44 minutes ago, moses19850 said:

Is this standard or not?

Yes this is the default.

You also don‘t have a password when logging into a Docker, pretty much the same.

You even can‘t login through ssh since ssh is not installed and root has no password that means by default sshd won‘t allow a connection anyways.

Thanks for the fast response. Understood.

Hi,

another question: for your the create scripts created by @ich777 : Can you please explain, why a termporary container is needed? Would it work without it? And if yes, which part need to be skipped?

37 minutes ago, moses19850 said:

for your the create scripts created by @ich777 : Can you please explain, why a termporary container is needed? Would it work without it? And if yes, which part need to be skipped?

I don‘t understand?

What scripts?

I have a feature request if it possible, have a gui to enable lxc configurations like cpu memory etc like in proxmox

1 hour ago, Del81 said:

I have a feature request if it possible, have a gui to enable lxc configurations like cpu memory etc like in proxmox

Still on my todo list but I have very limited time currently.

I just implemented a config editor in the GUI and you basically can limit it there with config entries.

On 6/15/2025 at 5:50 PM, ich777 said:

I don‘t understand?

What scripts?

This one for example: https://github.com/ich777/unraid_lxc_amp

4 hours ago, moses19850 said:

This one for example: https://github.com/ich777/unraid_lxc_amp

This is just for creating the container, the temporary container is just to create a base image and then run the scripts.

Do you want to create a container for other users or what is your goal?

Ok, my goal is same as yours, run a script, which

1.) creates a container

2.) install all dependencies

3.) configure the system for proper usage

My first container shall run https://hacksore.github.io/bluelinky-docs/

What works so far, after I create the container by hand:

.) install mariadb, needed to keep the data structured

.) install node.js and needed projcets (bluelinky, mariadb, nodejs-geolocation)

.) install grafana for having a vusialisation of the data

.) and some smaller staff

44 minutes ago, moses19850 said:

Ok, my goal is same as yours, run a script, which

So to speak you want to make the container archive accessible to other user correct?

Then I would always recommend doing it that way because it is way easier for you create a updated container and upload it to GitHub.

I also of course understand that you have to do it first by hand to test if everything is working but as said, I would recommend that you automate everything that you can create updated containers automatically.

You can find all examples here:
https://github.com/ich777?tab=repositories&q=lxc

Tailscale in LXC containers · Tailscale Docs

To configure this, this is the right way? is "lxc.cap.drop =" needed?

# Allow Tailscale to work
lxc.cap.drop =
lxc.cgroup2.devices.allow = c 10:200 rwm
lxc.mount.entry = /dev/net/tun dev/net/tun none bind,create=file

1 hour ago, L0rdRaiden said:

To configure this, this is the right way? is "lxc.cap.drop =" needed?

Yes, I'm not fully remembering if this is the correct way but it looks good to me, `lxc.cap.drop = ` is the easiest way at least.

Please note that you have to put this at the very end of the file.

On 7/4/2025 at 9:57 AM, ich777 said:

Yes, I'm not fully remembering if this is the correct way but it looks good to me, `lxc.cap.drop = ` is the easiest way at least.

Please note that you have to put this at the very end of the file.

As a reference is not needed, it works without it apparently, and it's an "unsafe" practice.

This is all network-related settings I use in my containers. Been working great since last year with Nginx Proxy Manager and more recently a container just for AdGuard Home.

LXC Config entries

# Network Config
lxc.net.0.type = veth

lxc.net.0.flags = up

lxc.net.0.link = br0

lxc.net.0.name = eth0

lxc.net.0.hwaddr=YOUR:MAC:ADDRESS:HERE

# Allow TUN access for Tailscale use

lxc.cgroup2.devices.allow = c 10:200 rwm

lxc.mount.entry = /dev/net/tun dev/net/tun none bind,create=file

Edited July 8, 2025Jul 8 by Espressomatic

10 hours ago, L0rdRaiden said:

it's an "unsafe" practice.

Exactly, easy means often times also unsafe. :)

What this does is basically drops no capabilities and has (almost) full access to the system.

I'm only running three LXC at the moment, 2 on one system and 1 on another. I love the simplicity of spinning them up and the easy access to so many relatively full distributions, unlike the pre-baked restrictions of (someone else's) docker container.

1 hour ago, Espressomatic said:

I'm only running three LXC at the moment, 2 on one system and 1 on another. I love the simplicity of spinning them up and the easy access to so many relatively full distributions, unlike the pre-baked restrictions of (someone else's) docker container.

I agree, the simplicity of setting up something real quick or try something out stands out as main benefit for me from LXC with the downside of course you have to update everything you do with the apps in the containers manually and of course the containers themselves too.

Again me,

since hours I struggle with connect a network share from the host and mount it within the LXC? Can sombody please help me?

I don't what's the problem: Within the container I do ping and enter the IP-address from the host and get response.

When I use the same IP-address in this command:

mount -t cifs -o username=user,password=password //192.xxx.xxx.xx/mnt/disk1/data/backup /mnt/data/backup

I get the error: special device xxxxxx does not exist.

Google does also not give me an answer.

Thanks upfront

22 minutes ago, moses19850 said:

Again me,

since hours I struggle with connect a network share from the host and mount it within the LXC? Can sombody please help me?

I don't what's the problem: Within the container I do ping and enter the IP-address from the host and get response.

When I use the same IP-address in this command:

mount -t cifs -o username=user,password=password //192.xxx.xxx.xx/mnt/disk1/data/backup /mnt/data/backup

I get the error: special device xxxxxx does not exist.

Google does also not give me an answer.

Thanks upfront

Why are you trying to mount it via smb when you can mount it directly?

lxc.mount.entry = /mnt/disk1/data/backup mnt/data/backup none bind 0 0

lxc.mount.entry = /mnt/disk1/data/backup mnt/data/backup none bind,create=dir 0 0

if the dir doesnt exist yet.

Edited July 28, 2025Jul 28 by Mainfrezzer

@Mainfrezzer Thanks a lot. Did not think about that.

question again: Maybe I did not see it, but is it possible to change the size of the LXC? 30GB is a little bit of an overkill for my puporse.

32 minutes ago, moses19850 said:

question again: Maybe I did not see it, but is it possible to change the size of the LXC? 30GB is a little bit of an overkill for my puporse.

What do you mean with that? RAM or what exactly?

Forget it, I missread the line memory usage. My bad.

Hi,
It's been a while since I started my AMP LXC which worked fine last time I span it up. But Now I get a cannot start error.

I have incluced by diagnostic.

Thanks :)

dous-diagnostics-20250813-2257.zip

9 hours ago, d0us said:

It's been a while since I started my AMP LXC which worked fine last time I span it up. But Now I get a cannot start error.

Sorry, you have a lot custom things in your AMP configuration file, remove all of that and see if it then works.

May I ask why are you passing through the Nvidia driver through to the container, also your method seems rather complicated and I assume this is the issue why it's not working, your installed Nvidia driver version: 570.86.16 you pass through files from driver: 565.77

You always can check from the command line what went wrong with:

lxc-start -F CONTAINERNAME