You are not logged in.
Well, pinning isn't required. You can disable search for updates in the options. 7.6.7.2 is the latest version before they switched their naming scheme. I had it on Daedalus, and I am using it on Excalibur now. And I am happy with what I have got, I use Writer and Calc, the rest is installed but not used. This way I have a constant UI. You may play around a bit with the symbol sets supplied to find a suitable one for you.
You download three packages from their website (main, help and language pack) and extract the contents to /tmp.
When you do an update of the main version, like in this case from 7.4 to 7.6, you need to remove the old stuff first of all. Settings remain in place.
# apt remove libreoffice*Installation:
# cd /tmp/Libreoffice_xyz_Linux_x86-64_deb/DEBS
# dpkg -i *.deb
# cd /tmp/Libreoffice_xyz_Linux_x86-64_deb_languagepack_nn/DEBS
# dpkg -i *.deb
# cd /tmp/Libreoffice_xyz_Linux_x86-64_deb_helppack_nn/DEBS
# dpkg -i *.debBasically you have the latest generation of AMD board, CPU and GPU. It is very much possible that your HW is too new for the kernel on the Excalibur netinstall. That works fine on Ventoy now. But my HW is about 6 years old by now.
Is there a working Freia netinstall available? On the internet I see a version dated June 30th? Is that good? Haven't tested it yet. Then I would try "Testing".
When LO 7.4 came out there were some voices that said it was not the best. I have downloaded the latest 7.6 version and that works fine. I am not using the LO version from the repo.
... you need to change "deb.devuan.org" to "archive.devuan.org" ...
Thanks, that made it work.
I am trying to download packages from Beowulf which is archived now. The Devuan Package View presents a download path like http://deb.devuan.org/merged/pool/DEBIA … _amd64.deb but that gives an error "Not Found
The requested URL was not found on this server."
How can I access the archive for downloading packages? I was searching for a path, but couldn't find it.
I am using the traditional "full disk"-encryption with luks, that leaves /boot unencrypted.
I am prepared to help if required.
Can that please be the last off-topic post in this thread that's supposed to be about display managers that work with both X and Wayland?
You are right, the xlibre stuff is sort of off-topic here. Please can you move that to e.g. https://dev1galaxy.org/viewtopic.php?id=7388?
@greenjeans: you got it.
I thought the doc is wrong and corrected my entry according to a correct deb822 syntax as used by Debian/Devuan where "Suites" stands for the release names and "Components" for main contrib ..... These "corrections" caused the release file issues.
When using your version, apt update works and my VM is getting the latest xlibre update. With one exception: xserver-xlibre-video-amdgpu. That is fixed when I run an
apt -t excalibur-backports install xserver-xlibre-video-amdgpuUnfortunately, the VM is no more starting to the graphical display. I get a blinking cursor on a black screen, and a bar that might be something like a pattern from a text output where the pixels are not aligned. Same symptons I got with another test on Freia.
Thats exactly what does not work for me.
@fsmithred: sorry, the link to xlibre you posted yesterday is not working as expected. That page can be read but contains severe problems. The sources file is bad, Suits and Components is mixed up. And when I try to change my "working" sources to what is displayed there with corrections I get "no release file found".
What works forme is this older information:
Types: deb
URIs: https://xlibre-deb.github.io/devuan
Suites: excalibur
Components: main
Architectures: amd64
Signed-By: /etc/apt/keyrings/xlibre-deb.asc
Enabled: yesNo updates for a while but no errors on apt update.
Not working, meanwhile disabled again:
Types: deb
URIs: https://xlibre-debian.github.io/devuan
# Suites: excalibur
Suites: stable
Components: main
Architectures: amd64
Signed-By: /usr/share/keyrings/NexusSfan.pgp
Enabled: noWould be nice if you could post these console outputs in code tags, refer to the button labelled code in rectangular brackets above the editor window. Its better readable.
What I see is a discrepancy between the System Info and the real kernel version. I see you are on Excalibur, and you use xlibre instead of xorg.
dmesg says: "Optional firmware "amdgpu/raven_ip_discovery.bin" was not found" which I would read as not required but stops with error. In any case its not part of the amdgpu firmware package which I also have installed.
Also looked at the latest firmware package on kernel.org, there it is included. So it was added sometime between the Excalibur version and June this year. What you may try is to install the firmware-amd-graphics from backports, if there is any.
It probably would make sense to show the output from
inxi -Fzxrto get some basic understanding what we are talking about.
inxi may need to be installed first of all.
BTW: I looked into my firmware folder and did not see any "raven_ip_discovery.bin" in there, being on Excalibur.
I can't recall any network failure during installations that was not due to operator error, i.e. mistyping IP numbers or similar. Anyhow, two nics is an unusual setup and maybe this has led to problems? I have only dealt with one nic on many installations.
AFAIK there is no way to skip modules for known nics, unlike this topic that you may only get ext2 and no ext3/4 when jumping over the "load extra modules" in expert mode.
- The installer did not activate the link for the Ethernet interface, so mirror, clock, etc. did not work while installing.
I would claim this is not an installer issue. If the network does not work its most likely an issue related to user input.
sysvinit just does its job. I have been using OpenRC when I started with Devuan on ASCII. I have moved back to sysvinit nowadays on new installations since I did not see any advantages as a user from using OpenRC.
As fas as I know it doesn't. (And I'm not sure, if this was changed with ntpsec?)
I does indirectly. When I shut down my PC the HW clock is updated. ntp and ntpsec are the same.
If NTP would be depending on the HW-clock, how should it work with devices (e.g. singleborad-computers) without any such?
I am pretty sure any device supporting a clock has a RTC on board.
Do you know if that sets OS time only, or hardware time too?
ntp sets the HW clock. The rest is job of the operating system.
In Linux its standard to set the HW re. the bios clock to UTC. During installation you configure your location and timezone, and the OS then generates the local time from that.
Windows per default is running on local time, i.e. the bios clock is set to local time. This creates 2h time shifts on a Linux/Windows dual boot system. The best fix is to set Windows to UTC.
i wonder, will the same mod be made to the xlibre packages?
There are xlibre packages that are especially made for Devuan.
Excrypted Excalibur with Cinnamon DE, static wired network defined in /etc/network/interfaces, no network manager used.
With increasing frequency it looks like the network is no more available after hibernation. How can I easily reset the network w/o rebooting?
~$ inxi -Fxzr
System:
Kernel: 6.12.94+deb13-amd64 arch: x86_64 bits: 64 compiler: gcc v: 14.2.0
Desktop: Cinnamon v: 6.4.10 Distro: Devuan GNU/Linux 6 (excalibur)
Machine:
Type: Desktop Mobo: ASUSTeK model: PRIME X570-PRO v: Rev X.0x
serial: <superuser required> UEFI: American Megatrends v: 5031
date: 01/13/2025
CPU:
Info: 12-core model: AMD Ryzen 9 5900X bits: 64 type: MT MCP arch: Zen 3+
rev: 2 cache: L1: 768 KiB L2: 6 MiB L3: 64 MiB
Graphics:
Device-1: Advanced Micro Devices [AMD/ATI] Navi 14 [Radeon RX 5500/5500M /
Pro 5500M] vendor: Sapphire driver: amdgpu v: kernel arch: RDNA-1
bus-ID: 0c:00.0
Display: x11 server: X.Org v: 21.1.16 driver: X: loaded: amdgpu
unloaded: fbdev,modesetting,vesa dri: radeonsi gpu: amdgpu
resolution: 2560x1440~60Hz
Network:
Device-1: Intel I211 Gigabit Network vendor: ASUSTeK driver: igb v: kernel
port: f000 bus-ID: 06:00.0
IF: eth0 state: up speed: 1000 Mbps duplex: full mac: <filter>
Swap:
ID-1: swap-1 type: partition size: 183.11 GiB used: 0 KiB (0.0%)
dev: /dev/dm-2 mapped: vg--exc--sys-2_swap
Sensors:
System Temperatures: cpu: 39.0 C mobo: 43.0 C gpu: amdgpu temp: 47.0 C
Fan Speeds (rpm): N/A gpu: amdgpu fan: 0
Repos:
Packages: 1837
Active apt repos in: /etc/apt/sources.list.d/signal-xenial.list
1: deb [arch=amd64 signed-by=/usr/share/keyrings/signal-desktop-keyring.gpg] https://updates.signal.org/desktop/apt xenial main
Active apt repos in: /etc/apt/sources.list.d/brave-browser-release.sources
1: deb [arch=amd64] https://brave-browser-apt-release.s3.brave.com stable main
Active apt repos in: /etc/apt/sources.list.d/devuan.sources
1: deb http://deb.devuan.org/merged/ excalibur excalibur-security excalibur-updates excalibur-backports main contrib non-free non-free-firmware
Info:
Memory: total: 64 GiB note: est. available: 62.71 GiB used: 5.56 GiB (8.9%)
Processes: 590 Uptime: 4h 21m Init: SysVinit runlevel: 2
Compilers: gcc: 14.2.0 Shell: Bash v: 5.2.37 inxi: 3.3.38What I miss in the man pages are examples for at least standard applications. manpages usually just list the options.
System: luks encrypted Excalibur on sysvinit with Cinnamon DE.
I have split up my data into a static part and a dynamic part. The static part are all the documents, pictures and other data I have. No .dot files in there. Resides on a SSD and is decrypted and mounted to /mnt/HomeDaten during bootup. fstab entry is:
/dev/mapper/homedata_crypt /mnt/HomeDaten ext4 defaults 0 2The folders on this drive are the ones used by xdg.
All .dot files reside in the normal /home/$USER folder. /home is on a separate drive, also luks encrypted. fstab entry is:
/dev/mapper/home_crypt /home ext4 defaults 0 2
I am using symbolic links to get access to the folders on the static drive. Typical command (in German):
ln -s /mnt/HomeDaten/$USER/Dokumente /home/$USER/Dokumente Problem:
In most folders I cannot move deleted files into the trashcan where I have the chance to review/undo the deletion. I must immediately delete any file. This is not only limited to the linked folders, I seem to have a partial problem in my /home too. Not everywhere. I haven't fully understood the systematics behind the behaviour yet. I played around and copied some empty trashcans on my HomeDaten folder and into the individual folders I have linked. No succes.
What is the reason and what can I do to fix the problem (keeping the setup)?
486 UEFI? No way. There is no 486 CPU supported by any bios with uefi.
Please show:
inxi -FzDon't forget, kernel sizes are growing too. What worked some releases ago must no more work with todays kernels.
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.