboot | 05:49 |
gnarface | most people just forget to put the kernel back in | 05:49 |
talismanick | rrq: Aside from using Alpine a bit in the past, I haven't used sysvinit/OpenRC much. What would the runit equivalents of lines 81-87 be? | 05:50 |
talismanick | gnarface: sounds like I should spin up a Devuan VM and move things over instead? | 05:51 |
gnarface | talismanick: well, no i'd still recommend rrq's script in this case, and just switch to runit after the migration is done | 05:51 |
gnarface | ... since you have backups | 05:52 |
gnarface | and this will be a good learning experience :) | 05:52 |
gnarface | if it goes well it should be a lot faster | 05:52 |
gnarface | just remember the thing about the kernel | 05:53 |
rrq | I don't know. Perhaps the best is to install sysvinit-core first and then replace to openrc (I don;t know the steps for that, but it's almost certainly easier that way) | 05:53 |
rrq | I think 81 -87 would stay as they are | 05:53 |
gnarface | talismanick: it might actually uninstall a lot of stuff, but you can just reinstall all of it | 05:54 |
gnarface | talismanick: just make sure you got the kernel, grub, and eudev installed again before you reboot | 05:55 |
gnarface | if it uninstalls your GUI you can reinstall that before or after reboot | 05:55 |
talismanick | There's no gui | 05:56 |
talismanick | it's a server | 05:56 |
gnarface | should be much easier then | 05:56 |
talismanick | yea | 05:56 |
gnarface | it's not dual-booting windows or anything right? | 05:56 |
talismanick | nope | 05:56 |
gnarface | and grub is on the first disk already? | 05:56 |
talismanick | yes | 05:57 |
gnarface | you should be fine if you don't panic and reboot before it's bootable again | 05:57 |
talismanick | just a plain Debian install, running several webservers | 05:57 |
gnarface | if for some reason you reboot before it's bootable again, it's usually pretty simple to just boot with a live image or other rescue image, chroot in, then install what's missing (usually consists of a kernel, eudev, and running update-grub again) | 05:58 |
rrq | mmm line 74 might need editing; there is a newer keyring | 05:58 |
rrq | you'd use _2022.11.15_all.deb nowadays | 06:00 |
rrq | ... and line 75 the same | 06:01 |
talismanick | Came back to see if it finished | 08:14 |
talismanick | it just dropped to a grub prompt | 08:14 |
talismanick | I think I'll just boot off a live ISO and pull everything else out of the backup | 08:14 |
gnarface | talismanick: no, don't do that, boot off the live image, chroot in, and run grub-install again | 08:20 |
gnarface | talismanick: and while you're in there make sure you have a kernel installed, like i said | 08:20 |
talismanick | gnarface: ...but why? | 08:20 |
talismanick | that only sounds like it's inviting more problems... | 08:20 |
gnarface | talismanick: we've been through this 1000 times or so by now | 08:21 |
gnarface | just trying to save you time | 08:21 |
talismanick | What usually happens if I do the opposite? | 08:21 |
gnarface | what do you mean the opposite? | 08:21 |
gnarface | what you did was the opposite; ignore me and just let it reboot without a kernel or grub configuration in plce | 08:22 |
gnarface | there's only so many things it can be missing here, it's a list of 2 | 08:22 |
talismanick | fresh install from an ISO, boot the backup in a VM, and move over the files I need | 08:22 |
gnarface | you just waste another 1-4 hours when you could be up and running in 5 minutes | 08:22 |
talismanick | Oh, I see what you thought I meant | 08:23 |
talismanick | I meant downloading an ISO, installing to disk, and adding back what state I need from the backup | 08:23 |
gnarface | yea that's still going to take way more time than just finishing what you started | 08:23 |
gnarface | do you need instructions for how to chroot properly? i can give them to you step by step | 08:24 |
talismanick | I know how to chroot into another system from a live image, but I don't think there's an option to boot a live ISO and access the prior system | 08:25 |
gnarface | what???? | 08:25 |
gnarface | false | 08:25 |
gnarface | you can access the harddrive from a live iso easily | 08:25 |
gnarface | you could do this without the live iso even if you know what to type into the grub prompt i just don't remember the exact commands | 08:26 |
gnarface | once you're chrooted in, with /sys, /proc, /dev, and /dev/pts all bind-mounted inside it, i guarantee that installing a kernel and re-running grub-install or update-grub will make it bootable | 08:27 |
gnarface | no need to restore from backup, you're like 99.9% of the way there and if you'd paid attention this would already have happened before you rebooted | 08:27 |
gnarface | and you're easily the 10th person i've had this argument with | 08:28 |
talismanick | alright, I'll listen | 08:28 |
gnarface | ok | 08:29 |
gnarface | mount /dev/sda1 /mnt | 08:29 |
gnarface | mount -o bind /proc /mnt/proc | 08:29 |
gnarface | mount -o bind /sys /mnt/sys | 08:29 |
gnarface | mount -o bind /dev /mnt/dev | 08:29 |
gnarface | mount -o bind /dev/pts /mnt/dev/pts | 08:29 |
gnarface | chroot /mnt /bin/bash -i | 08:29 |
gnarface | apt-get update && apt-get install linux-image-amd64 | 08:30 |
gnarface | update-grub | 08:30 |
gnarface | then just reboot | 08:30 |
talismanick | yeah, yeah, ye olde `for i in dev proc sys run; do mount --rbind /$i /mnt/$i; mount --make-rslave /mnt/$i; done` | 08:30 |
talismanick | or variations thereof | 08:31 |
gnarface | uh, i'm not sure whether that'll work or not, but what's important is that grub-install or update-grub sees the hardware right, or it'll misfire | 08:31 |
talismanick | can't say I can recall messing around with pts, though | 08:31 |
gnarface | pts may only be necessary if you're also running a tf2 server in there honestly, i just want to be sure you have everything for grub to be able to properly orient itself | 08:32 |
talismanick | What's special about TF2? | 08:32 |
talismanick | My curiosity is piqued | 08:32 |
gnarface | it's crappily programmed, i think, or it has a valve backdoor, fuck if i know. if you do this wrong it'll try to install grub to the livecd which will be useless | 08:32 |
talismanick | that's hilarious | 08:33 |
talismanick | "return to sender" | 08:33 |
gnarface | the bind mounts are so the kernel on the livecd, while in the chroot, knows where the real hardware is | 08:33 |
gnarface | it wouldn't be necessary if you'd done this step before rebooting | 08:34 |
gnarface | while you're in there make sure you have eudev installed too or you will get past the grub prompt but probably not much further | 08:34 |
talismanick | I just followed the instructions on the website verbatim | 08:34 |
gnarface | i didn't look at them so i can't tell you why they didn't work, this is the way i recovered from the mess the last dozen times it came up that's all | 08:35 |
gnarface | the bios found grub, grub didn't find a kernel | 08:38 |
gnarface | that means the kernel is missing or grub is out of place | 08:38 |
gnarface | the fix of reinstalling them both works either way | 08:38 |
talismanick | fair's fair | 08:39 |
talismanick | that makes sense to me | 08:39 |
gnarface | your main harddrive is called /dev/sda right? | 08:39 |
gnarface | this isn't something weird like an ARM machine or a VM right? | 08:39 |
talismanick | ...it is a vm | 08:40 |
talismanick | a vps | 08:40 |
* gnarface facepalms | 08:40 | |
talismanick | I thought I said this was a server | 08:40 |
talismanick | I was wondering why you were so adamant about this method | 08:40 |
gnarface | the instructions you were given most likely assumed it was a bare metal server | 08:40 |
gnarface | calling it a server also implies bare metal | 08:40 |
talismanick | bold of you to assume I have money | 08:40 |
gnarface | i don't have any money, i'm still using pentium3's here | 08:41 |
gnarface | wait if it's a vps how do you boot a live image on it? | 08:41 |
talismanick | Hetzner lets you | 08:41 |
talismanick | via the rescue mode reboot | 08:41 |
gnarface | this might still work, but we need to make sure grub is installed in the right place | 08:42 |
talismanick | Yeah, I do think this'd work - I don't see why it shouldn't | 08:42 |
talismanick | but, I hope my confusion about /why/ is clearer now | 08:42 |
gnarface | yea, being a VM it just introduces a lot of other unknowns, you should specify that more clearly in the future | 08:43 |
talismanick | mea culpa | 08:43 |
gnarface | bare metal, x86 hardware boots very uniformly, relatively | 08:43 |
gnarface | if you start getting into virtual setups, there's tons of weird different ways the hypervisor can work | 08:43 |
talismanick | true | 08:43 |
talismanick | I think I'll just install all the way as usual from the ISO | 08:44 |
gnarface | i'd still like to know if this will work though | 08:44 |
gnarface | you might learn something too | 08:44 |
gnarface | should only take a couple minutes to try | 08:44 |
talismanick | Well, if I can ever get it to recognize that I'm telling it to pull the ISO from an address... | 08:46 |
talismanick | Been a long time since I fell back on this option | 08:46 |
gnarface | you said they had a rescue image already preloaded? | 08:46 |
talismanick | To boot a custom ISO, you have to edit a conf file | 08:47 |
gnarface | see, i assumed you were sitting by the machine itself | 08:47 |
talismanick | And, I don't know how to get back to the default menu from this conf file | 08:47 |
talismanick | so it looks like I'm trapped in limbo until I can get it to go through without giving up and nuking the system | 08:48 |
gnarface | you literally can't figure out how to exit the conf file editor or something? | 08:48 |
talismanick | It drops you back into it if it doesn't recognize what it's suppose to do lol | 08:49 |
gnarface | can you show a screenshot? | 08:49 |
rrq | you know that the installer itself is like a live system? shift to a shell with C-A-F2 and go from there | 08:50 |
rrq | .. ah probably irrelevant here.. (where's my stone?) | 08:51 |
talismanick | https://0x0.st/o0Uu.png | 08:52 |
talismanick | ah, found how to get out | 08:53 |
talismanick | (surprisingly, not the "Cancel" button) | 08:54 |
gnarface | alright, good because i'm still not clicking on that domain | 08:54 |
talismanick | dude, it's just a pastebin | 08:54 |
talismanick | lol | 08:54 |
talismanick | (I like it because the default way to upload is via curl, so it's easy to script and attach to a button) | 08:54 |
gnarface | funny enough whenever i float my own links nobody clicks on them either | 08:55 |
talismanick | https://0x0.st/ | 08:55 |
gnarface | so anyway, boot anything that can mount the drive | 08:56 |
gnarface | or virtual drive or whatever it is | 08:56 |
gnarface | if you got it as far as a grub prompt that still suggests this will work | 08:56 |
gnarface | it has grub after all | 08:56 |
talismanick | lsblk just says the biggest partition is sda1, at 38G, so that should be it | 08:56 |
gnarface | ok, when you run grub-install make sure to install to sda not sda1 | 08:57 |
talismanick | ofc | 08:57 |
gnarface | but make sure a kernel is installed first | 08:57 |
talismanick | I've shot myself in the foot like that before :) | 08:57 |
gnarface | or you can run grub-install, then install the kernel, then just run update-grub | 08:58 |
gnarface | but i think grub-install will also imply update-grub | 08:58 |
gnarface | shouldn't hurt to run update-grub once extra as a last step just to be sure | 08:58 |
gnarface | the chroot and bind mount steps are still important, i only recommended the live image because i didn't realize you already had some sort of virtual manager with preloaded images | 09:00 |
talismanick | Well, that's a problem | 09:01 |
gnarface | ? | 09:01 |
talismanick | no networking in chroot | 09:01 |
talismanick | so, no apt-get | 09:01 |
gnarface | there probably is networking, it probably just didn't get as far as creating your /etc/resolv.conf | 09:02 |
gnarface | try to ping something by IP | 09:02 |
gnarface | or run /sbin/ifconfig -a | 09:02 |
talismanick | yeah, eth0 up and running | 09:03 |
gnarface | just create a resolv.conf manually | 09:03 |
gnarface | were you using NetworkManager before or something like that? | 09:03 |
talismanick | no | 09:03 |
talismanick | not that I | 09:03 |
talismanick | 'm aware of | 09:03 |
gnarface | is resolv.conf missing? | 09:03 |
talismanick | ls say it's there, but cat says no such file | 09:05 |
gnarface | ls -l | 09:05 |
gnarface | it's probably an orphaned symlink | 09:05 |
talismanick | ls -l says it symlinks to /run/resolvconf/resolv.conf | 09:05 |
talismanick | yeah | 09:05 |
gnarface | delete it and just create a static one | 09:05 |
gnarface | doesn't have to be permanent, just has to be there long enough to finish this | 09:05 |
gnarface | you probably want to UNINSTALL the resolvconf package too, since that'll point it elsewhere looking for dns entries | 09:06 |
gnarface | what does this command return? dpkg -l |grep linux-image | 09:07 |
talismanick | rc linux-image-5.10.0-17-amd64 5.10.136-1 amd64 Linux 5.10 for 64-bit PCs (signed) | 09:08 |
talismanick | rc linux-image-amd64 5.10.136-1 amd64 Linux for 64-bit PCs (meta-package) | 09:08 |
gnarface | traces of two uninstalled kernels, that's the smoking gun | 09:08 |
gnarface | somehow this went off the rails before the reboot | 09:08 |
gnarface | well, linux-image-amd64 is just a meta-package that points to the latest stable kernel | 09:09 |
gnarface | but the "rc" means they've been uninstalled | 09:09 |
gnarface | that should be "ii" | 09:09 |
talismanick | As much as I like a good detective mystery, I'm feeling too tired for another episode | 09:10 |
talismanick | I think I'm going to bed for now | 09:10 |
talismanick | thanks for the help | 09:10 |
rrq | talismanick: I wonder if you where running my script, or di you DIY ? | 09:10 |
talismanick | rrq: I DIY'd from the migration instructions on the website, then tried your script when those didn't work | 09:11 |
talismanick | ergo nasal demons | 09:11 |
rrq | ok; maybe someone could try tohe opposite order sometime | 09:11 |
rrq | (ideally running my script should avoid the need for DIY :)) | 09:12 |
talismanick | I read it over, and didn't see anything which tipped me off that it'd conflict | 09:13 |
talismanick | (at most, it repeated already-idempotent commands or added another) | 09:13 |
gnarface | talismanick: it doesn't matter why it didn't work, what matters is that if you put the resolv.conf back and reinstall them it starts working | 09:13 |
rrq | the key issue is to only repalce init before going further | 09:13 |
talismanick | gnarface: Where can I find the default resolv.conf? | 09:14 |
talismanick | to copy-paste | 09:14 |
gnarface | just use anything it's one line | 09:14 |
gnarface | don't you know the IPs of your DNS servers? | 09:14 |
talismanick | like, "foo bar baz"? | 09:14 |
talismanick | um, no | 09:14 |
gnarface | no, not literally anything but | 09:14 |
gnarface | hmm | 09:14 |
gnarface | check in /etc/network/interfaces | 09:14 |
gnarface | do you have lines for dns-nameservers and dns-search in there? | 09:15 |
talismanick | mmm, nyet | 09:15 |
gnarface | you can probably just use google's public dns temporarily until you can find the real ones | 09:15 |
talismanick | 8.8.8.8? | 09:16 |
gnarface | yea | 09:16 |
gnarface | nameserver 8.8.8.8 | 09:16 |
gnarface | that one line should be enough | 09:16 |
talismanick | oh | 09:16 |
talismanick | ah, works now | 09:16 |
gnarface | most likely your ISP will have provided their own nameservers for you, you should locate them later when you have a chance | 09:17 |
gnarface | some systemd component must have absorbed this, is my guess | 09:18 |
gnarface | so naturally they buried the evidence | 09:19 |
talismanick | wait fuck why did i reboot without exiting and umounting | 09:20 |
talismanick | oh, it didn't even do anything | 09:20 |
gnarface | for sysvinit you want to make sure you have: sysv-rc, sysvinit, sysvinit-core and sysvinit-utils | 09:20 |
gnarface | also make sure you have eudev, and linux-image-amd64 | 09:21 |
gnarface | (we're assuming the system is 64-bit, is that wrong?) | 09:21 |
talismanick | oh, nice, it works | 09:21 |
talismanick | And, I fetched runit-init instead of sysvinit-core | 09:22 |
gnarface | well i just wanted to make sure you had stuff i knew would work but as long as it works that's what matters | 09:22 |
talismanick | ty | 09:22 |
gnarface | i thought it would be safer to fix this mess first then you can migrate to runit after you've got it booting | 09:22 |
talismanick | The two are orthogonal, no? | 09:23 |
gnarface | well, it's not a difficult migration and i'm not sure if the runit install can fall back on sysvinit scripts or not, the openrc one does | 09:23 |
talismanick | openrc is a skin over sysvinit iirc | 09:24 |
talismanick | runit is its own thing (it even works on BSDs, etc) | 09:24 |
gnarface | openrc is in debian but not in other distros... anywayit seemed safer to have sysvinit in place and working as a known quantity first | 09:24 |
talismanick | part of the daemontools family, like s6 if you've used that | 09:25 |
gnarface | anyway is it booting now? | 09:25 |
talismanick | yeah | 09:25 |
gnarface | great | 09:25 |
talismanick | I'm using runit on Void, actually | 09:25 |
talismanick | (the default) | 09:25 |
gnarface | i want to just point out for the record we spent more time debating the solution than actually doing it | 09:25 |
talismanick | fair | 09:25 |
gnarface | anyway, i'm stepping away for now, have fun with your devuan | 09:25 |
talismanick | thanks | 09:26 |
uvos | hi, devuan chimaera includes the iio-sensor-proxy package imported from debian | 11:29 |
uvos | however this package contains only a systemd unit file, makeing this deamon package pretty useless | 11:30 |
uvos | for previous beowulf we had a fork with a init script: https://github.com/maemo-leste-upstream-forks/iio-sensor-proxy/blob/maemo/beowulf/debian/iio-sensor-proxy.init | 11:30 |
uvos | i would be neat if this could be included in devuan | 11:30 |
fsmithred | uvos, if debian dropped the init script, then the script needs to be added to the orphan-sysvinit-scripts package. Both are debian packages, and I'm not sure if the bug report should go to iio-sensor-proxy or orphan-sysvinit-scripts. It doesn't look like we forked that package - it's not in the main repo. Someone must have done it just for maemo-leste. | 12:35 |
uvos | yes i did | 12:36 |
uvos | (forked it for leste) | 12:36 |
fsmithred | lol | 12:36 |
fsmithred | we try to avoid forking because that requires maintenance. | 12:36 |
fsmithred | do you know if the old init script still works? | 12:37 |
uvos | ok so in beowulf it was not in your repos | 12:37 |
uvos | but now in chimeara it is https://pkginfo.devuan.org/cgi-bin/package-query.html?c=package&q=iio-sensor-proxy=3.0-2 | 12:37 |
uvos | imported from debian i gues | 12:37 |
fsmithred | yeah, it's also in beowulf-backports | 12:37 |
fsmithred | all are from debian | 12:37 |
uvos | ok, but in this state its broken, so i gues a init script somewhere is needed | 12:38 |
uvos | not sure how you do this | 12:38 |
fsmithred | neither am I | 12:38 |
fsmithred | some info here: https://salsa.debian.org/matthew/orphan-sysvinit-scripts | 12:40 |
uvos | "[12:37] <fsmithred> do you know if the old init script still works?" <-- there was no old init script, previous version of iio-sensor-proxy hard dependend on systemd in its code | 12:42 |
uvos | this dependancy was lifted some time after beowulf was frozen | 12:43 |
fsmithred | your forked version has an init script? | 12:43 |
uvos | for openrc yes, as linked | 12:43 |
fsmithred | oh, I guess that would not belong in a package for sysvinit scripts. | 12:44 |
ravehaver9000 | hey everyone, ive been having an issue with the propietary nvidia drivers and my installation of devuan. the issue is that some stuff with the drivers seem to have not installed, and its worrying me. every time i boot up my i see an error message along the lines of "error running install command 'modproble -i nvidia" for module nvidia: retcode 1, and when i install any program with apt--get it tries | 13:00 |
ravehaver9000 | to set up the nvidia persistence daemon, but always fails. how do i solve this problem? im using daedalus and sysvinit if you want to know more | 13:00 |
gnarface | if ravehaver9000 comes back tell them to install the linux-headers-* package corresponding to their current kernel then re-run dpms as root, and just give up on nvidia-persistenced and set environment variable __GL_SHADER_DISK_CACHE_SKIP_CLEANUP=1 instead | 13:56 |
gnarface | also warn them that i've got a heads-up from another channel that the latest nvidia driver release crippled pascal family cards just fair warning, upgrading might be a bad idea right now | 13:57 |
gnarface | (and incidentally, there being a catastrophic crippling regression in the official nvidia linux drivers on the same day as a major world of warcraft release is no longer even remotely plausibly a coincidence) | 14:01 |
unclouded | Hi, debootstrap is giving me "chroot: failed to run command 'dpkg-deb': No such file or directory". It's dead right: There is no dpkg-deb in the target directory. The command is just "debootstrap beowulf /tmp/target http://mirror:3142/devuan", run from a chimaera installation | 21:13 |
unclouded | Does it matter than /tmp is tmpfs? | 21:15 |
markizano | unclouded, mirror:3142 is a valid mirror, right? | 21:35 |
___used | Is vlc 3.0.18 coming up for beowulf? It's not in packages, I'm on 3.0.17 and that has a nasty hole: https://securityonline.info/cve-2022-41325-vlc-media-player-remote-code-execution-vulnerability/ | 21:45 |
Jjp137 | you'll probably have to wait for Debian to update it since I don't think Devuan forks vlc, but you can keep an eye on it here: https://security-tracker.debian.org/tracker/source-package/vlc | 21:48 |
Jjp137 | just remember what each Debian release maps to in Devuan of course | 21:48 |
unclouded | markizano: It's apt-cacher-ng, but the problem is the same when I use http://deb.devuan.org/devuan as the mirror | 21:49 |
unclouded | The problem is the same when run from a beowulf installation too | 22:04 |
unclouded | Doh! The problem is that I was using http://deb.devuan.org/devuan instead of http://deb.devuan.org/merged. Working now | 22:18 |
golinux | Sometimes it's the little things that get you . . . | 22:20 |
unclouded | :) It got me good | 22:22 |
ravehaver9000 | back here, thanks for the advice gnarface | 22:28 |
ravehaver9000 | my card is an ampere one so itd prob should be safe to do that | 22:28 |
ravehaver9000 | whats dpms btw | 22:37 |
ravehaver9000 | i dont really recogmize it at all and want to know what it means before i screw up anything | 22:37 |
unclouded | ravehaver9000: Is it this? https://en.wikipedia.org/wiki/VESA_Display_Power_Management_Signaling | 22:39 |
tux12 | apropos dpms says X11::Protocol::Ext::DPMS (3pm) - Perl module for the X11 Protocol DPMS Extension | 22:40 |
rrq | yes | 22:40 |
ravehaver9000 | ooh, so its a module for perl | 22:47 |
ravehaver9000 | how do i restart it? i tried switching to root and using something like, "system dpms restart" but that ofc wouldnt wokr | 22:48 |
ravehaver9000 | it prob doesnt help that the only init system i actually used was systemd. i used a bit of openrc when i was in artix but ive mostly unlearned it aside from everything starting with rc-service | 22:48 |
ravehaver9000 | ive also tried runit. not for me to say the least | 22:49 |
rrq | no, DPMS is a monitor facility; power saving | 22:49 |
rrq | some docs via "man xset" | 22:50 |
rrq | not so important for LCD and LED monitors but CRTs where costlier | 22:53 |
ravehaver9000 | yeah i can see that, crts def had an issue with that t. owned crt tvs until 2016 | 22:54 |
bbliss | Hey all. I was hoping to figure out why my Intel i5-based 2020 MacBook's keyboard and trackpad do not work at all with the current Daedalus 11/28 preview, and possibly get the necessary support folded into Daedalus before it is released. USB keyboards work fine, and Daedalus generally installs okay, but not having KB+trackpad work out of the box is sorta frustrating. Ideas? Pointers to where | 23:00 |
bbliss | I should be looking? | 23:00 |
rwp | bbliss, That support is entirely within the Linux kernel. I would be looking there for solutions and support. | 23:46 |
rwp | If an older kernel works but a newer kernel fails then that would be a regression in the kernel. That's always easier to argue needs to be fixed. | 23:47 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!