You are not logged in.
Short update: the 6.1.1 netinstall in a VM led to a working cli installation with a similar setup as yours.
Please tell us your work sequence and where the problem finally appears.
Strange. Have you ticked the format option in the partitioner? I am without any clue what maybe wrong, at least without a better idea about what you have entered. Can you post your settings line by line please? When exactly does this message appear? When leaving the partitioner?
What kind of HW are we talking about? Brand of the laptop and the NVME would be interesting.
I haven't used the 6.1.1 netinstall myself yet, maybe I give it a try in a VM when I have a bit of time.
Well, that one thing I forgot to mention. I am normally putting that into an encrypted LVM together with /.
If you want to use hibernation, make it a bit bigger that the size of the RAM. In old times the saying was twice the RAM, nowadays 1,1 x size of RAM is fine. Your best option is to delete the ext4 partition, first of all create the swap, and then use the rest for /.
Strangely enough, when I go into BIOS the USB shows up twice but booting the first one brings up the installer (haven't tried the second).
UEFI boards do that. If you boot the one without a note/prefix/postfix saying its efi or uefi then you boot the installer in legacy mode. And if you have a brand new laptop it will be uefi only.
Make sure you boot the install stick in UEFI mode.
to my understanding I should be all clear to fully wipe the drive and try again?
The easiest way is to use the installer during partitioning, and to write a new GPT to the drive. That makes sure the old stuff is deleted. There is nothing in the bios that could do that.
Then you first create a 1024MiB large partition, assign it to be ESP, and the rest is done by the installer.
Next you create a whatever you like large partition for /, ext4 as file system and mount point /.
Take a read on the webpage I already mentioned, there is a installtion instruction with pictures. But as far as I know it just describes the auto mode. Manual partitioning isn't mentioned in there. Its a bit arkward.
Hi there. I do not get the full picture yet, your post is a bit confused.
I had Windows 11 up and running up till this point.
What is your goal? Replace Win11 completely or a parallel installation to Win11? The steps to take are different. Before going into details, we should clarify this topic.
Before installing, I tinkered with the BIOS a bit. I disabled and cleared the TPM chip and disabled the "Absolute" tracking system.
TPM isn't relevant for the installation of Excalibur.
Personally, I would go and switch off Secure Boot. On some PCs this is called "Other OS". And I delete all SB keys. Don't need them.
What is this "the "Absolute" tracking system"?
Each time I tried to install I was warned that the intel/sof/sof-tgl.ri firmware was missing.
This is Audio stuff, you can ignore it until Excalibur is up and running.
The first time through I didn't even manage to get past NW configuration. I had to try again with an Ethernet connection.
You may install w/o network. I often do.
Setting up wifi can be very challening in the installer, depending on the HW used. The better solution: Ethernet normally is no big problem, except for on some exotic chipsets.
I had the option to try installing the GRUB boot loader again, but it once again failed.
Sounds like there is an issue with getting the efi grub installed. That can have several reasons. Maybe a confusion between legacy/efi install?
Some remarks:
Your PC must be an modern UEFI capable system, that means only an installation in efi mode makes sense. Make sure that you have disabled CSM in the bios.
When booting the install media, make sure you boot it in efi mode. The bios should show a prefix like efi or uefi before the name of the device.
For an efi installation you need a FAT32 formatted ESP partition. When you do manual partitioning, the ESP has to be configured in the partitioner as ESP, it does not need to be partitioned. Win leaves a relatively small ESP of a bit more than 100MB. When you do a clean Devuan only install, spend 1024 MiB.
It should be no deal to use /dev/nvme0n1p2 as / with ext4 file system. When its empty, you can have it formatted by the installer without data loss. Just the assignment has to be correct.
Good luck.
PS: Here you can find some reading: https://www.devuan.org/os/install
Well, it depends on the kernel you use. No issue with 6.12, starts with 6.13 to I think 6.18 (not 100% sure, I am still on 6.12 and there it still works). Then its fixed again.
Strangely enough tools like gvfs do their job as intended. I need it on my Cinnamon to connect to my smartphone every now and then.
That is more than a little sickly for folks like me that tend to stay away from bloated products like gvfs (wiki) virtual filesystem software for the GNOME desktop environment
So what is so "bloated" about gvfs? Its from Gnome, ok. But its not the desktop. I recently looked at Gnome again because some people were raving about the desktop, but its not my beer.
My dear friends, I think your are on the wrong track. I am not talking about the file system trash can but the one built in into Evolution. Its a folder like the Inbox, Archive, Send Messages, .... Similar to the setup in Thunderbird and other mail clients. Was my description that bad? In any case, thank your for your feedback.
I am aware of the built in functions "Save/Restore Evolution Data" or how its called in the English menu (I am from Germany so my menu is in German). Have used it many times to transport the Evolution settings and mails to a new release or a different computer over the years. And I am running a backup of my data with rsync normally. Please don't worry, I am no beginner.
also evolution has a "remove duplicate messages" function as well(you mentioned "repeating mails" but i suppose that isn't the same as "duplicates" so that probably won't work for you)
I am sorry, I do not mean duplicates. I am talking about something like the weekly mailing from one of my email providers who spams his customers with a "magazine" I do filter out to the trash can without looking at it. Repeating messages of similar structure and slightly differing content. Or shippig information from ... whatever.
As of today I have not found a way to partially clean out am single one of the more than 10000 deleted mails in the Evolution trash can like I could do with Outlook. I miss this function.
I am using Evolution as mail client. Many years of mails, and the trash can is filled with a lot of stuff. I am keeping certain stuff in there in case I deleted it by mistake. Instead of completely deleting the trash can I would like to selectively remove certain repeating mails. Is there any chance I can do that?
I am coming from my business experience with Outlook. There this was possible.
I guess the problem is that you are mixing several approaches, and they don't go together very well. I have taken my setup from here: https://dev1galaxy.org/viewtopic.php?pid=45747#p45747. This is the initial Daedalus approach. The same thread then talks about different approaches and daemons ... never fully understood the differences, and never changed my setup. It works 18 months now.
I have simply upgraded my Daedalus installation to Excalibur, ignored the release notes, and the old stuff simply works. Also on my travelling laptop.
You followed a different guide, and I guess you have to undo that stuff first of all. Sorry, I should have mentioned that.
I got the same snippet in front of my .xsessionrc:
if [ -f ~/.profile ]; then
. ~/.profile
fiBut according to my notes this was because of some conf change for gnupg. I don't think this hurts.
I would remove or comment out the addition from the release note in .profile. Give it a try.
Welcome to dev1galaxy.
What worked for me on my Excalibur with Cinnamon on sysvinit, basically the same setup you have:
apt install pipewire-audioEdit the users .xsessionrc and add:
# kill any existing pipewire instance to restore sound
pkill -u "$USER" -fx /usr/bin/pipewire-pulse 1>/dev/null 2>&1
pkill -u "$USER" -fx /usr/bin/wireplumber 1>/dev/null 2>&1
pkill -u "$USER" -fx /usr/bin/pipewire 1>/dev/null 2>&1
exec /usr/bin/pipewire &
# wait for pipewire to start before attempting to start related daemons
while [ "$(pgrep -f /usr/bin/pipewire)" = "" ] ; do
sleep 1
done
exec /usr/bin/wireplumber &
exec /usr/bin/pipewire-pulse & The file is a config file, does not need to be executable. Octal rights 644.
If you have more than one user you have to repeat that for these too.
I re-installed Devuan 6.1.1 (Excalibur). During installation, I supplied only a User Name and User Password. (I bypassed the prompt to enter a Root Password.)
Good to know that you have deciuded to follow the sudo path. Important for further support.
Opening the Dolphin file manager, I found that I needed only to click my partitions to mount them ...
My guess would be that these partitions have been mounted already during system boot, and you are just opening them with Dolphin. Without looking more into the details, we can't say for sure, but the inxi output is enough for a qualified guess.
From there I can read that you have / on sda11, /boot/efi on sda1, swap on sda10 and /home on sda9. These partitions can be accessed by Dolphin w/o any additional mount because they are listed in the fstab which defines which partitions will be mounted during system boot (swap will remain hidden). To confirm please execute:
cat /etc/fstab... -- except for just one partition, Media1 (sda7). After clicking on it, I got the following message:
Authentication is required to mount WDC WD1003FZEX-00MK2A0 (dev/sda7).
I had to enter my user account password to mount Media1.
I opened the Gnome Disk Utility, deleted the entire sda7 partition (Media1) and reformatted it as EXT4. After restarting, I opened Dolphin and tried to remount Media1 again by clicking on it. I got the same message: Authentication is Required. I had to enter my account password to mount it.
Looks very normal to me. sda7 seems not to be listed in the fstab since its not listed in the inxi report, so when you want to mount it you must enter your admin pw for elevated privilegues for this activity. This requirement is part of the Linux safety philosophy.
You can't bypass this procedure, except for when you follow pcalvert's advice: add sda7 to the fstab. Then its mounted during bootup and everything required is handled in the background.
The only exception to this that I have seen (so far) is when I use a file manager to mount a partition, formatted as VFAT or exFAT, that is located on a USB flash drive.
All major DEs like Cinnamon, Gnome, KDE, LXQT, Mate ... have so called automounters on board that take care for removable drives. These automounters handle the elevated privilegues in the background and supply the drives mostly in /media/$USERNAME/drive with access for the user. This is also valid for e.g. USB sticks or drives that are extx formatted or usb disk drives with any file system or encryption.
... has additional EXT4 partitions that are non-Linux and had nothing to do with the OS install.
ext4 is Linux in any case. For the second part: as long as this/these partitions are not mounted, nobody sees them. No user, no root, no installer. No matter who the owner is.
Each time that I boot up (or restart), Devuan demands that I enter the Root SUDO password to authorize the mounting of the Media1 partition (which is sd7). When I enter the password, Media1 gets mounted, and I am logged onto a desktop that the ROOT created.
First of all, you have not mentioned anything about what you have installed. Which release? Which desktop? You mention something about KDE later on. Would you mind to share the output from
inxi -FzrYou may need to install inxi first.
Assuming you have not setup something like autologin, you should be asked for Username and password and not asked for any root/sudo password during booting and land into any root account.
Another fact: Root SUDO are not the same, they a two separate approaches to gain admin rights. We are here on a Debian based distro, not Ubuntu or like. The installer asks you if you want to assign a root password. If you say no, sudo gets installed and configured for your username (the Ubuntu setup). For admin actions you are asked for the user pw. If you say yes, the root account is fully generated and no sudo is setup at all. For admin actions you need to enter the root password.
Basically I can't follow your complaints and make any sense out of your problems. Why do you automount your media1 partition? What happens if you simply abort the mount request?
BTW: mounting is a admin action and requires root privilegues. Always.
Is it possible that Devuan somehow got the bootable UBS Install drive confused with the Install ISO on Media1?
Simple answer: NO
I hope my comments do help you to clear up some mist.
First of all: welcome to the Dev1Galaxy.
Other distros, like Linux Mint, require that you use FlatPak. Using something else appears to cause problems.
I am sorry, your statement definitely isn't close to correct. Mint, no matter if native or the LMDE variant, can fully be used w/o any flatpack. They have their "Anwendungsverwaltung" (don't know the English term) for installation of packages, they have ditched Synaptics from the default installation against a new tooling, but you can still install and use it. You also may use the console and e.g. apt.
You may install flatpacks, but that is an exception but no way the rule.
Here under Debian/Devuan Synaptics is fine, using the console is fine too.
Discover as a KDE app seems to habe problems every now and then. I have seen one or the other thread about Discover not connecting to the internet, not finding apps ... Well, I don't use KDE, I prefer Cinnamon.
More than one desktop: It may work, but I have also seen threads about conflicts due to Desktop dependent different configs for the same program. Give it a try, it may work, but don't be too much disappointed if you run into problems. When I try out something I use one desktop in a VM.
Well, in the long years I have watched Debian and Devuan forums, I have seen several threads where the use of deb-multimedia packages have broken a system.
Myself I did install Acrobat Reader from deb-multimedia in the beginning of my Debian existance. That was an isolated package, the repo was enabled for the installation, and immediately afterwards it was disabled again. That starategy worked for me. Until I decided that the on board PDF tools were fine for me.
If you can, make a test installation in a disposable VM.
Nothing was removed, Devuan Daedalus installers won't boot from Ventoy. Period. Known fact, please search the forum. ISOs up to Chimaera and Excalibur again do work.
Well, chrony is an alternative. I am used to ntp(sec) ...
... but the time was broken in the GUI. I couldn't set it, nor enable network time, probably due to using a systemd component.
Its not the fault of Cinnamon nor a systemd influence. The package gnome-system-tools* is broken, and this for at least 4 years already. Don't try to use the gui, you can easily make it work on the console.
To be installed: ntpsec, ntpsec-ntpdate and ntpsec-ntpdig. Configuration in /etc/ntpsec/ntp.conf.
* gnome-system-tools is a Debian package w/o systemd dependency. The maintainer got lost somewhere in the Ukrainian planes, and there seems to be no replacement.
Addendum: according to your sources your are using xlibre instead of xorg. Is there any error in the x11.log? A switch to xorg could be another action.
You have to make sure that you just copy the relevant command which is [ -d /sys/firmware/efi ] && echo UEFI || echo BIOS, not the leading prompt and not any response. The later leads to that bash: UEFI: command not found.
I'll reset the bios to default I guess, then update grub. That would fix it, right?
Don't know. I wouldn't dare to guarantee anything. Anyhow, on one hand it should not hurt. Give it a try.
My advice: when you go and modify any setting in the bios or in your sw, take notes you can later refer to. Makes life easier in the long run.
Another thing I would try: reset this autologin to normal login with username/password until the blank screen is solved. Removes an additional variable.
Hi Robin, thanks for the information. My thoughts:
Bios version: Your PC has an efi Bios dated from 2013. I would make sure that this really is the latest that is available for this PC. And if not upgrade it in any case.
Legacy or efi mode: looks like your PC is booting in legacy mode now. Fits to the last update as indicated in response #7. Do you recollect how you installed your PC? Which mode did you use?
Dual boot: And you say you are dual booting? Which is the second OS on the other HDD?
Please show a
# fdisk -lSources.list: I think line 2 is a cripped deb-src entry. In any case its partially duplicating line 1, so it superfluous, it overwrites line 1. My suggestion: delete lines 2, 4 and 6. I think you are not compiling and linux software on your PC, are you?
And out of
[ -d /sys/firmware/efi ] && echo UEFI || echo BIOSis: nothing, no response from the terminal.
On my PC it works fine with copy/paste:
~$ [ -d /sys/firmware/efi ] && echo UEFI || echo BIOS
UEFIMy PC is running in efi mode obviously. Do not try to type that line in, you are prone to mistakes, use copy/paste to a terminal. Works fine in a user terminal.
Again:
On the other hand, just switching from efi to legacy should not at all affect the graphics, ...
I still think this is true. There must be another reason for the problems.
Do you remember what exactly, which settings, you modified in the bios a few days ago?
Maybe its worth checking the bios battery. Maybe that is borderline?
So I went into the BIOS and made sure Secure Boot was disabled, UEFI-only is enabled (it was Legacy when the trouble started). That was supposed to fix it but didn't...
Hi Robin, according what you just showed it looks like you switched your PC from efi mode to legacy mode at 2026-05-09 10:59:46. The update is a consequence of that.
greenjeans thought its the other way round.
In any case, looks like you played around in the bios without understanding what you did. As a matter of fact, in efi mode Secure Boot can be on (=often called Windows OS) or off (Other OS or similar). It is better to turn it off since there is no real benefit in having it enabled. In legacy mode with CSM on, SB is automagically disabled.
On the other hand, just switching from efi to legacy should not at all affect the graphics, except for whatever reason?
Maybe its a good idea to show us details about what we are talking about. Please show us the result from
inxi -Fzrfrom prompt to prompt in code tags please. It is possible that you have too install inxi first of all.
And in addition, the result from
[ -d /sys/firmware/efi ] && echo UEFI || echo BIOS What do you expect from a X11 replacement?
From my Excalibur on XLibre VM:
~$ env | grep -E -i 'x11|xorg|wayland'
XDG_SESSION_TYPE=x11Do also set on hold: linux-headers-amd64 linux-image-amd64
Typically there are no regular iso's prepared in the early stages of testing. This is happening when the release is due.
Normally, you take a stable, e.g. netinstall, do your installation, and then change the sources.list to freia. apt update/apt full-upgrade then will do the job.