5 hours ago, ich777 said:

I'll look into that, please give me a few days.

 

Please update the plugin, tested and working with the provided URL.

it works perfect now,  thanks

I decide to fork your instance:
https://github.com/bmartino1/unraid_lxc_debian_novnc/tree/Production

https://github.com/bmartino1/unraid_lxc_debian_novnc/releases/tag/Production

As I needed/wanted a debin lxc with a vnx web access for a app latter. Found yours and though it be faster then a fresh install and dabbling in a simlar install method... For now i gues manual Debian install and run the build configs manual for the time being to atleast have something...

I was able to install your test template version. My only issues i was locked out of sudo/root to install any applicaitons.


I've recently done a overhaul and reinstalled  / tested some settings.



I'm not sure if zfs storage broke the create script or not.
I've seem to have hit a snag somewhere and wanted your opinion:
 

WARNING: apt does not have a stable CLI interface. Use with caution in scripts.

Stopping temporary container
Copying over build logs from container
Performing final cleanup from container
Generating build.log
Packing up container
tar: -: Wrote only 8192 of 10240 bytes
tar: Error is not recoverable: exiting now
./createLXCarchive.sh: line 105: 1100742 Exit 2                  tar -cf - .
     1100743 Killed                  | xz -9 --threads=$(nproc --all) > ${LXC_PACKAGE_DIR}/${LXC_PACKAGE_NAME}.tar.xz
Creating md5 checksum
Stopping and destroying temporary container

 

11 minutes ago, bmartino1 said:

I'm not sure if zfs storage broke the create script or not.

No it does not because the build script doesn't use ZFS if you haven't modified it.

 

12 minutes ago, bmartino1 said:

WARNING: apt does not have a stable CLI interface. Use with caution in scripts.

Most likely because you are using apt alone ant not apt-get, in headless scripts you should always use apt-get not apt.

 

13 minutes ago, bmartino1 said:

tar: -: Wrote only 8192 of 10240 bytes

Are you sure you have enough free space where the temporary directory is located at?

 

15 minutes ago, bmartino1 said:

I was able to install your test template version. My only issues i was locked out of sudo/root to install any applicaitons.

What are you doing here in the screenshot? Please type in `su` and not sudo su

The root password is set here and it's Unraid.

 

Please make issues on GitHub and mention me in your repositories/Issues, these are development related questions and should not really belong here in this thread.

5 minutes ago, ich777 said:

No it does not because the build script doesn't use ZFS if you haven't modified it.

 

Most likely because you are using apt alone ant not apt-get, in headless scripts you should always use apt-get not apt.

 

Are you sure you have enough free space where the temporary directory is located at?

 

What are you doing here in the screenshot? Please type in `su` and not sudo su

The root password is set here and it's Unraid.

 

Please make issues on GitHub and mention me in your repositories/Issues, these are development related questions and should not really belong here in this thread.

yes, used a fresh fork downlaod

Stopping temporary container
Copying over build logs from container
Performing final cleanup from container
Generating build.log
Packing up container
Creating md5 checksum
Finished Container still exist

more then enogh free space...

 

# create container archive and md5 sum
echo "Packing up container"
tar -cf - . | xz -7 --threads=$(nproc --all) > ${LXC_PACKAGE_DIR}/${LXC_PACKAGE_NAME}.tar.xz
echo "Creating md5 checksum"
md5sum ${LXC_PACKAGE_DIR}/${LXC_PACKAGE_NAME}.tar.xz | awk '{print $1}' > ${LXC_PACKAGE_DIR}/${LXC_PACKAGE_NAME}.tar.xz.md5
echo "--------------------END--------------------" >> ${LXC_PACKAGE_DIR}/build.log

# upload to GitHub
#TBD

# remove packages (comment if you don't want to remove them)
#rm -rf ${LXC_PACKAGE_DIR}/${LXC_PACKAGE_NAME}.tar.xz ${LXC_PACKAGE_DIR}/${LXC_PACKAGE_NAME}.tar.xz.md5 ${LXC_PACKAGE_DIR}/build.log

echo Finished Container still exist
# remove container
#echo "Stopping and destroying temporary container"
#lxc-stop -k -n ${LXC_CONT_NAME} 2>/dev/null
#lxc-destroy -n ${LXC_CONT_NAME}

had to drop tar compression from 9 to 7

Thanks for taking a look

39 minutes ago, bmartino1 said:

had to drop tar compression from 9 to 7

Looking again at the message, it is most likely the case that you where running out of RAM.

You should lower the cores which are used for compression not the ratio.

 

Again, please lets discuss this on GitHub since this has nothing to do with the plugin itself.

On 5/23/2022 at 11:18 PM, ich777 said:

cgroup v2 (ONLY NECESSARY if you are below Unraid version 6.12.0): Distributions which use systemd (Ubuntu, Debian Bookworm+,...) will not work unless you enable cgroup v2

To enable cgroup v2 append the following to your syslinux.conf and reboot afterwards:

Hello, thank you very much for developing this plugin. However, I encountered some issues when limiting the CPU and memory of the container. My system is 6.12.13, but I cannot set the container using cgroupv2, which causes my lxc container to fail to start. If using cgroup settings, the CPU can be successfully limited, but the memory cannot be set.

2 hours ago, cdf5b9fac9ad said:

Hello, thank you very much for developing this plugin. However, I encountered some issues when limiting the CPU and memory of the container. My system is 6.12.13, but I cannot set the container using cgroupv2, which causes my lxc container to fail to start. If using cgroup settings, the CPU can be successfully limited, but the memory cannot be set.

Please post your Diagnostics for further troubleshoting or at least how your config looks or what you added.

 

I really can‘t tell what‘s going on with the information you provided.

looks like unraid 7 beta 4 has a broken backup for lxc:



snapshots work, backup does not..

weird the backup trie dto make a zfs snapshot?

root@BMM-Unraid:~# zfs destroy vm-zfs/zfs_lxccontainers/PBS/snap2
cannot unmount '/mnt/vm-zfs/LXC/General/PBS/snaps/snap2/rootfs': pool or dataset is busy
root@BMM-Unraid:~# 



treid a seconed time reveiwing the system log... went to fuse link orgianl was direct to zfs pool...


root@BMM-Unraid:~# zfs destroy  vm-zfs/zfs_lxccontainers/PBS/
vm-zfs/zfs_lxccontainers/PBS/PBS    vm-zfs/zfs_lxccontainers/PBS/snap0  vm-zfs/zfs_lxccontainers/PBS/snap1  vm-zfs/zfs_lxccontainers/PBS/snap2

cant seem to delete broken backup and 2 weird snapshots exist..

bmm-unraid-diagnostics-20241024-1417.zip

bmm-unraid-diagnostics-20241024-1451.zip

Edited October 24, 20241 yr by bmartino1
aditional data

1 hour ago, bmartino1 said:

looks like unraid 7 beta 4 has a broken backup for lxc:

Nope, what browser are you using?

 

Your LXC container is simply to big so that your browser times out while creating the backup, please use lxc-autobackup from the command line.

 

1 hour ago, bmartino1 said:

weird the backup trie dto make a zfs snapshot?

I really don't understand you even configured it to create a snapshot...

 

When you click on the little question mark in the top right corner when you are on that page: you would actually see what this means:

1 hour ago, bmartino1 said:

root@BMM-Unraid:~# zfs destroy  vm-zfs/zfs_lxccontainers/PBS/
vm-zfs/zfs_lxccontainers/PBS/PBS    vm-zfs/zfs_lxccontainers/PBS/snap0  vm-zfs/zfs_lxccontainers/PBS/snap1  vm-zfs/zfs_lxccontainers/PBS/snap2

Can you please format such things as code, it is really hard to read...

You first have to destroy snap2 then snap1 and then snap0

 

However are you sure the backup process isn't running in the background any more?

3 hours ago, ich777 said:

Nope, what browser are you using?

 

Your LXC container is simply to big so that your browser times out while creating the backup, please use lxc-autobackup from the command line.

 

I really don't understand you even configured it to create a snapshot...

 

When you click on the little question mark in the top right corner when you are on that page: you would actually see what this means:

Can you please format such things as code, it is really hard to read...

You first have to destroy snap2 then snap1 and then snap0

 

However are you sure the backup process isn't running in the background any more?

Time got away from me, sorry. 

Will do in the future. Ran out of time for an event and caught the bug right at time i had to leave. my bad...

yes, as I rebooted the unraid system... after adding the aditional data and remoted unraid... I will use terminal comand next time. Ironically after the reboot I have 2 backups...



the nul snapshots are gone as well... not sure I trust them, though. 
 

4 hours ago, bmartino1 said:

the nul snapshots are gone as well... not sure I trust them, though. 

What do you mean with that?

Before reboot when gateway times out.
Snapshot had a null date data picture in diag post.
After reboot, snapshot data gone, but ?backup data exists?

a drastic change with no ability to verify if the backup was successful or not.
So I don't trust the backup file/files...

so I delete them, and used the terminal command to certify a good backup.

Hi, guy. You did a good job and still keep going on this. I got machine running lxc container perfectly before,now i got a new one, newest version 2024.10.29, downloaded debian and openwrt images, unfortunately two of them failed to start. 

diagnostics file as following

m5p-diagnostics-20241107-0941.zip

4 hours ago, hunterson said:

I got machine running lxc container perfectly before

Sorry no support from my side since it seems to me that you are running a counterfeit version from Unraid from what I can see from your Diagnostics.

 

Please buy a copy from Unraid and we can look into that further.

@SpencerJ

21 hours ago, hunterson said:

Hi, guy. You did a good job and still keep going on this. I got machine running lxc container perfectly before,now i got a new one, newest version 2024.10.29, downloaded debian and openwrt images, unfortunately two of them failed to start. 

diagnostics file as following

m5p-diagnostics-20241107-0941.zip 94.97 kB · 5 downloads

Booooo. Please buy a legitimate version 

17 hours ago, ich777 said:

Sorry no support from my side since it seems to me that you are running a counterfeit version from Unraid from what I can see from your Diagnostics.

 

Please buy a copy from Unraid and we can look into that further.

@SpencerJ

wo, Seemed to be warned  that It was a version for personally testing use when I downloaded the system.

@ich777

i am not sure if you noticed this but i think this needs to be fixed 🤣
 

Edit: seems like adding a just a / to the href fixes it

Edited December 14, 20241 yr by Mainfrezzer

8 minutes ago, Mainfrezzer said:

@ich777

i am not sure if you noticed this but i think this needs to be fixed 🤣
 

 

Nope, will look into this, for now you have to click two times. 😉

Just now, ich777 said:

Nope, will look into this, for now you have to click two times. 😉

i did already fix it by adding a / at the beginning of href in the LXCBackup.page

Certainly wasnt expecting that to pop up

26 minutes ago, Mainfrezzer said:

i did already fix it by adding a / at the beginning of href in the LXCBackup.page

Certainly wasnt expecting that to pop up

Fixed with todays update.

Can anyone give me guidance on passing nvidia gpu to the LXC container, i tried to follow the pinned post but it is too complex and not easy to understand for me , need a stepwise tutorial in simple language.

 

thanks

2 hours ago, drmetro said:

Can anyone give me guidance on passing nvidia gpu to the LXC container, i tried to follow the pinned post but it is too complex and not easy to understand for me , need a stepwise tutorial in simple language.

I just took a look at the tutorial and it is very well written...

 

If you just need Nvidia GPU in the container then do:

1. Install the Nvidia Driver plugin on your Unraid Server
2. Set up LXC
3. Download the zip file from the linked post and open readme.txt
4. Create a container and but don't start it
5. Edit the container config and follow step 6 from readme.txt what to put into the config
6. Start the container and open up a terminal from the container
7. Follow everything from step 8 (granted you have all necessary dependencies installed)
8. Done

 

I can't help any further since I don't have a Nvidia GPU installed on my LXC system however even if you follow the readme.txt from the pinned post step by step you will eventually end up with a container with Nvidia GPU support and a Desktop environment.

3 minutes ago, ich777 said:

I just took a look at the tutorial and it is very well written...

 

If you just need Nvidia GPU in the container then do:

1. Install the Nvidia Driver plugin on your Unraid Server
2. Set up LXC
3. Download the zip file from the linked post and open readme.txt
4. Create a container and but don't start it
5. Edit the container config and follow step 6 from readme.txt what to put into the config
6. Start the container and open up a terminal from the container
7. Follow everything from step 8 (granted you have all necessary dependencies installed)
8. Done

 

I can't help any further since I don't have a Nvidia GPU installed on my LXC system however even if you follow the readme.txt from the pinned post step by step you will eventually end up with a container with Nvidia GPU support and a Desktop environment.

How do I know if i have all the dependencies installed ?
 

4 minutes ago, drmetro said:

How do I know if i have all the dependencies installed ?

If something is not working or you get an error you don't have all dependencies installed.

 

Sorry but these are the basics, you should really look into the basics and then start with that project.

 

Have you already looked into if there are Docker containers out there for the apps that you want to run because you could run multiple containers with one GPU in Docker.

No the app I want to use doesn’t have any docker image or docker container in unraid community apps , but on github it does have a docker compose yml file but I don’t know how to install it , also there is no active support from the repo admin.

 

https://github.com/CatchTheTornado/pdf-extract-api

 

im a med student i have multiple handwritten note books which i want to OCR but there are no good OCR apps on community apps so i want to install the above github repo OCR tool but i dont know how to , only way im seeing is LXC container with GpU support because docker im not following it.

Edited December 15, 20241 yr by drmetro