You are not logged in.
Your initial post sounds like you are not aware what Devuan really is in relation to Debian.
Devuan developers look at the packages, if they are systemd or have systemd dependencies. These are replaced. All the rest is pure Debian. And there are some additional init systems that are supported.
If you are interested in which packages are on that list, have a look at https://pkgmaster.devuan.org/bannedpackages.txt
Me too. I don't use fstab, I wrote a cron script that is executed @reboot on the clients. My setup is a bit special anyhow. The server is shut down at nigth when all clients are off, and woken up in the morning by the first client in the morning. It checks if the network is available, if the server can be pinged, if the shares have been setup. Then the shares are mounted on the client(s). Server and clients are encrypted, the server is on a RAID6.
Great news. Congratulations to the team. Can't appreciate your work enough.
is there already a good wiki on how to start with NFS on Devuan?
I do not see any big difference between Debian and Devuan regarding the implementation of a nfs server and client. The guides you can find on the internet do apply for both in general.
I have started a file server on Sarge, first of all with Samba for the Windows PCs on my family network. When I started with dual booting with Linux on my home PC, I discovered that I could export the shares I had prepared for Samba easily for nfs too. I started off with nfs(3), and moved to nfs4 some years later. There are a few differences in the way you export the shares, but there is no relation to any init system.
On the client side, I wasn't successful with mounting the shares via fstab due to the fact that the file server was put to sleep at night, and it would wake up past the client, and it needed some additional time for decrypting the shares. So I created a script that is invoked by cron @reboot. The script checks if the server is online, then if the shares are exported, and then it mounts the shares. When the server is off the operation of my workstation isn't affected, just the shares aren't there. I am still using this proven and reliable approach nowadays.
No, neither of them ever were in the Debian/Devuan repos.
Veracrypt to be obtained from: https://veracrypt.io/en/Home.html
You may find a similar package in the repo: zulucrypt.
That hash at the front is there to indicate that this is a command to be used with root priviligues.
My findings:
Debian User Forum: forums.user.net page not found, what works is https://forums.debian.net/. Attention: Anti-bot software in place.
debianuserforums.org: 403, unsafe, probably certificate outdated.
Well, that does not happen by chance or arbitrary. There is a definite command at least in Excalibur:
apt modernize-sourcesHas to be called deliberately.
I don't think you are missing some packages. One explanation for the different grub display might be a different screen resolution. Have a look at the data sheets or the output from
inxi -FzThe inxi package may need to be installed.
A way to tweak the grub screen font size is to modify the /etc/default/grub file.
# nano /etc/default/grubChange the contents to show e.g.:
GRUB_GFXMODE=1024x768
GRUB_GFXPAYLOAD_LINUX=keepYou may modify the screensize further.
Then run:
# update-grub You can uncomment excalibur-security and excalibur-updates. They've been active since around when trixie was released.
Thanks, didn't know that. Good news, makes things a bit easier.
Yes.
I tend to upgrade to excalibur. So I du just the usual?
Not quite, there is one more thing to do and some remarks.
Before changing the sources to excalibur, you need to run the usrmerge app (I use apt instead of apt-get, I like the progress bar). If you don't do that you get an error message and the upgrade is stalled.
# apt install usrmergeThen, as long as Excalibur still is testing, the sources.list must contain just one line:
deb http://deb.devuan.org/merged/ excalibur main contrib non-free non-free-firmwareAnd the usual precautions do apply: make a backup and remove/deactivate all foreign repos from your sources.list.
Nowadays before buying a stick I do look at if there are read/write specs on the websites, and if a data sheet is available. Some only offer read rates, they are dropped from the list. Sticks without any of that stuff are sorted out. I do not look at the cheapest price, I look for the size and a reasonable performance.
A check if the memory on the stick really is the advertized one can be done with the f3 utility.
The first thing I do with any USB stick I buy.
Daedalus backports is on 6.12, as Excalibur.
Backports is the safe route, Excalibur means a slightly higher risk of ooups.
The Devuan installer is a pure text based installer, it does not need any graphics driver like nouveau ...
I never had any issues booting 5.0.0 or 5.0.1 on one of my PCs, no matter if legacy or uefi, except for when using Ventoy.
My guess: your PC has some strange configuration that does not like the basic settings the installer uses. Maybe related to your NVidea card (I only use AMD). And if 5.0.0 works, install with it, an apt update/apt full-upgrade installs everything to the latest version in the repo.
Addition: Devuan pulls in all Debian packages on a regular basis, filters out the systemd related stuff, adds the replacements where applicable. They end up in a "merged" repo which must be used to update any Devuan system.
I regularly see this "Stopping early crypto disks" during system shutdown. But I consequently use encryption. Tells me the system is mostly down, and now cryptodisks are being closed as one of the last steps in the shutdown process.
Maybe this is a step that happens all the time? I don't know.
How should messages been logged when the system is going down already? Disks have been unmounted already? No way.
I have just raised a bug report against Debian:
Bug#1117664: lightdm: Per default it is possible to login as user root graphically
nvme drives are no different from spinning rust or SSDs. On some older mainboards the bios must be prepared to support nmve's.
I do my partitioning before installation with gparted live. In a VM I use the installer but manual partitioning. I do not like the automatisms build in, in my eyes they are outdated in terms to size allowed. Usually too little. I want to determine size, order etc. myself based on a long time experience.
I have done that since the first tools came up on DOS like Partition Magic.
@greenjeans: Why do I consider graphical root login as security risk?
Here is my answer: I am on Daedalus with Lightdm/Cinnamon. I log in as user, the root account exists and has a password. I do not use sudo except for Veracrypt.
For root access I can either call up graphical programs like gparted where root access is granted by policykit, start a root terminal, or perform file copy/paste activities in Nemo via "Open as admin" if required. I do get along with these features very well. These features support everything I ever needed to do with root privilegues. Running root on the desktop is not required in my opinion.
BTW: everything started because a beginner in the German Debian forum complained about not being able to easily work as root despite the access being possible, something with LXDE and Lightdm.
The naive standard user who logs in as root has all possibilities to damage and wreck his system. Maybe "security" is not a perfect term for these risks. Well, somebody with deep knowledge and experience will easily get a long with root on the desktop. Me too.
In the IRC chat somebody said that root on the desktop isn't forbidden, just discouraged.
In my opinion lightdm shouldn't allow root access per default. It should be locked, and the experienced user may open this gate.
PS: other DMs like sddm and gdm inhibit root login.
Definitely lightdm has a leak. lightdm allows root login. Since I do not use other DM's, I can't speak for their performance.
Based on a web research and a chat on the Devuan Developers IRC, I have identified two possible fixes.
1.) Add a line to /etc/pam.d/lightdm
auth required pam_succeed_if.so user != root quiet2.) Set up a group that is allowed to login:
# groupadd dmlogin
# usermod -aG dmlogin urmel # urmel is a dummy. Replace with real usernames. Repeat for all other users that shall be permittedChange /etc/pam.d/lightdm and add:
auth required pam_succeed_if.so user ingroup dmloginIn both cases trying to login as root returns "wrong password".
Hi kapqa, I think it makes sense to open a new thread for your topic. I don't see any relation to the old thread.