fluffywolf | I have a laptop with a fresh beowulf install. installed back when beowulf was stable, did nothing since except very light web browsing. today I tried upgrading it to daedalus in anticipation of using it for other tasks. it failed miserably. apt has hosed itself and won't run, missing libcrypt. | 04:09 |
---|---|---|
fluffywolf | this is a release-critical failure. a completely standard unused install with no frankendevuaning very much should not implode on an apt-get dist-upgrade. | 04:10 |
gnarface | fluffywolf: i think you forgot that you have to upgrade through chimaera first | 04:13 |
gnarface | you would also have needed to make sure beowulf was fully up-to-date including beowulf-updates and beowulf-security first | 04:14 |
gnarface | and nothing about that process is new with devuan | 04:15 |
gnarface | skipping those steps would have borked a debian upgrade too | 04:15 |
Xenguy | fluffywolf, The truth hurts | 04:16 |
phogg | Skipping releases is a chancy business; sometimes you can get lucky, but it was never expected to work. | 04:17 |
Xenguy | Upgrade from release to release, full stop | 04:17 |
Xenguy | Otherwise, 'yer doin it fuckin wrong' | 04:18 |
gnarface | it's true that people often get away with it, but it depends a lot on what you had installed ahead of time | 04:19 |
gnarface | (the less the better) | 04:19 |
Xenguy | Sure, but it's just bad practice, so why do it | 04:19 |
gnarface | impatience? | 04:20 |
Xenguy | Details matter | 04:20 |
fluffywolf | this was a completely minimal install. | 04:21 |
Xenguy | Impatience in this case leads to wrong procedure, and broken systems, why not minimize one's own pain and do it in 2 steps? | 04:21 |
Xenguy | fluffywolf, Doesn't matter | 04:21 |
fluffywolf | it had... firefox. and that's about it. | 04:21 |
fluffywolf | the only thing I used it for was setting up my christmas lights. lol | 04:21 |
Xenguy | Irrelevent | 04:21 |
Xenguy | You did it wrong | 04:21 |
Xenguy | Maybe you deserve the pain | 04:21 |
fluffywolf | apt shouldn't delete libraries it needs to run. that's not me doing it wrong. that's a bug. | 04:21 |
gnarface | my guess is there was just one security update to one library (possibly libcrypt) that you were missing before you started, and that derailed the whole process | 04:22 |
Xenguy | You refuse to adopt a recommended approach, and then point fingers when it doesn't work out for you | 04:22 |
fluffywolf | I didn't refuse something I had no idea existed. lol | 04:23 |
fluffywolf | and apt imploding due to very normal user behavior is a bug. | 04:23 |
Xenguy | Do I need to serve up the URL on a silver platter? You need to do some basic reading, but it's in the documentation | 04:23 |
gnarface | well, even #debian would often give out bad advice on this matter, but the release guide has always been very clear on it. | 04:23 |
Xenguy | And it's a simple, known, principle | 04:24 |
Xenguy | It's definitely documented, what else can one do? | 04:24 |
fluffywolf | if you just have a release (testing) in your sources, and you update and upgrade after a long time, the entire system self-destructing is not ok. lol | 04:24 |
Xenguy | Don't try to cut corners, or sometimes you're going to get bit | 04:24 |
fluffywolf | I didn't try to cut corners? lol | 04:25 |
Xenguy | OK, so Debian is wrong, and you take no responsibility | 04:25 |
Jjp137 | here's an example of it happening to a Debian user: https://forums.debian.net/viewtopic.php?t=155761 | 04:25 |
Jjp137 | read the first reply. | 04:25 |
Xenguy | (Debian *has* been wrong before ; -) | 04:25 |
gnarface | fluffywolf: wait, so that means you actually tried to upgrade from beowulf -> excalibur directly! | 04:25 |
gnarface | that's bad news! | 04:25 |
gnarface | fluffywolf: daedalus went stable | 04:26 |
gnarface | (also this is a very well known risk of using the "stable/testing/unstable" monikers directly) | 04:26 |
fluffywolf | I changed my sources to say daedalus, so it was definitely to daedalus. | 04:26 |
gnarface | oh | 04:26 |
Xenguy | Same issue | 04:27 |
gnarface | well, still, unless you can reproduce the situation following the proper upgrade path, nothing can be done about it | 04:27 |
gnarface | nothing rational, anyway | 04:27 |
Xenguy | There's nothing *to* be done about it, it's not a bug | 04:27 |
brocashelm | that is why you always upgrade to the release after your current one, instead of the latest release | 04:27 |
brocashelm | a lot of packages get deprecated in between that it confuses apt | 04:27 |
fluffywolf | yes it is a bug. apt should never hose itself without very explicitly making it do so. | 04:27 |
Xenguy | User doesn't follow recommended upgrade path | 04:27 |
Xenguy | Now I shrug | 04:27 |
Xenguy | boo | 04:27 |
Xenguy | hiss | 04:28 |
fluffywolf | having broken packages or such would be less-terrible, but breaking apt itself is very bad. | 04:28 |
Xenguy | You're a bug | 04:28 |
fluffywolf | and you're a dick? | 04:28 |
Xenguy | Only sometimes | 04:28 |
Xenguy | You deserve it though, I've been saving it up | 04:28 |
Xenguy | OT, I know | 04:29 |
brocashelm | as long as you're not doing anything too foolish with your package management (e.g. using third-party repos, trying to get wine to "work", mixing official release repos), apt should work 99.9% of the time | 04:29 |
brocashelm | had to learn this the hard way | 04:29 |
brocashelm | i realized i still had some dmo-versioned packages that i needed to correct to match devuan's | 04:29 |
fluffywolf | this was a completely plain install with no frankendebianing. | 04:30 |
* Xenguy offers some cheese with that... | 04:30 | |
brocashelm | i know, but it was still a mistake to jump to daedalus from beowulf (if i'm understanding correctly) | 04:30 |
Xenguy | Bingo | 04:31 |
Xenguy | That's the issue | 04:31 |
fluffywolf | apt should be smart enough to install packages in the correct order to not break itself. | 04:31 |
Xenguy | It's not rocket science | 04:31 |
brocashelm | things that happened during chimaera get skipped | 04:31 |
Xenguy | Don't do that okay? | 04:31 |
Xenguy | Case closed | 04:31 |
fluffywolf | "the wheel falls off if I turn the car too sharp to the left" means you need to fix the car, not tell the driver not to turn so sharp. | 04:31 |
brocashelm | such as python 2 -> 3 | 04:31 |
Xenguy | aghhhhh | 04:31 |
Jjp137 | just found this lol: https://www.debian.org/releases/bookworm/amd64/release-notes/ch-information.en.html#libcrypt-upgrade-from-buster | 04:32 |
Jjp137 | you might be able to salvage it ikd. | 04:32 |
Jjp137 | idk* | 04:32 |
gnarface | hmm, fluffywolf ^ this is worth a try | 04:33 |
fluffywolf | you're taking an obvious bug and blaming it on the user. | 04:33 |
fluffywolf | Jjp137: thanks! I didn't find that googling. trying it now. | 04:33 |
Jjp137 | np | 04:33 |
brocashelm | apt doesn't think that way, so it has to be done very meticulously | 04:34 |
brocashelm | i usually try aptitude dist-upgrade to get an all-clear on the upgrade | 04:35 |
brocashelm | *safe-upgrade | 04:35 |
fluffywolf | then apt should give a warning about trying to install more than one release number and require forcing to do it. | 04:35 |
fluffywolf | I wish I had aptitude on that laptop... trying to install aptitude manually was going to be my next step. heh. | 04:35 |
brocashelm | i would agree that it should give more warnings like that, but has it been suggested already to debian? perhaps file a bug report with this suggestion | 04:37 |
gnarface | hah, that won't work | 04:37 |
brocashelm | forgot the /s | 04:37 |
fluffywolf | hrmm. apt-get update is giving "error: timeout was reached" without specifying WHAT timed out, all files look ok? | 04:37 |
gnarface | no traction on fixes for this from them will ever happen. i've tried. same thing with forbidden downgrades... nobody over there is interested in warning people not to shoot their foot off | 04:38 |
fluffywolf | the instructions on that page give a really cryptic permission denied message | 04:38 |
* fluffywolf attempts to type | 04:39 | |
brocashelm | gnarface: i ctrl+f plenty of "warning" in the title of apt's bug reports that are nearly two decades old still active: https://bugs.debian.org/cgi-bin/pkgreport.cgi?dist=unstable;package=apt#_0_6_0 | 04:39 |
fluffywolf | download is performed unsandboxed as root as file [...]libcrypt[...].deb couldn't by accessed by user _apt pkgacquire:run 13 permission denied | 04:39 |
gnarface | fluffywolf: make sure mktemp didn't fail | 04:40 |
gnarface | it's probably assuming you have normal permissions on /tmp, i'd assume | 04:40 |
fluffywolf | it seems to have worked despite that message | 04:41 |
fluffywolf | specifically, apt is trying to actually download and install packages again, not immediately fail. | 04:42 |
fluffywolf | yay! | 04:42 |
gnarface | oh, probably then the warning was failure to preserve permissions of the extracted files | 04:42 |
gnarface | (maybe due to umask settings or whatever the /tmp directory permissions were) | 04:43 |
Jjp137 | eh I think usually the _apt user is used as a sandbox user but it failed to work so it just did it as root | 04:43 |
fluffywolf | the bug linked there says there's a trivial fix, but they're not adding it because... the maintainer of the package involved is a dick and no one wants to do any work? | 04:45 |
fluffywolf | I like debian less and less with time. | 04:45 |
fluffywolf | adding a single fucking breaks: to the control file fixes it. | 04:46 |
fluffywolf | let's use systemd, ditch debian menus, force freedesktop crap into absolutely everything, and have a broken process where individual maintainers can screw everyone. | 04:47 |
gnarface | fluffywolf: i think it's clear that's why we're all here, but take the ranting to #devuan-offtopic | 04:48 |
fluffywolf | yay, /etc/issue says daedalus now. | 04:50 |
fluffywolf | grr. forgot firefox got even shittier and needs extensive fucking with to make usable. can we make killing sanwich menus a goal for devuan? :P | 04:51 |
fluffywolf | sigh... one of the main reasons I upgraded was hibernate didn't work. hibernate still does not work. clicking xfce's hibernate button does exactly nothing. | 04:57 |
fluffywolf | what do I install to make power management work? | 04:58 |
gnarface | acpi and uh... | 04:58 |
gnarface | and pm-utils | 04:58 |
gnarface | might need something besides that but start with that if you don't have it | 04:59 |
fluffywolf | installing | 04:59 |
gnarface | if it still doesn't work but /usr/sbin/pm-hibernate does work (as root) then you might need to switch graphical logins or run Xorg suid root or something | 05:00 |
gnarface | if it pm-hibernate doesn't even work, you need to make sure your swap partition is large enough and that the initramfs tools has the right UUID for it | 05:00 |
fluffywolf | well, I have pm-hibernate now at least, which I didn't before... still does nothing. checking logs. | 05:00 |
fluffywolf | nothing in syslog, messages, daemon.log, or dmesg, just fails silently as far as I can see... | 05:01 |
gnarface | make sure you have acpi-support-base too but i think it should have come with acpi | 05:01 |
gnarface | fails silently even when running pm-hibernate directly as root? | 05:01 |
fluffywolf | yes | 05:02 |
fluffywolf | tried it from vt1 just to make sure it wasn't dumping anything to console, nope | 05:02 |
gnarface | is your swap partition at least as large as your amount of physical ram? | 05:03 |
gnarface | swap partition has to be equal or larger than the size of physical ram for hibernate to work | 05:03 |
fluffywolf | I should check. usually I always do that, but it's been so long since I installed this laptop... | 05:03 |
fluffywolf | yes, I have a 2GB swap with my 2GB ram. | 05:03 |
fluffywolf | none of which is used, because nothing is running | 05:04 |
gnarface | if you run "update-initramfs -u" it should mention in the output what the UUID of the swap partition it will resume from is, and if it's wrong you have to fix it in the config in there somewhere (it will say where) and run update-initramfs again | 05:04 |
fluffywolf | usually power management failures get logged somewhere... I've certainly seen errors about not enough free swap. heh. | 05:05 |
fluffywolf | checked kern.log too, nothing there | 05:05 |
gnarface | check the initramfs thing. bbl | 05:05 |
fluffywolf | hrmm. it seems "man" is broken. | 05:06 |
fluffywolf | man fails with bad system call errors. that's one I've never seen before. | 05:06 |
brocashelm | for only 2 gb of physical ram, i'd immediately look into doubling that amount for swap | 05:06 |
brocashelm | although even with more swap space, hibernate hardly ever "works" for me; suspend is usually good enough to take a breather from using my laptop | 05:07 |
brocashelm | although i don't know if it's due to using swapfiles instead of swap partitions on my part | 05:07 |
fluffywolf | you only need more swap than the compressed size of used system ram... it's not a swapfile size issue. | 05:08 |
fluffywolf | I'm trying to figure out why man is broken now. I'm finding google stuff about apparmor problems from 2018 that don't seem current enough. | 05:08 |
fluffywolf | https://bugs.archlinux.org/task/57557 this | 05:09 |
fluffywolf | now as much as you like blaming me, I'm pretty sure I did not cause this, and it's a bug. lol | 05:09 |
fluffywolf | I can't find anything newer than 2018. | 05:10 |
fluffywolf | removing apparmor and rebooting didn't fix it | 05:12 |
fluffywolf | found bug reports on it, all saying it's been fixed... I think something didn't get upgraded that needs to be. | 05:14 |
* fluffywolf re-upgrades | 05:14 | |
fluffywolf | apt-get upgrade wants to upgrade another 866 packages now. let's see if that does it... | 05:15 |
fluffywolf | this will be a while... (slow laptop) | 05:15 |
gnarface | fluffywolf: you did "apt-get dist-upgrade" right? | 05:27 |
gnarface | maybe better to do "apt-get --no-install-recommends dist-upgrade" | 05:27 |
fluffywolf | yep. (apt-get dist-upgrade) | 05:32 |
fluffywolf | the last apt-get upgrade seems to have broken wifi. it fails with bad password. I did not change the password. lol | 05:32 |
fluffywolf | rebooting... | 05:33 |
fluffywolf | it got new wifi management stuff? | 05:35 |
fluffywolf | says "device not ready" | 05:35 |
fluffywolf | I seem to now have "NetworkManager Applet 1.30.0" and it won't do shit? | 05:35 |
fluffywolf | I don't have the possibility of a wired connection here. | 05:36 |
fluffywolf | googling finds only systemd problems? | 05:37 |
fluffywolf | wpa_supplicant can do its thing just fine, and I have the connection up, but whatever the fuck NetworkManager Applet is seems to be broken. | 05:40 |
fluffywolf | the wifi is working perfect; this problem seems to be bugged. | 05:41 |
fluffywolf | program | 05:41 |
fluffywolf | I'm guessing it's impossible to reinstall wicd? | 05:42 |
fluffywolf | grrrrrrrrrr. why the fuck was a working tool replaced with what's apparently a total shit non-working tool? | 05:50 |
fluffywolf | I very fucking much do not want to be manually configuring wpa_supplicant for each network. | 05:50 |
fluffywolf | anyone have any ideas why NetowkrMaanagerApplet is refusing to function in any way? I am very much not in the mood to dig through source to this crap. | 05:51 |
fluffywolf | yay, I think I found something useful googling. | 05:53 |
fluffywolf | there's some bug between networkmanager and certain firmware-iwlwifi versions. if it says device not ready, you need to either upgrade or downgrade firmware-iwlwifi. firmware-iwlwifi got moved to non-free-firmware and apt forgot it exists and will do neither until you add that section. | 05:54 |
fluffywolf | adding the new section and upgrading the package seems to have made networkmanagerapplet stop saying that | 05:55 |
fluffywolf | looks like the bug reports are entirely ignored | 05:59 |
fluffywolf | .... now back to figuring out why pm-hibernate still silently fails | 06:00 |
fluffywolf | pm-suspend works | 06:01 |
fluffywolf | but now pm-hibernate logs errors! "hibernation is restricted". what the hell? | 06:02 |
fluffywolf | apparently hibernation is now disabled as a feature. | 06:03 |
fluffywolf | because you can twiddle the hibernation image | 06:03 |
fluffywolf | I am very, very, very unconcerned about someone twiddling the hibernation image on the onboard flash on this shitty laptop I use for trivial tasks and don't even have a user account with my name on. I do, however, want it to fucking suspend! | 06:08 |
gnarface | fluffywolf: someone else just got it working a couple days ago, could it be hardware specific? | 06:16 |
gnarface | it shouldn't be, but i'm running out of ideas | 06:16 |
gnarface | suspend doesn't use swap, it's completely different | 06:16 |
gnarface | hibernate is supposed to be the one that doesn't require any special hardware | 06:17 |
fluffywolf | I just got it working. the problem is it's now disabled by default as a feature. | 06:18 |
gnarface | in the window manager UI only maybe? | 06:18 |
fluffywolf | hibernate is insecure because you could use another OS to modify the saved kernel image and haxx0r it. | 06:19 |
fluffywolf | so it's disabld. | 06:19 |
fluffywolf | it's a kernel feature | 06:19 |
gnarface | my google searches about the matter seem to suggest it's a problem with "kernel_lockdown" | 06:19 |
fluffywolf | yep | 06:19 |
gnarface | you running a non-stock kernel? | 06:19 |
gnarface | or some non-stock config? | 06:20 |
fluffywolf | and debian has also disabled all the ways to disable kernel lockdown except for turning off secure boot, apparently. | 06:20 |
fluffywolf | completely stock | 06:20 |
gnarface | oh, it's a problem with secure boot | 06:20 |
gnarface | now i see | 06:20 |
fluffywolf | so I turned off secure boot, thus disabling all security with a large hammer, because debian disabled all the fine-grained control. | 06:20 |
fluffywolf | heh. | 06:20 |
gnarface | the person who got it working last may have been using a legacy bios | 06:20 |
fluffywolf | as of daedalus, secure boot = no hibernate | 06:21 |
gnarface | good to know... | 06:21 |
fluffywolf | the man failure is also a security thing. you can't use old man binary with new kernels. | 06:21 |
fluffywolf | have to upgrade man-db if you upgrade kernel | 06:21 |
fluffywolf | why the bloody fuck there's ANY relation between a utility to show text manual pages, and kernel versions, is a bug itself... but there you have it. | 06:22 |
fluffywolf | https://bugs.archlinux.org/task/57557 if you get these errors trying to view man pages, you have a kernel version and man-db version issue. | 06:24 |
gnarface | i would assume it has to do with the man page auto-formatting whatever, i forget what it's called. | 06:24 |
gnarface | (you could always disable it but then man pages stop stretching to terminal width automatically unless you're running man as root) | 06:24 |
fluffywolf | no, it's kernel security stuff | 06:24 |
gnarface | i remember that being a security issue | 06:25 |
gnarface | considered one in general | 06:25 |
gnarface | maybe that's the "why" you're looking for | 06:25 |
gnarface | because like you said, i can't imagine what else it would be... | 06:25 |
fluffywolf | this upgrade has been very painful in multiple ways. | 06:26 |
gnarface | so far honestly i haven't had trouble anywhere i've tried it | 06:26 |
gnarface | the feedback has been good for the most part | 06:26 |
fluffywolf | is there documentation saying you have to add non-free-firmware to sources before upgrading, that I didn't read? | 06:26 |
gnarface | probably the release notes | 06:26 |
gnarface | hmm... or maybe not | 06:27 |
fluffywolf | it's not in https://files.devuan.org/devuan_daedalus/Release_notes.txt , not even in the non-free firmware section | 06:27 |
gnarface | someone probably should add that though | 06:27 |
fluffywolf | debian moved firmware from non-free to non-free-firmware, so your firmware won't get upgraded unless you add the new section | 06:28 |
gnarface | yea, i was aware but people were warning us about it in here well ahead of the daedalus release | 06:29 |
fluffywolf | and networkmanager apparently has bugs with specific firmware-iwlwifi versions, where you have to downgrade or upgrade it (either works according to posts) or networkmanager gets stuck at "device not ready" and you have no connection unless you manually configure wpa_supplicant. | 06:29 |
gnarface | noted | 06:29 |
fluffywolf | and wicd went away. | 06:30 |
gnarface | i think that happened last release | 06:30 |
fluffywolf | even though it actually, you know, worked. | 06:30 |
fluffywolf | lol | 06:30 |
gnarface | no maintainer | 06:30 |
deadbabies | Hey uh | 06:30 |
deadbabies | did I do a bad by changing my sources to daedalus | 06:31 |
deadbabies | or is that how I'm supposed to upgrade | 06:31 |
fluffywolf | I haven't been here much lately, didn't see any warnings... life has been keeping me very busy. | 06:31 |
deadbabies | :V | 06:31 |
deadbabies | same | 06:31 |
fluffywolf | deadbabies: if you're using chimaera, that's the first step to upgrading, yes. if you're using something older, upgrade to chimaera first. | 06:31 |
deadbabies | I was using chimaera | 06:32 |
deadbabies | what else am I supposed to do | 06:32 |
deadbabies | I apt-get update and upgraded | 06:32 |
gnarface | deadbabies: if you were using non-free for drivers they moved to non-free-firmware | 06:32 |
gnarface | bbl | 06:32 |
deadbabies | Ah | 06:32 |
deadbabies | thanks | 06:32 |
fluffywolf | then change sources.list to daedalus, apt-get update, apt-get dist-upgrade. if you have non-free firmware, add non-free-firmware after main contrib non-free or whatever sections you use. | 06:32 |
deadbabies | okay | 06:32 |
deadbabies | thanks | 06:32 |
deadbabies | let me do that | 06:33 |
deadbabies | thankfully this is on a second computer so I can keep this open lol | 06:33 |
fluffywolf | bbl, wolfy bedtime | 06:34 |
fluffywolf | side note: can we please stop sticking "ae" in names? ;P | 06:35 |
fluffywolf | bbl | 06:35 |
deadbabies | ah | 06:36 |
deadbabies | it was that dist-upgrade part that I was missing I think | 06:36 |
deadbabies | thanks a lot | 06:41 |
deadbabies | Um | 07:13 |
deadbabies | What locale do I want if I use english and want yyyy-mm-dd and 24 hour time? | 07:14 |
rrq | used to be LC_TIME=en_DK I think .. or maybe en_DK.UTF-8 | 07:17 |
deadbabies | I keep getting these cannot set lc_ctype and lc_messages and lc_all errors | 07:17 |
deadbabies | ok | 07:17 |
deadbabies | ... any idea how I set that? xD | 07:18 |
rrq | nonot off=hand | 07:19 |
deadbabies | let's see how bad my google fu is.... | 07:20 |
rrq | maybe: update-locale LC_TIME=en_DK.UTF-8 | 07:20 |
gnarface | deadbabies: make sure the locales package is installed then run "dpkg-reconfigure locales" | 07:39 |
deadbabies | gnarface I think that's gonna work | 07:39 |
deadbabies | update-locale LC_TIME spit me out errors because I apparently didn't have the other ones set? | 07:40 |
deadbabies | thanks | 07:40 |
gnarface | if you were missing locales you might be missing tzdata too | 07:40 |
[-_-] | hi | 08:58 |
[-_-] | where are the logs for Xorg ? | 08:58 |
rrq | either /var/log/ or ~/.local/share/xorg/ | 08:59 |
[-_-] | thanks | 09:01 |
[-_-] | Hiiiii | 10:57 |
[-_-] | guys, I tried to disable modesetting from kernel parameters, and this solved the issue for me ( changing to other tty freezes Xorg), but debian page (https://wiki.debian.org/KernelModesetting) says "On Intel and Radeon cards KMS is disabled at run time in Debian kernel images by default.1 It is enabled via modprobe configuration files provided by the xserver-xorg-video-intel and | 10:59 |
[-_-] | xserver-xorg-video-radeon packages." | 11:00 |
[-_-] | where are they loading this module from ? | 11:00 |
[-_-] | how do I disable that? | 11:00 |
[-_-] | I mean if it is default, why doing it to GRUB works? | 11:02 |
[-_-] | and how do I prevent Xorg from loading those without going GRUB way? | 11:02 |
gnarface | it's probably referring to /etc/modprobe.d/radeon-kms.conf and /etc/modprobe.d/i915-kms.conf | 11:03 |
[-_-] | none of those files exist | 11:04 |
gnarface | do you have xserver-xorg-video-radeon and xserver-xorg-video-intel installed? | 11:04 |
gnarface | each of them are one line | 11:05 |
gnarface | options i915 modeset=1 | 11:05 |
gnarface | options radeon modeset=1 | 11:05 |
gnarface | you could create a .conf file easily and set them to 0 | 11:05 |
gnarface | i never tried it that way but it should work | 11:05 |
[-_-] | I don't have the intel package installed | 11:06 |
[-_-] | debian wiki suggests default one | 11:06 |
[-_-] | .xserver-xorg-core | 11:06 |
[-_-] | I do have the radeon one, but radeon is idk, PRIME GPU OFF Loading thinggy | 11:07 |
[-_-] | does that affect my Xorg? idk | 11:07 |
gnarface | the files in /etc/modprobe.d/ are only relevant to the kernel modules | 11:08 |
gnarface | the kernel modules may be loaded even without the xorg package that's supposed to accompany them | 11:08 |
gnarface | (since the kernel modules come with the kernel, not the xorg package) | 11:08 |
gnarface | you can create any file name in /etc/modprobe.d/ for yourself that you want as long as it ends with .conf and it will be obeyed | 11:09 |
gnarface | you can set modeset to 0 or you can just blacklist the kernel modules if you prefer | 11:10 |
[-_-] | hmmmm | 11:10 |
gnarface | i feel like you probably want them and the xorg packages though | 11:10 |
[-_-] | will this work ' echo "nomodeset" > /etc/modprobe.d/modesetting.conf ' | 11:15 |
[-_-] | lemme test what I did | 11:29 |
[-_-] | guys, turning intel modesetting off takes away /sys/class/backlight/intel_backlight | 12:40 |
[-_-] | ;_; | 12:40 |
[-_-] | and the display remains at constant super ultra pro max brightness | 12:41 |
[-_-] | this fixed my issue : https://dev1galaxy.org/viewtopic.php?id=5865 | 15:05 |
[-_-] | but but I did not test multi Xorg multi user session | 15:05 |
[-_-] | wasted a whole day on this ;_; | 15:05 |
[-_-] | does this happen to you guys? | 15:09 |
fluffywolf | [-_-]: all the fucking time. also, yak shaving. | 15:42 |
[-_-] | :D | 15:43 |
rwp | Did https://dev1galaxy.org/viewtopic.php?id=5865 list a fix? I don't see a fix there. Just confirmation of the problem. | 18:47 |
fsmithred | rwp, no fix on that problem yet. I also tried running a second xserver as the same user, and that didn't work, either. | 19:37 |
rwp | I guess it is one step forward and two steps backward then. Sigh. (I am going to be installing on a new laptop for me a little later today.) | 19:39 |
fsmithred | I haven't tried it with a display manager | 19:41 |
fsmithred | might not be a problem there | 19:41 |
rwp | I'll report what I find. | 19:42 |
fsmithred | thanks | 19:43 |
fsmithred | rwp, I just added lxdm and I can now start a second xsession with a different user. | 19:48 |
fsmithred | oh, and the xserver is now running as root instead of user | 19:53 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!