The Linux Questions Thread: a bunch of pitfalls, but technically it's possible

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > The Linux Questions Thread: a bunch of pitfalls, but technically it's possible

«‹›1037 »

AlexDeGruven: Jun 29, 2007; Watch me pull my dongle out of this tiny box

That's fair. And it would also help with the LUKS issue, since the system will already be booted.

# ? Feb 17, 2024 06:49

Adbot: ADBOT LOVES YOU

# ? May 31, 2024 17:29

Mega Comrade: Apr 22, 2004; Listen buddy, we all got problems!

My new 7800xt arrived yesterday and I just spent 5 hours trouble shooting.
Thought I was going crazy, but no https://gitlab.freedesktop.org/drm/amd/-/issues/3062
It works on cold boot but fails on restart for late 6.6 and all 6.7 kernels.

Guess I'm staying on 6.5 for a while.

# ? Feb 17, 2024 09:24

VictualSquid: Feb 29, 2012; Gently enveloping the target with indiscriminate love.

I use this script to check the health of my zfs disks and email me:
https://gist.github.com/petervanderdoes/bd6660302404ed5b094d

It checks for this list of failure warnings:

code:

condition=$(/sbin/zpool status | egrep -i '(DEGRADED|FAULTED|OFFLINE|UNAVAIL|REMOVED|FAIL|DESTROYED|corrupt|cannot|unrecover)')

It started reporting my pool as faulty after I removed a redundant device. Because it says :

code:

remove: Removal of vdev 0 copied 2.61T in 16h27m, completed on Thu Feb  8 15:05:22 2024
        4.87M memory used for [b]removed[/b] device mappings

The easy fix is to just delete the condition to check for REMOVED. Is there any actual fault that I would miss from that? And alternatively is there a way to remove the removal report from "zpool status"?

# ? Feb 17, 2024 10:59

Saukkis: May 16, 2003; Unless I'm on the inside curve pointing straight at oncoming traffic the high beams stay on and I laugh at your puny protest flashes.
I am Most Important Man. Most Important Man in the World.

Bash tab completion has an annoyance with quotes and directories with spaces. I can tab-complete the directory but it will end with ("/). If I want to continue tab-completing for files and retain the quote I need to remove the closing quote before continuing tabbing. Is there a way around this, like if the initial tab complete would end with only (/).

code:

$ mkdir "Spaced Directory"
$ touch "Spaced Directory/Spaced File.txt"
$ vim "Spa<TAB>
$ vim "Spaced Directory"/
$ vim "Spaced Directory"/Sp<TAB>
$ vim Spaced\ Directory/Spaced\ File.txt

$ vim "Spaced Directory/<TAB>
$ vim "Spaced Directory/Spaced File.txt"

# ? Feb 18, 2024 11:57

Subjunctive: Sep 12, 2006; ✨sparkle and shine✨

I don�t think so, but I�ll admit I only skimmed the info page to double-check, so there could be some setopt sysv_alternate_path_quoting thing hiding out.

Is this an aesthetic concern, or is there actually a scenario where the result is different?

# ? Feb 18, 2024 15:32

hazzlebarth: May 13, 2013

Saukkis posted:

Bash tab completion has an annoyance with quotes and directories with spaces. I can tab-complete the directory but it will end with ("/). If I want to continue tab-completing for files and retain the quote I need to remove the closing quote before continuing tabbing. Is there a way around this, like if the initial tab complete would end with only (/).

You don't need the quotes, bash will escape the space when tab completing(\):

code:

$ vim Spa<TAB>
$ vim Spaced\ Dir/Spaced\ File

# ? Feb 18, 2024 15:33

Nitrousoxide: May 30, 2011; do not buy a oneplus phone

To return briefly to BTRFS, I did find a neat gui client which lets you manage the common functions of BTRFS and Snapper without having to dive into the terminal.

https://gitlab.com/btrfs-assistant/btrfs-assistant

It gets you (pretty much) the timeshift experience with the much greater snapper flexibility.

Nitrousoxide fucked around with this message at 15:54 on Feb 18, 2024

# ? Feb 18, 2024 15:50

Saukkis: May 16, 2003; Unless I'm on the inside curve pointing straight at oncoming traffic the high beams stay on and I laugh at your puny protest flashes.
I am Most Important Man. Most Important Man in the World.

hazzlebarth posted:

You don't need the quotes, bash will escape the space when tab completing(\):

Or, right, the main point was that I don't want escaping. This annoyance usually comes up when I want to rename files and escaping all the spaces and other special characters would be a hassle.

# ? Feb 18, 2024 17:58

Subjunctive: Sep 12, 2006; ✨sparkle and shine✨

But the shell will escape them for you, which is why it�s doing that? You can open a " anywhere, not just at the beginning, so

Sh code:

subjunctive@host $ rm Dir\ With \Spaces/"file* {name"

works fine. just open a " or ' when you want one, the shell will take care of itself when tabbing

# ? Feb 18, 2024 18:04

Saukkis: May 16, 2003; Unless I'm on the inside curve pointing straight at oncoming traffic the high beams stay on and I laugh at your puny protest flashes.
I am Most Important Man. Most Important Man in the World.

Subjunctive posted:

But the shell will escape them for you, which is why it�s doing that? You can open a " anywhere, not just at the beginning, so

works fine. just open a " or ' when you want one, the shell will take care of itself when tabbing

That doesn't work for me, maybe bash-completion changes the behavior. For example if I try complete the file "Spaced Directory/Spaced File.txt" I get

code:

# vim Spaced\ Directory/"Sp<TAB>
# vim Spaced\ Directory/"Spaced Directory"/

which definitely seems like the wrong behavior. When I need to do the rename I usually complete to some existing file name and the edit the target, so I would have to add spaces and escapes manually.

# ? Feb 18, 2024 18:27

Subjunctive: Sep 12, 2006; ✨sparkle and shine✨

Saukkis posted:

That doesn't work for me, maybe bash-completion changes the behavior. For example if I try complete the file "Spaced Directory/Spaced File.txt" I get
code:
# vim Spaced\ Directory/"Sp<TAB>
# vim Spaced\ Directory/"Spaced Directory"/
which definitely seems like the wrong behavior. When I need to do the rename I usually complete to some existing file name and the edit the target, so I would have to add spaces and escapes manually.

just put quotes in around the part you�re editing? they don�t have to be around the whole thing, so you don�t have to do any non-local editing, just type "* middle" wherever you need it. I�m not sure what user experience you�re trying to have exactly, tbh.

if this is a case of wanting the second argument to be a variant of the first, like mv "thing stuff 2.txt" �thing-stuff-2.txt� then you might be able to get there better with something like mv "thing stuff 2.txt" #!:1gs/ /-/ (on my phone so I can�t test it, but bash history expansion can also be applied to previous parts of the current command line)

this might also be one of those times that writing a function or short script is worthwhile, if the renames are of a similar transformation each time

# ? Feb 18, 2024 18:46

Saukkis: May 16, 2003; Unless I'm on the inside curve pointing straight at oncoming traffic the high beams stay on and I laugh at your puny protest flashes.
I am Most Important Man. Most Important Man in the World.

Okay yeah, it does seem to work by inputing
# mv Spaced\ Directory/Spaced\ File.txt Spaced\ Directory/Spaced\ File" 2 ".txt
Never thought quotes could be in the middle.

# ? Feb 18, 2024 19:01

Subjunctive: Sep 12, 2006; ✨sparkle and shine✨

Ah yeah, that�s what I meant by �just open a " or ' when you want one� but without an example that probably wasn�t clear, sorry!

# ? Feb 18, 2024 19:04

Framboise: Sep 21, 2014; To make yourself feel better, you make it so you'll never give in to your forevers and live for always.; Lipstick Apathy

Reaching my wit's end with this suspend issue. I get that it's most likely because of my GPU but I've tried everything to the extent of my understanding at the moment. I even started a fresh install of EndeavourOS with KDE instead of GNOME to see if it made a difference. It doesn't, but I do like it more as a DE I guess so there's that.

I've tried working with information from this link here https://download.nvidia.com/XFree86/Linux-x86_64/510.47.03/README/powermanagement.html, and "Video Memory Self Refresh is "Not Supported". Does that mean I'm 100% screwed here until I update my GPU?

I've tried blacklisting nvidiafb as suggested here https://wiki.archlinux.org/title/Power_management/Suspend_and_hibernate

I've tried adding "mem_sleep_default=deep" to /etc/default/grub according to this https://devnull.land/laptop-s2idle-to-deep and got nothing there, too.

It's silly that this is really the one struggle I'm having (outside of having trouble converting Windows 10 to GPT so I can boot to it from grub) and I'm enjoying everything else, but it's hard to get settled in when I can't use a very basic function.

edit:
fwiw I'm using the lts kernel on EndeavourOS, Nvidia GeForce GTX980 Ti GPU.

# ? Feb 19, 2024 21:26

zhar: May 3, 2019

Bit of a long shot but the only problem I have had with the display when waking from suspend was if I had it switched off rather than in standby and switched it on after waking from suspend.

It should be easy to switch between kde x11 and wayland if you want to isolate a bit more (iirc there's a button on the display manager, you may also have to install some package) but otherwise sounds like nvidia striking again.

# ? Feb 19, 2024 22:52

Framboise: Sep 21, 2014; To make yourself feel better, you make it so you'll never give in to your forevers and live for always.; Lipstick Apathy

Seems to behave the same regardless of whether I'm using X11 or Wayland.

To be precise, what is happening is that the PC will go to sleep almost instantly after I give the command-- and it is indeed in a low-power state as my PC's power light begins blinking rather than shutting it off.

When I try to wake it, it just wakes to a blank screen. I cannot tell if it is actually waking up in the background or not.

Don't know if this is helpful at all or not, but I also tried this (snipping out previous attempts, they're all the same):

code:

framboise@mochi ~> journalctl | grep "suspend"
Feb 19 16:55:53 mochi systemd-logind[480]: The system will suspend now!
Feb 19 16:55:54 mochi systemd-sleep[17885]: Performing sleep operation 'suspend'...
Feb 19 16:55:54 mochi kernel: PM: suspend entry (deep)
framboise@mochi ~> journalctl | grep "wake"
Feb 19 16:56:41 mochi kernel: rtc_cmos 00:02: RTC can wake from S4
framboise@mochi ~> journalctl | grep "resume"
Feb 19 16:56:41 mochi kernel: Command line: BOOT_IMAGE=/@/boot/vmlinuz-linux-lts root=UUID=6a982759-ee6f-450d-9c4d-6f5c960978fd rw rootflags=subvol=@ mem_sleep_default=deep nowatchdog nvme_load=YES resume=UUID=303d2b09-8b7e-4604-adc9-e6075ca2705a nvidia-drm.modeset=1 loglevel=3
Feb 19 16:56:41 mochi kernel: Kernel command line: BOOT_IMAGE=/@/boot/vmlinuz-linux-lts root=UUID=6a982759-ee6f-450d-9c4d-6f5c960978fd rw rootflags=subvol=@ mem_sleep_default=deep nowatchdog nvme_load=YES resume=UUID=303d2b09-8b7e-4604-adc9-e6075ca2705a nvidia-drm.modeset=1 loglevel=3

Framboise fucked around with this message at 23:04 on Feb 19, 2024

# ? Feb 19, 2024 22:58

Subjunctive: Sep 12, 2006; ✨sparkle and shine✨

Do you see other activity in the syslog after you wake, but before you reboot? Just seeing it through a grep isn�t going to tell that much, you�ll need to look at the timestamps and see if it�s starting service activity again.

(You could also start sshd and test that way.)

# ? Feb 20, 2024 00:23

ziasquinn: Jan 1, 2006; Fallen Rib

Framboise posted:

Seems to behave the same regardless of whether I'm using X11 or Wayland.

To be precise, what is happening is that the PC will go to sleep almost instantly after I give the command-- and it is indeed in a low-power state as my PC's power light begins blinking rather than shutting it off.

When I try to wake it, it just wakes to a blank screen. I cannot tell if it is actually waking up in the background or not.

Don't know if this is helpful at all or not, but I also tried this (snipping out previous attempts, they're all the same):
code:
framboise@mochi ~> journalctl | grep "suspend"
Feb 19 16:55:53 mochi systemd-logind[480]: The system will suspend now!
Feb 19 16:55:54 mochi systemd-sleep[17885]: Performing sleep operation 'suspend'...
Feb 19 16:55:54 mochi kernel: PM: suspend entry (deep)
framboise@mochi ~> journalctl | grep "wake"
Feb 19 16:56:41 mochi kernel: rtc_cmos 00:02: RTC can wake from S4
framboise@mochi ~> journalctl | grep "resume"
Feb 19 16:56:41 mochi kernel: Command line: BOOT_IMAGE=/@/boot/vmlinuz-linux-lts root=UUID=6a982759-ee6f-450d-9c4d-6f5c960978fd rw rootflags=subvol=@ mem_sleep_default=deep nowatchdog nvme_load=YES resume=UUID=303d2b09-8b7e-4604-adc9-e6075ca2705a nvidia-drm.modeset=1 loglevel=3
Feb 19 16:56:41 mochi kernel: Kernel command line: BOOT_IMAGE=/@/boot/vmlinuz-linux-lts root=UUID=6a982759-ee6f-450d-9c4d-6f5c960978fd rw rootflags=subvol=@ mem_sleep_default=deep nowatchdog nvme_load=YES resume=UUID=303d2b09-8b7e-4604-adc9-e6075ca2705a nvidia-drm.modeset=1 loglevel=3

in my experience this is an extension issue crashing Gnome.

# ? Feb 20, 2024 01:40

Klyith: Aug 3, 2007; GBS Pledge Week

(For reference, here's what Subjunctive is asking for: after rebooting due to the black screen, use journalctl -b -1 to see the journal for the previous boot, press end to look at the last stuff that happened.)

-b shows only the journal from a particular boot session. -b 0 (or no number) is the current one, -1 is the previous boot, and so on.

# ? Feb 20, 2024 01:54

Framboise: Sep 21, 2014; To make yourself feel better, you make it so you'll never give in to your forevers and live for always.; Lipstick Apathy

ziasquinn posted:

in my experience this is an extension issue crashing Gnome.

I'm using KDE now fwiw. Don't really know if that changes much.

Subjunctive posted:

Do you see other activity in the syslog after you wake, but before you reboot? Just seeing it through a grep isn�t going to tell that much, you�ll need to look at the timestamps and see if it�s starting service activity again.

(You could also start sshd and test that way.)

Klyith posted:

(For reference, here's what Subjunctive is asking for: after rebooting due to the black screen, use journalctl -b -1 to see the journal for the previous boot, press end to look at the last stuff that happened.)

-b shows only the journal from a particular boot session. -b 0 (or no number) is the current one, -1 is the previous boot, and so on.

Okay, here's what I have-- I got logs from testing the issue with both Wayland and X11, if it helps at all:

code:

Wayland:
Feb 19 22:07:46 mochi plasmashell[781]: qt.qpa.wayland: Wayland does not support QWindow::requestActivate()
Feb 19 22:07:47 mochi systemd-logind[476]: The system will suspend now!
Feb 19 22:07:47 mochi NetworkManager[518]: <info>  [1708398467.8940] manager: sleep: sleep requested (sleeping: no  enabled: yes)
Feb 19 22:07:47 mochi NetworkManager[518]: <info>  [1708398467.8941] device (wlan0): state change: disconnected -> unmanaged (reason 'sleeping', sys-iface->
Feb 19 22:07:47 mochi NetworkManager[518]: <info>  [1708398467.9404] device (wlan0): set-hw-addr: reset MAC address to  (unmanage)
Feb 19 22:07:47 mochi kernel: rtl8192cu: MAC auto ON okay!
Feb 19 22:07:47 mochi kernel: rtl8192cu: Tx queue select: 0x05
Feb 19 22:07:48 mochi NetworkManager[518]: <info>  [1708398468.4357] device (p2p-dev-wlan0): state change: disconnected -> unmanaged (reason 'sleeping', sy>
Feb 19 22:07:48 mochi NetworkManager[518]: <info>  [1708398468.4358] manager: NetworkManager state is now ASLEEP
Feb 19 22:07:48 mochi NetworkManager[518]: <info>  [1708398468.4360] device (enp3s0): state change: activated -> deactivating (reason 'sleeping', sys-iface>
Feb 19 22:07:48 mochi kded5[752]: org.kde.plasma.nm.kded: Unhandled active connection state change:  3
Feb 19 22:07:48 mochi wpa_supplicant[578]: wlan0: CTRL-EVENT-DSCP-POLICY clear_all
Feb 19 22:07:48 mochi systemd[1]: Starting Network Manager Script Dispatcher Service...
Feb 19 22:07:48 mochi systemd[1]: Started Network Manager Script Dispatcher Service.
Feb 19 22:07:48 mochi NetworkManager[518]: <info>  [1708398468.4772] device (enp3s0): state change: deactivating -> disconnected (reason 'sleeping', sys-if>
Feb 19 22:07:48 mochi avahi-daemon[471]: Withdrawing address record for  on enp3s0.
Feb 19 22:07:48 mochi avahi-daemon[471]: Leaving mDNS multicast group on interface enp3s0.IPv6 with address .
Feb 19 22:07:48 mochi NetworkManager[518]: <info>  [1708398468.4774] dhcp4 (enp3s0): canceled DHCP transaction
Feb 19 22:07:48 mochi avahi-daemon[471]: Joining mDNS multicast group on interface enp3s0.IPv6 with address .
Feb 19 22:07:48 mochi NetworkManager[518]: <info>  [1708398468.4774] dhcp4 (enp3s0): activation: beginning transaction (timeout in 45 seconds)
Feb 19 22:07:48 mochi avahi-daemon[471]: Registering new address record for  on enp3s0.*.
Feb 19 22:07:48 mochi NetworkManager[518]: <info>  [1708398468.4774] dhcp4 (enp3s0): state changed no lease
Feb 19 22:07:48 mochi avahi-daemon[471]: Withdrawing address record for  on enp3s0.
Feb 19 22:07:48 mochi avahi-daemon[471]: Leaving mDNS multicast group on interface enp3s0.IPv6 with address .
Feb 19 22:07:48 mochi avahi-daemon[471]: Interface enp3s0.IPv6 no longer relevant for mDNS.
Feb 19 22:07:48 mochi avahi-daemon[471]: Withdrawing address record for  on enp3s0.
Feb 19 22:07:48 mochi avahi-daemon[471]: Leaving mDNS multicast group on interface enp3s0.IPv4 with address .
Feb 19 22:07:48 mochi avahi-daemon[471]: Interface enp3s0.IPv4 no longer relevant for mDNS.
Feb 19 22:07:48 mochi wpa_supplicant[578]: wlan0: CTRL-EVENT-DSCP-POLICY clear_all
Feb 19 22:07:48 mochi wpa_supplicant[578]: nl80211: deinit ifname=wlan0 disabled_11b_rates=0
Feb 19 22:07:48 mochi NetworkManager[518]: <info>  [1708398468.5321] device (enp3s0): state change: disconnected -> unmanaged (reason 'sleeping', sys-iface>
Feb 19 22:07:48 mochi kscreenlocker_greet[2561]: kf.kirigami: Failed to find a Kirigami platform plugin
Feb 19 22:07:48 mochi kscreenlocker_greet[2561]: qt.qpa.wayland: Wayland does not support QWindow::requestActivate()
Feb 19 22:07:48 mochi systemd[1]: Reached target Sleep.
Feb 19 22:07:48 mochi kscreenlocker_greet[2561]: qt.qpa.wayland: Wayland does not support QWindow::requestActivate()
Feb 19 22:07:48 mochi systemd[1]: Starting System Suspend...
Feb 19 22:07:48 mochi systemd-sleep[2625]: Performing sleep operation 'suspend'...
Feb 19 22:07:48 mochi kernel: PM: suspend entry (deep)
Feb 19 22:07:48 mochi kernel: Filesystems sync: 0.027 seconds

X11:
Feb 19 22:12:31 mochi systemd-logind[482]: The system will suspend now!
Feb 19 22:12:31 mochi NetworkManager[523]: <info>  [1708398751.9227] manager: sleep: sleep requested (sleeping: no  enabled: yes)
Feb 19 22:12:31 mochi NetworkManager[523]: <info>  [1708398751.9229] device (wlan0): state change: disconnected -> unmanaged (reason 'slee>
Feb 19 22:12:31 mochi NetworkManager[523]: <info>  [1708398751.9434] device (wlan0): set-hw-addr: reset MAC address to (>
Feb 19 22:12:31 mochi kernel: rtl8192cu: MAC auto ON okay!
Feb 19 22:12:31 mochi kernel: rtl8192cu: Tx queue select: 0x05
Feb 19 22:12:32 mochi kscreenlocker_greet[4753]: Qt: Session management error: networkIdsList argument is NULL
Feb 19 22:12:32 mochi NetworkManager[523]: <info>  [1708398752.4395] device (p2p-dev-wlan0): state change: disconnected -> unmanaged (reas>
Feb 19 22:12:32 mochi NetworkManager[523]: <info>  [1708398752.4397] manager: NetworkManager state is now ASLEEP
Feb 19 22:12:32 mochi NetworkManager[523]: <info>  [1708398752.4398] device (enp3s0): state change: activated -> deactivating (reason 'sle>
Feb 19 22:12:32 mochi kded5[3223]: org.kde.plasma.nm.kded: Unhandled active connection state change:  3
Feb 19 22:12:32 mochi wpa_supplicant[581]: wlan0: CTRL-EVENT-DSCP-POLICY clear_all
Feb 19 22:12:32 mochi systemd[1]: Starting Network Manager Script Dispatcher Service...
Feb 19 22:12:32 mochi systemd[1]: Started Network Manager Script Dispatcher Service.
Feb 19 22:12:32 mochi NetworkManager[523]: <info>  [1708398752.4805] device (enp3s0): state change: deactivating -> disconnected (reason '>
Feb 19 22:12:32 mochi avahi-daemon[477]: Withdrawing address record for  on enp3s0.
Feb 19 22:12:32 mochi avahi-daemon[477]: Leaving mDNS multicast group on interface enp3s0.IPv6 with address :>
Feb 19 22:12:32 mochi NetworkManager[523]: <info>  [1708398752.4807] dhcp4 (enp3s0): canceled DHCP transaction
Feb 19 22:12:32 mochi avahi-daemon[477]: Joining mDNS multicast group on interface enp3s0.IPv6 with address .
Feb 19 22:12:32 mochi NetworkManager[523]: <info>  [1708398752.4807] dhcp4 (enp3s0): activation: beginning transaction (timeout in 45 seco>
Feb 19 22:12:32 mochi avahi-daemon[477]: Registering new address record for  on enp3s0.*.
Feb 19 22:12:32 mochi NetworkManager[523]: <info>  [1708398752.4807] dhcp4 (enp3s0): state changed no lease
Feb 19 22:12:32 mochi avahi-daemon[477]: Withdrawing address record for  on enp3s0.
Feb 19 22:12:32 mochi avahi-daemon[477]: Leaving mDNS multicast group on interface enp3s0.IPv6 with address .
Feb 19 22:12:32 mochi avahi-daemon[477]: Interface enp3s0.IPv6 no longer relevant for mDNS.
Feb 19 22:12:32 mochi avahi-daemon[477]: Withdrawing address record for  on enp3s0.
Feb 19 22:12:32 mochi avahi-daemon[477]: Leaving mDNS multicast group on interface enp3s0.IPv4 with address .
Feb 19 22:12:32 mochi avahi-daemon[477]: Interface enp3s0.IPv4 no longer relevant for mDNS.
Feb 19 22:12:32 mochi wpa_supplicant[581]: wlan0: CTRL-EVENT-DSCP-POLICY clear_all
Feb 19 22:12:32 mochi wpa_supplicant[581]: nl80211: deinit ifname=wlan0 disabled_11b_rates=0
Feb 19 22:12:32 mochi NetworkManager[523]: <info>  [1708398752.5280] device (enp3s0): state change: disconnected -> unmanaged (reason 'sle>
Feb 19 22:12:32 mochi kscreenlocker_greet[4753]: kf.kirigami: Failed to find a Kirigami platform plugin
Feb 19 22:12:32 mochi kscreenlocker_greet[4753]: file:///usr/share/plasma/look-and-feel/org.kde.breeze.desktop/contents/components/Virtual>
Feb 19 22:12:32 mochi systemd[1]: Reached target Sleep.
Feb 19 22:12:32 mochi systemd[1]: Starting System Suspend...
Feb 19 22:12:32 mochi systemd-sleep[4803]: Performing sleep operation 'suspend'...
Feb 19 22:12:32 mochi kernel: PM: suspend entry (deep)
Feb 19 22:12:33 mochi kernel: Filesystems sync: 0.094 seconds

I don't know what sshd is but I can start looking into it.

edit: snipped out addresses. Not sure if that matters or not or if it's of concern, I'm still learning

Framboise fucked around with this message at 04:24 on Feb 20, 2024

# ? Feb 20, 2024 04:18

Klyith: Aug 3, 2007; GBS Pledge Week

Framboise posted:

Okay, here's what I have-- I got logs from testing the issue with both Wayland and X11, if it helps at all:

Yep, that looks like your system is going to sleep and the OS is never alive enough to log anything after that, so it's not just the display or something.

Framboise posted:

I don't know what sshd is but I can start looking into it.

sshd is the daemon (server) for ssh, ie remote terminal log in for your system.

I don't think that you need to bother -- that would be helpful if your system was still alive after sleep, with just the display or DE dead. But your situation would require you to ssh via ouija board.

What you may want to do is check what BIOS options your mobo / PC has around sleep & power, if they can be set to more generic or basic modes. <- rear end-pull guesses here

# ? Feb 20, 2024 05:18

Framboise: Sep 21, 2014; To make yourself feel better, you make it so you'll never give in to your forevers and live for always.; Lipstick Apathy

Klyith posted:

Yep, that looks like your system is going to sleep and the OS is never alive enough to log anything after that, so it's not just the display or something.

sshd is the daemon (server) for ssh, ie remote terminal log in for your system.

I don't think that you need to bother -- that would be helpful if your system was still alive after sleep, with just the display or DE dead. But your situation would require you to ssh via ouija board.

What you may want to do is check what BIOS options your mobo / PC has around sleep & power, if they can be set to more generic or basic modes. <- rear end-pull guesses here

Well, drat. That just makes things more complicated.

As far as my BIOS goes, I see a setting for "Suspend to RAM" that is set to "Auto", enabling ACPI S3 power saving (whatever that means), and a "Check Ready Bit" that's enabled, which says "Enable to enter the operating system after S3 only when the hard disk is ready. this is recommended for better system stability." And then a bunch of "allow PC to be woke by xyz entry methods" with USB keyboard and mouse enabled.

# ? Feb 20, 2024 06:24

zhar: May 3, 2019

If it was me I�d try disabling the check ready bit setting to start as I�ve never heard of such a thing. If that didn�t help I�d try nouveau drivers and then maybe try suspend from a Ubuntu livecd or similar to make sure it�s not some wonky config in endeavouros.

# ? Feb 20, 2024 10:01

cruft: Oct 25, 2007

Klyith posted:

the thing that's caused all my past data loss

The thing that's caused all my past data loss is me. So checkpoints are pretty great.

# ? Feb 21, 2024 02:30

Framboise: Sep 21, 2014; To make yourself feel better, you make it so you'll never give in to your forevers and live for always.; Lipstick Apathy

zhar posted:

If it was me I�d try disabling the check ready bit setting to start as I�ve never heard of such a thing. If that didn�t help I�d try nouveau drivers and then maybe try suspend from a Ubuntu livecd or similar to make sure it�s not some wonky config in endeavouros.

No luck on the check ready bit. I'll try to figure out how to set up the nouveau drivers next. However you do that.

# ? Feb 21, 2024 04:21

Klyith: Aug 3, 2007; GBS Pledge Week

Framboise posted:

No luck on the check ready bit. I'll try to figure out how to set up the nouveau drivers next. However you do that.

nouveau is built into the kernel so all you need to do is uninstall the proprietary nvidia drivers and you should go back to it

endeavour right? so looks like they have their own nvidia manager nvidia-inst, so running that with --nouveau will switch you back.

edit: uhhh also you did use that in the first place to get the nvidia proprietary drivers, right?

# ? Feb 21, 2024 04:29

Framboise: Sep 21, 2014; To make yourself feel better, you make it so you'll never give in to your forevers and live for always.; Lipstick Apathy

Klyith posted:

nouveau is built into the kernel so all you need to do is uninstall the proprietary nvidia drivers and you should go back to it

endeavour right? so looks like they have their own nvidia manager nvidia-inst, so running that with --nouveau will switch you back.

edit: uhhh also you did use that in the first place to get the nvidia proprietary drivers, right?

code:

01:00.0 VGA compatible controller: NVIDIA Corporation GM200 [GeForce GTX 980 Ti] (rev a1) (prog-if 00 [VGA controller])
        Subsystem: ASUSTeK Computer Inc. GM200 [GeForce GTX 980 Ti]
        Flags: bus master, fast devsel, latency 0, IRQ 48
        Memory at f6000000 (32-bit, non-prefetchable) [size=16M]
        Memory at e0000000 (64-bit, prefetchable) [size=256M]
        Memory at f0000000 (64-bit, prefetchable) [size=32M]
        I/O ports at e000 [size=128]
        Expansion ROM at 000c0000 [virtual] [disabled] [size=128K]
        Capabilities: <access denied>
        Kernel driver in use: nvidia
        Kernel modules: nouveau, nvidia_drm, nvidia

I didn't have to do anything beyond simply selecting the nvidia option when loading up the live environment/installer. I guess it kind of just figured it out from there. I'll try switching over to nouveau now, will report back momentarily.

# ? Feb 21, 2024 06:52

Framboise: Sep 21, 2014; To make yourself feel better, you make it so you'll never give in to your forevers and live for always.; Lipstick Apathy

I either hosed up something real bad, or it really doesn't like the nouveau drivers.

Back on Windows for now. Not really sure what to do beyond wiping it all out and starting fresh again, I'm not sure how to fix it otherwise.

I now understand just how much nvidia does not play nice with linux.

# ? Feb 21, 2024 07:10

Volguus: Mar 3, 2009

Framboise posted:

I either hosed up something real bad, or it really doesn't like the nouveau drivers.

Back on Windows for now. Not really sure what to do beyond wiping it all out and starting fresh again, I'm not sure how to fix it otherwise.

I now understand just how much nvidia does not play nice with linux.

You can boot with "nomodeset" kernel param to tell it to not load video drivers. Then from the command line you can fix whatever is broken. No need to go medieval on it and reinstall.

# ? Feb 21, 2024 14:23

Less Fat Luke: May 23, 2003; Exciting Lemon

That distribution choice seems pretty rough for a first timer, I'd suspect you'll have a much easier time with Ubuntu as it's still the gold-standard for "works out of the box".

# ? Feb 21, 2024 14:46

Klyith: Aug 3, 2007; GBS Pledge Week

Framboise posted:

I either hosed up something real bad, or it really doesn't like the nouveau drivers.

Back on Windows for now. Not really sure what to do beyond wiping it all out and starting fresh again, I'm not sure how to fix it otherwise.

I now understand just how much nvidia does not play nice with linux.

Also just how much Arch is a "learn the hard way" distro.

Volguus posted:

You can boot with "nomodeset" kernel param to tell it to not load video drivers. Then from the command line you can fix whatever is broken. No need to go medieval on it and reinstall.

Endeavour should have fallback options in grub to give you text only boot.

Also it's possible that hitting ctrl-alt-F# on that screen will get a usable terminal, because that mess is from the display server failing to start up?

# ? Feb 21, 2024 14:49

cruft: Oct 25, 2007

I'm just about to the point where I'm going to tell people that NVidia cards are not compatible with Linux. Not "compatible asterisk", just straight up incompatible.

What a nightmare.

# ? Feb 21, 2024 15:18

Mega Comrade: Apr 22, 2004; Listen buddy, we all got problems!

cruft posted:

I'm just about to the point where I'm going to tell people that NVidia cards are not compatible with Linux. What a nightmare.

Every person you see who says "I've got an Nvidia card and not had a problem" should just get "not yet" in reply.

Hopefully once the new FOSS drivers have matured this will be a thing of the past, but right now I don't even consider Nvidia GPUs when looking at cards.

# ? Feb 21, 2024 15:20

ziasquinn: Jan 1, 2006; Fallen Rib

honestly after spending 5 hrs troubleshooting an audio issue in arch once that a wipe and pave resolved, I set a hard limit of <=3 hr troubleshooting Linux now before "going medieval" especially if I have /home on its own partition.

sometimes it's worth it to learn how but sometimes it ain't

# ? Feb 21, 2024 15:25

Volguus: Mar 3, 2009

Mega Comrade posted:

Every person you see who says "I've got an Nvidia card and not had a problem" should just get "not yet" in reply.

Hopefully once the new FOSS drivers have matured this will be a thing of the past, but right now I don't even consider Nvidia GPUs when looking at cards.

Well, I've got a nvidia card and not had a problem. For more than 20 years now. "Not yet" - should be due any day.

Volguus fucked around with this message at 15:33 on Feb 21, 2024

# ? Feb 21, 2024 15:26

VictualSquid: Feb 29, 2012; Gently enveloping the target with indiscriminate love.

ziasquinn posted:

honestly after spending 5 hrs troubleshooting an audio issue in arch once that a wipe and pave resolved, I set a hard limit of <=3 hr troubleshooting Linux now before "going medieval" especially if I have /home on its own partition.

sometimes it's worth it to learn how but sometimes it ain't

I am surprised that worked. Aren't most audio configs in /home these days?

# ? Feb 21, 2024 15:28

mawarannahr: May 21, 2019

Mega Comrade posted:

Every person you see who says "I've got an Nvidia card and not had a problem" should just get "not yet" in reply.

Hopefully once the new FOSS drivers have matured this will be a thing of the past, but right now I don't even consider Nvidia GPUs when looking at cards.

Hope you don't need cuda lmao

# ? Feb 21, 2024 15:29

Less Fat Luke: May 23, 2003; Exciting Lemon

Volguus posted:

Well, I've got a nvidia card and not had a problem. For more than 20 years now. "Not yet" - would be due any day now.

Yeah same under Debian primarily; I need to use it for CUDA stuff but I also play a lot of games with the binary drivers on Debian, no real issues other than the usual Linux bullshit occasionally. That being said I would buy anything else if I didn't need it for work since their GPUs are so goddamned expensive.

# ? Feb 21, 2024 15:33

xzzy: Mar 5, 2009

cruft posted:

I'm just about to the point where I'm going to tell people that NVidia cards are not compatible with Linux. Not "compatible asterisk", just straight up incompatible.

What a nightmare.

We just evaluated a "gpu accelerated software raid card" that requires an nvidia gpu in the system to work. So now not only can driver problems break your display, it can break access to your data!

https://www.graidtech.com/product/sr-1010/

# ? Feb 21, 2024 15:34

Adbot: ADBOT LOVES YOU

# ? May 31, 2024 17:29

Less Fat Luke: May 23, 2003; Exciting Lemon

What in the gently caress

# ? Feb 21, 2024 15:35

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > The Linux Questions Thread: a bunch of pitfalls, but technically it's possible

«‹›1037 »