You are not logged in.
HevyDevy wrote:i dont use elogind or consolekit, ive been using devuan with no dbus and it runs great for what i need.
That seems very interesting to me. Could you, please, elaborate on it? May be a link or a new thread? Thank you.
I don't know what HevyDevy is using, but I made some nodbus live isos to see how much I could install. It wasn't difficult, and I got a fairly complete set of desktop apps. You can read about it here:
https://dev1galaxy.org/viewtopic.php?id=2158
Newer isos than the link in that thread are here.
https://get.refracta.org/files/experimental/
(refracta9 is ascii, refracta10 is beowulf.)
refracta2usb definitely doesn't solve the encrypted suspend problem. I suppose you could unmount the loopback file that has the encrypted filesystem and then suspend, but I've never tried. I just turn it off if I need it to be secure.
The other problem r2u doesn't solve is the user ID when moving home across different machines. It's meant for a live-usb, so you just take your own OS with you.
The only problem I might have with homed is if it gets in the way of doing things differently. I guess we'll see...
It doesn't look like we have a pinebook image here: https://files.roundr.devuan.org/devuan_ascii/embedded/
It might be possible to install one of the other images and then migrate to devuan, but I don't see plain vanilla debian in the list that you linked. So a migration from something else to devuan might be difficult.
If you have a pinebook to play with, you have an opportunity to be the first one to make a devuan image for it.
Boots for me in qemu:
qemu-system-x86_64 -enable-kvm -m 512 -smp cores=2,threads=2 -cdrom KNOPPIX_V8.6-2019-08-08-EN.iso
Boots fast.
Um, we did that. It's in /etc/os-release. ID=debian magically makes third-party software work with devuan. I think there's another reason for it, but I don't remember what it is. What I don't get is why /etc/lsb-release doesn't override it like it used to.
" withouth the need of deinstalling firefox 60 esr"
Hm... that's a good to know. Now I can have both versions installed without conflict because the older version is outside the apt system.
You might run into conflicts with extensions. Maybe there's a way to set the second firefox to use something other than ~/.mozilla to store its files.
Whoa! OK, so gdm is going to manage access to this encrypted home, user records will be json files to be more compatible with the internet (wtf?), and when you convert the json to a more traditional format, some metadata will be lost. (i.e. you won't see all that's in the user record.)
What does "Queriable via Varlink interface." mean? Is it time to put on my metal hat?
Lennart actually describes it as home-on-a-stick in the video presentation, which amused me.
Me too. Been doing luks-encrypted home on a stick for years, with the additional feature of a read-only operating system. https://refracta.org/docs/readme.refracta2usb.txt
I don't have interchangeable UID's or automatic close-luks-on-suspend. You'll have to turn the laptop off when you go through customs.
Let me steer this into a more general devuan topic.
When you need to roll back an update or install an older version of a package, how would you go about it?
Find the older package in /var/cache/apt/archives and install it wtih
dpkg --force-downgrade -i <package>I don't mean /root/.cache but /.cache
a folder is created at boot time
Oh! I checked two asciis, one jessie and a beowulf. I don't have one, and I don't recall ever seeing that.
Gave up waiting for suspend/resume device
This one can often be fixed by adding RESUME=none to /etc/initramfs-tools/initramfs.conf and then running update-initramfs -u or my manually unpacking the initrd and removing the resume file before repacking it. I suppose it would also be valid to put the correct swap uuid in the resume file if you're planning to hibernate.
Gave up waiting for root file system device
I don't know an initrd edit for this. You could try adding rootdelay=10 to the linux line in grub.cfg as suggested in the screen. If you know your way around grub command line, you could muck around there and see if you can find the root device yourself.
It gets created when you run a root application on the user's desktop. My /root/.cache has subdirectories named after the apps I've run as root.
Does your laptop normally boot in uefi mode, like the one in the article you posted? Have you tried installing with gpt and uefi? Do you know if it needs a 32-bit bootloader to do that? If so, there's a way to use the amd64 desktop-live iso to install on such a system.*
It sounds like Ralph identified the problem. Get rid of that erroneous uuid, and it should work. (Run blkid to prove that the uuid doesn't exist, if you haven't already done that.)
* You would install the grub-efi-ia32 package that's in the root of the live system before you run the installer. When it asks about installing the bootloader, say No. Then run the installer, and when that asks about bootloader, let it install.
This was tested at least one time, on a macbook pro, by someone other than me.
_
x-terminal-emulator is set in the alternatives system. That will be a symlink to the actual terminal, which is probably the terminal that came with your desktop or possibly the last terminal you installed.
ls -l /etc/alternatives/x-terminal-emulator
The man page for x-terminal-emulator will be the man page for the actual terminal. On one of my systems, it shows me the man page for lxterminal, and on another it's xfce4-terminal. That means the available options for x-terminal-emulator will not be the same for everyone. If the hold is needed (I suspect it is, but I'm not certain) we can make a function to test which terminal is used and which is the appropriate 'hold' option.
I think it's perfectly reasonable to leave out the terminal. Anyone who prefers upgrading in a terminal already knows how to open one and get root.
That one gives me two numbers, 13 and 12, which corrspond to:
↳ Microsoft Wired Keyboard 600 id=13 [slave pointer (2)]
↳ Microsoft Wired Keyboard 600 id=12 [slave keyboard (3)]The first line with XIKeyClass on my desktop is my mouse. On the laptop, it's Virtual Core Keyboard.
⎜ ↳ Logitech USB Receiver id=11 [slave pointer (2)]
Reporting 6 classes:
Class originated from: 11. Type: XIButtonClass
Buttons supported: 7
Button labels: "Button 0" "Button Unknown" "Button Unknown" "Button Wheel Up" "Button Wheel Do
wn" "Button Horiz Wheel Left" "Button Horiz Wheel Right"
Button state:
Class originated from: 11. Type: XIKeyClassYou can't find 60.2 because it's not there, and it's been gone for some time.
Run:
apt updateand try again. You should see 60.9.0esr-1~deb9u1
not quite. That way leaves white space at the end. Maybe add another sed that removes the white space. (not sure if it's a tab or several spaces.)
$ echo "$keyboard_id"
10
$echo "$keyboard_id"x
10 xMaybe more creative grepping will do it. Hm... not so easy. I think I can do it, but I'd need to see what your xinput shows and which line you want.
Assuming I want the one that says "keyboard" but not "Virtual core" on the left side, and says "slave keyboard" on the right side, this works on two of my machines. (One is "Microsoft Wired Keyboard 600" and the other is "AT Translated set 2 keyboard" on a Thinkpad.)
keyboard_id=$(xinput --list | grep 'slave keyboard' | sed -e 's/\[.*//g' | grep -i keyboard | grep -v core | cut -d"=" -f2)setxkbmap
Thanks! I forgot about that one. It works instantly.
Also, I mis-reported one thing above. After killing the desktop with alt-sysrq-k, the change in keyboard-configuration took effect, and I got the us keyboard.
I didn't get to try the installer but hope to do that some time. Too much other stuff to do now. If anyone else tries it, please let me know.
You don't need to install unattended-upgrades to use the periodic update function. Just adding 10periodic to /etc/apt/apt.conf.d/ is enough. The script that runs it is in /etc/cron.daily and runs once a day. In my case, it's running around 6:30am. The actual time it updates is randomized, so that everyone is not hitting on the servers at the same time. In other words, if you use a cron job with a set time, a million miyolinux users will burn out the server on the next update.
It logs in /var/log/syslog. Look for 'cron.daily' to see when it ran.
Oh, I like the @reboot trick. That might be especially useful on a laptop that is asleep at 6:30am. Of course, the usual way to run stuff at boot is to put it in rc.local.
I tried the avmultimini.iso. Thanks for putting that up. Here are some notes you might find useful.
I can get it to boot in qemu (with kvm) in legacy bios mode with 'eno' boot code. If I try uefi mode, it hangs on the archivista splash screen. The vertical size of the window increases to 2-3 times normal height, and I have to resize the window to see the bottom of the screen. After a few minutes of this, the cpu starts to get hot.
I can't get it to boot in virtualbox. Tried uefi and legacy bios.
It boots normally from usb. Booted it in legacy bios mode with 'eno' boot code.
The language was English except for kodi, which came up in German.
Changing the keyboard layout presented some problems. First I tried 'dpkg-reconfigure keyboard-configuration'. I wanted to log out and in again to get it to take effect, but there's no way to log out of the desktop. I tried 'init 1' and nothing happened. I finally tried alt-sysrq-k and was able to restart the desktop. Keyboard was still German. This was especially frustrating for me because it made it impossible to do some things in the terminal.
Then I went into the desktop settings and found the keyboard layout settings. The drop-down menu for countries goes off-screen, and I can't scroll down.
I did not have any problems with the display orientation. The buttons on the panel work and are very cool. That's the first time I've ever been able to rotate the screen. (I don't need it, but I have tried it out of curiosity. Nice to see it finally work.)
It was also nice to see the pulseaudio volume control work. I haven't had good luck with that in the past.
Getting it to run in virtualbox or qemu is probably not essential to using it, but a lot of people like to test new isos in a virtual environment, for convenience.
Here's an example from 'man rsync showing how to keep one file and remove everything around and above it'
+ /some/
+ /some/path/
+ /some/path/this-file-is-found
+ /file-also-included
- *I think you can do it like this:
+ /home/*/.moonchild_productions/
+ /home/*/.moonchild_productions/pale_moon/
+ /home/*/.moonchild_productions/pale_moon/*.default/
+ /home/*/.moonchild_productions/pale_moon/*.default/prefs.js
- /home/*/.moonchild_productionsOr, you could probably do what the snapshot excludes file does with ~/.mozilla - exclude Cache and the sqlite files, and let everything else stay.
- /home/*/.mozilla/*/Cache/*
- /home/*/.mozilla/*/urlclassifier3.sqlite
- /home/*/.mozilla/*/places.sqlite
- /home/*/.mozilla/*/cookies.sqlite
- /home/*/.mozilla/*/signons.sqlite
- /home/*/.mozilla/*/formhistory.sqlite
- /home/*/.mozilla/*/downloads.sqliteI'm more stumped by the update-sync script.
Me too. It shouldn't work unless you're running it as root. When I split that up into two functions, I had it record the exit code of 'apt-get update', and while it kept running at the expected interval, it spit out an error code every time.
Also stumped by the inability to find sudo. Maybe the quoting needs to be changed? "sudo $command" instead of 'sudo "$command"' with or without the full path. (grasping at straws here.)
Maybe xterm needs the full path. Change 'sudo "$command"' to '/usr/bin/sudo "$command"' in all three xterm lines.