You are not logged in.
Ceres (Sid) never freezes, but the rate of new packages slows down right before a release. That's because everyone is working on finishing the next release. When it's released, then people start working on new packages again and the unstable branch gets faster changes.
apt remove gtk3-nocsd libgtk3-nocsd
Tested on desktop-live iso.
You have to remove libgtk3-nocsd. Just getting rid of the main package doesn't fix it.
For zoom, I use chromium browser. If you have a link to a zoom meeting, ignore all the suggestions to install zoom and scroll to the bottom of the page to enter the meeting in your web browser.
For skype, you need to use a third-party package (thanks, microsoft!). Use the package for the corresponding version of debian. (bullseye = chimaera, bookworm = daedalus)
Is skype not installable in gnuinos? I have no idea how the libre kernel handles that.
This one was locally brewed - Simple Update Notifier
https://dev1galaxy.org/viewtopic.php?pid=32990#p32990
I didn't mention it, but another option is to use one of the installer isos instead of a live iso. The installer isos use the debian-installer (modified for devuan) that you may be more accustomed to. That one will let you select a partition for home and not format it, so you can re-use your files.
Either way, you should have all your files backed up somewhere before installing.
IMPORTANT!!!
The live installer does not know how to re-use an existing /home partition. You have to arrange it manually.
If you select an existing /home partition with the live installer, it will be formatted and you will lose your data.
(Exception is if you set the installer not to format any partitions, and you format all of them in advance, but even then, some of your desktop configs may get clobbered with the new versions.)
Install without choosing a separate partition for /home.
Then rename the new /home directory.
Make a new /home directory.
Edit /etc/fstab to mount the old home partition to the new /home directory.
You can do the above tasks while still in the live session or after rebooting into single user mode. If you need more specific directions, let us know what your partition layout looks like so we can use the correct device names.
If you installed from a live-iso, see if this file exists:
ls -l /usr/sbin/anacron.orig.anacron
If it does, run the following (as root):
rm /usr/sbin/anacron
dpkg-divert --rename --remove /usr/sbin/anacron
if it is possible to configure package management so as it first pulls packages requested by user from the install cd(s) and then only if not present or outdated downloads data from the central repository (mirrors)
It might be possible to do that by pinning online sources to lower priority than the file or dvd source. Then you would need to specify the online source with the -t option to get packages from there.
As time goes on, any packages you pull from online sources will be more likely to need newer versions of associated packages that are in the dvd source, so that the package in the dvd won't be installable.
I think another way to do it is to mount the iso file and add it as a cdrom source.
So everyone who uses Devuan gets new updates at least two months later than the "slowest" distribution.
No, that's not the case at all. Devuan users have to wait a few hours after debian posts updates. The actual time it takes for the mirrors to update varies, but I think the longest is four hours.
The part that might not be ready is the installer isos and live isos. Those have to be put together and all the pieces might not be done today. I think we finished writing the release docs, but I'm not certain. Daedalus itself is ready. You can take your chances installing it from one of the new isos that might still have some bugs, or you can do a minimal install of chimaera and upgrade to daedalus, or if you already have chimaera, you can upgrade to daedalus.
If you use one of the new isos and find some bugs, you could let us know so we can fix it.
Did you boot legacy bios or uefi? If you did boot legacy bios, did you create a bios_grub partition?
I just tried expert install with the same iso and ran into some problems. I used gpt partition table but did not create the bios_grub partition. The installer didn't warn me about it, but I did get the window that asks if I want to install grub.
I answered No and it proceeded to ask me where I wanted grub installed. I chose "Select location" and didn't put in any location. It tried to install grub to nowhere. (Says so in the log)
When the window came up asking me to finish, I selected Go Back. The screen flashed a few times and it came back to the finish window. I tried to go back to the menu several times, but eventually I had to choose Finish the Install.
Yeah, I'm pretty sure you're right about the delay. You can also run into problems if the server goes down while the client still has the share mounted. (I just did that to myself about an hour ago.)
I find it better to use autofs, which will mount the share when you try to use it.
bash: dpkg-reconfigure: kommando ikke fundet
Use 'su -' instead of just 'su' to become root. That way you will get root's $PATH.
The other way to fix this is to use the full path to the command...
/usr/sbin/dpkg-reconfigure...
See 'man su' for yet another solution.
In the installed system, you should be able to set language and keyboard by running (as root or with sudo)
dpkg-reconfigure locales
dpkg-reconfigure keyboard-configuration
You can configure more than one language and keyboard and choose which one is the default.
If you want to change the language and keyboard when you boot a live-CD or live-USB, edit the boot command to add
locales=de_DE.UTF-8 keyboard-layouts=de
Note: To access the boot command at the boot menu, press TAB if you booted legacy bios or press e if you booted UEFI.
root@FM2:~# mkfs.ext2 -c /dev/sdb
Is this the command you used, or did you try writing to /dev/sdb1 (a partition) instead of the whole device?
It looks like you did things in the right order. When I do installs like that, it sometimes takes me a few tries to get it right.
Check /etc/fstab, /etc/crypttab, /etc/mdadm/mdadm.conf and /boot/grub/grub.cfg
I think something is aiming for the wrong target.
The simple answer is no, you'll have to make the swap file after the installation is finished.
The more complicated answer is that you can probably drop to a shell some time late in the install and add the swapfile manually.
If you want encryption and you want a swap partition, it's easier to do it with encrypted lvm. Then the OS partition(s) and swap will all be part of the encrypted volume. If you just want a single partition, you could use a swap file.
As an example, I use a 256MB swapfile on a laptop. None of that is being used at the moment. On my desktop machine I have a 1GB swapfile, and 16 MB of that are in use. On a server that runs radio automation software (constantly playing music from a playlist) that has LVM and RAID, the 256MB swapfile is about half full. These machines have 6 or 8 GB RAM.
Note: Encryption only works when the computer is turned off. It's useful in case someone steals your computer or you have to send the hard drive out for replacement or repair.
If you're worried about writes to the ssd, you could put the swap on one or both of the spinning disks. You would need to make the RAID from partitions instead of the whole drive, to leave room for a swap partition. Oh, I suppose you could put a swapfile inside the RAID. That might be easier. Then you don't need to make the extra partition.
In debian and devuan, runit will use init scripts if there are no run scripts for that service. Make sure libvirt-daemon-system-sysv is installed so you have those init scripts.
This one is a few days old. It's for ceres, but there should be a screen early in the install that asks what suite you want. You might need to choose expert install to get that question.
https://pkgmaster.devuan.org/devuan/dis … s/netboot/
.
Make sure policykit-1-gnome is installed. If that doesn't do it, check some other discussions here about the same thing. If all the installs are supposed to be the same, compare the package lists. This will create a package list:
dpkg -l | awk '/^ii/ { print $2 }' > package-list
What if someone wants luit and xorg ?
Then they might need to avoid metapackages that pull in things that might conflict with other things they want.
For better granularity in package selection, un-check everything at the tasksel window except maybe Standard System Utilities. You probably want those. Then add what you want after you reboot into the new system.
Use the option "--no-install-recommends" when you install packages. If that will exclude something you want, then add that package to your list of things to install. You can use "--simulate" with apt-get apt or aptitude to check what will happen before you actually do it.
Here's a list of what task-console-productivity pulls in.
apt show task-console-productivity
<snip>
Recommends: dialog, zsh, entr, gddrescue, gdisk, htop, iftop, iotop, iw, mtr-tiny, multitail, ncdu, parted, pciutils, psmisc, sudo, time, wavemon, wireless-tools, wpasupplicant, ed, zile, bind9-host, bittornado, curl, dnsutils, edbrowse, fetchmail, ftp, geoip-bin, irssi, lftp, links2, lrzsz, mcabber, minicom, mosh, msmtp, mutt, netcat, net-tools, nfacct, nrss, openssh-client, openssh-server, procmail, rsync, telnet, tin, traceroute, w3m, wget, whois, abook, apcalc, aspell, aspell-en, calcurse, clex, dvtm, fbi, fbterm, ghostscript, gnupg, gnupg2, gnupg-agent, mc, parallel, poppler-utils, rpl, rename, sc, screen, taskwarrior, tmux, bastet, bombardier, bsdgames, cavezofphear, crawl, curseofwar, empire, freesweep, gnuchess, greed, matanza, moria, nethack-console, ninvaders, omega-rpg, pacman4console, pente, sudoku, beep, brltty, espeak, espeakup, yasr, alsa-utils, caca-utils, hasciicam, imagemagick, jhead, moc, radio, sox, cmatrix, cowsay, eject, figlet, fortunes-min, fortune-mod, gpm, man-db, manpages, manpages-dev, mlocate, termsaver, toilet, toilet-fonts, ttyrec, unzip
'apt show luit' give me this:
Breaks: x11-utils (<< 7.7+6)
Replaces: x11-utils (<< 7.7+6)
which suggests that luit might work with newer x11-utils than what is in ceres/daedalus (7.7+5) so maybe you can't have both luit and xorg or maybe a newer xorg is coming at the last hour before bookworm release. Have you looked for any debian bug reports about this? That might tell you if they have plans to fix it.
If you have an installation with an encrypted root partition, then os-prober will not find it. Either you have to let the encrypted system be the one in charge of boot, or you have to create a boot menu entry in /etc/grub.d/40_custom for the encrypted system. Run update-grub to generate the new menu.
I don't remember what refind does with encrypted systems. And it might depend on whether /boot is encrypted or not.
One way to do it would be to install the second system without installing a bootloader and then run 'update-grub' in the first system. The new installation will be added (if os-prober is installed and enabled in /etc/default/grub).
If you want the second system to be the one in charge of booting, you can give it a different name. Boot into second system and run
grub-install --bootloader-id=somecoolname
update-grub
and then the boot files will be in /boot/efi/EFI/somecoolname
Edit: If your computer's uefi works like mine, the new one will be set to boot first.
And, if your computer's uefi works the way it's supposed to work, you'll also be able to change the boot order with efibootmgr.
9.6.5 was replaced with 9.6.6 that has a better fix. Those links above to 9.6.5 on sourceforge will still work if you click on the link for "Problems downloading?" and it will link you to the 9.6.6 debs. Or, you could just change the link like this...
https://sourceforge.net/projects/refrac … b/download
https://sourceforge.net/projects/refrac … b/download
Also, there are new Chimaera (Stable) and Daedalus (Testing) live-isos with the patched installer, available at your favorite mirror:
https://www.devuan.org/get-devuan
.
This looks like what I get on one laptop when I run the debian/devuan installer. You might be able to specify a screen resolution at the boot screen. If you're booting uefi, you can edit the selected entry in the boot menu by pressing e and then adding a line before the linux line that says set gfxpayload=<some resolution that works> then press ctrl-x to boot.
If you're booting legacy bios, you get an isolinux boot menu and you can press TAB to edit the boot command and add vga=<some vga mode that works>
Maybe check what the screen resolution is when you're running the desktop-live system and use that.