Keith Ellis

It seems the latest versions of cAdvisor are not compatible with Unraid v7.2 and above. I've pinned cAdvisor to v0.49.1 and cAdvisor works. But above this version, cAdvisor does not connect to the docker backend. Is there an alternative to cAdvisor for Unraid 7.2? Or is there a way to get the latest versions of cAdvisor working with Unraid 7.2? Here is some summary output from ChatGPT on the matter: How cAdvisor ≥0.53 worksModern cAdvisor routes Docker metrics through containerd: Docker itself runs on top of containerd internally. The old Docker stats API (directly via /var/run/docker.sock) is deprecated in cAdvisor ≥0.53. If containerd socket is not accessible → cAdvisor cannot get container metrics at all. On Unraid 7.2: Docker uses containerd internally. containerd socket is hidden / in a private namespace. /var/run/docker.sock alone is not enough for cAdvisor v0.53+.

Hello, I've just compared our docker run commands and note the following: Netbox: Since you are using a custom docker network you can reference the Postgres and Redis containers by hostname, you don't need to use IP address or port number. The hostname of the docker is the Container Name in the Docker template. I note your docker run commands are not using a custom docker network, but I suspect you changed this after the initial post. So -e 'DB_HOST'='xxx.xxx.xx.xx'can be -e 'DB_HOST'='netbox-postgresql18' obviously change this to your Container name. For this to work, your Postgres container must be on the same custom docker network. -e 'DB_PORT'='xxxx' can be deleted from the template, you probably need to enable 'advanced' mode to do this at the top right of the docker template form. The same for -e 'REDIS_HOST' Change this to your Redis container name and delete -e 'REDIS_PORT' -e 'REDIS_PASSWORD' Since the Redis container is only available on the Netbox custom docker network and we are removing the Port, only containers inside this custom docker network can access Redis, so I think it is safe to remove the password. This is what solves your Authentication issues. REDIS You have a lot more defined in your Redis Docker template. I think I deleted a lot here. Try deleting the following from your template: -e 'ALLOW_EMPTY_PASSWORD'='no' -e 'REDIS_PASSWORD'='xxxxxxxxxxx'I think deleting the password bits will solve your authentication issue. I also do not have the following, but not sure this is an issue: -e 'REDIS_EXTRA_FLAGS'='--auto-aof-rewrite-percentage 100 --auto-aof-rewrite-min-size 64mb' -e 'CA_TS_FALLBACK_DIR'='/bitnami/redis' Give this ago, let me know how you get on. cheers. Keith.

OK, I’ve found a solution. Basically, stop the container, delete the media symlink and recreate a normal ‘media’ directory, start the container. Here is a summary from ChatGPT with a fuller explanation. Solution: NetBox “Symbolic link loop” on /config/media (Unraid + LinuxServer.io) If you see this error when starting the NetBox container: find: ‘/config/media’: Symbolic link loop chown: cannot dereference ‘/config/media’: Symbolic link loop **** Permissions could not be set. This is probably because your volume mounts are remote or read-only. **** …it’s caused by a recursive symlink created by the LinuxServer.io NetBox container when running on Unraid’s FUSE-based /mnt/user filesystem. What happened The container creates a symlink: /config/media → /config/media Because /config is mapped to /mnt/user/appdata/netbox, this becomes a loop, and the container can’t set permissions on the folder. Fix 1. Stop the NetBox container 2. Remove the symlink: rm /mnt/user/appdata/netbox/media 3. Create a real media folder: mkdir /mnt/user/appdata/netbox/media chown nobody:users /mnt/user/appdata/netbox/media chmod 775 /mnt/user/appdata/netbox/media 4. Start the container again After that, the container starts cleanly and media uploads work normally. Why this happens It’s an interaction between: - LinuxServer.io’s startup script - Unraid’s FUSE filesystem (/mnt/user) This isn’t documented by LSIO or NetBox, but it’s a known behaviour with several LSIO containers on Unraid.

Hello all, it seems there are issues with this netbox template. I’ve got NetBox working, but I have a problem with the media directory. The first log entry after starting the NetBox docker is find: ‘/config/media’: Symbolic link loop chown: cannot dereference '/config/media': Symbolic link loop **** Permissions could not be set. This is probably because your volume mounts are remote or read-only. **** **** The app may not work properly and we will not provide support for it. ****in my appdata folder I see: root@tower:/mnt/user/appdata/netbox# ls -l total 16 -rw-r--r-- 1 nobody users 11659 Nov 26 21:42 configuration.py -rw-r--r-- 1 nobody users 0 Nov 26 21:47 ldap_config.py lrwxrwxrwx 1 root root 13 Nov 26 21:45 media -> /config/media drwxr-xr-x 1 nobody users 0 Nov 26 21:42 scripts/My docker run command for NetBox looks like this: docker run -d --name='netbox' --net='netbox_net' --pids-limit 2048 -e TZ="Europe/London" -e HOST_OS="Unraid" -e HOST_HOSTNAME="tower" -e HOST_CONTAINERNAME="netbox" -e 'SUPERUSER_EMAIL'='[email protected]' -e 'SUPERUSER_PASSWORD'='secret_password' -e 'ALLOWED_HOST'='netbox.mydonmain.uk' -e 'DB_NAME'='netbox_db' -e 'DB_USER'='netbox_db_user' -e 'DB_PASSWORD'='secret_password' -e 'DB_HOST'='netbox-postgresql18' -e 'REDIS_HOST'='netbox-redis' -e 'REDIS_DB_TASK'='0' -e 'REDIS_DB_CACHE'='1' -e 'BASE_PATH'='' -e 'REMOTE_AUTH_ENABLED'='' -e 'REMOTE_AUTH_BACKEND'='' -e 'REMOTE_AUTH_HEADER'='' -e 'REMOTE_AUTH_AUTO_CREATE_USER'='' -e 'REMOTE_AUTH_DEFAULT_GROUPS'='' -e 'REMOTE_AUTH_DEFAULT_PERMISSIONS'='' -e 'PUID'='99' -e 'PGID'='100' -e 'UMASK'='022' -l net.unraid.docker.managed=dockerman -l net.unraid.docker.icon='https://raw.githubusercontent.com/linuxserver/docker-templates/master/linuxserver.io/img/netbox-logo.png' -p '8001:8000/tcp' -v '/mnt/user/appdata/netbox':'/config':'rw' 'lscr.io/linuxserver/netbox'My docker run command for Postgres is: docker run -d --name='netbox-postgresql18' --net='netbox_net' --pids-limit 2048 -e TZ="Europe/London" -e HOST_OS="Unraid" -e HOST_HOSTNAME="tower" -e HOST_CONTAINERNAME="netbox-postgresql18" -e 'POSTGRES_PASSWORD'='secret_password' -e 'POSTGRES_USER'='netbox_db_user' -e 'POSTGRES_DB'='netbox_db' -e 'TZ'='Europe/London' -e 'CA_TS_FALLBACK_DIR'='/var/lib/postgresql' -l net.unraid.docker.managed=dockerman -l net.unraid.docker.icon='https://raw.githubusercontent.com/Skylinar/unraid_templates/refs/heads/main/images/postgresql.png?raw=true' -v '/mnt/user/appdata/netbox-postgresql18':'/var/lib/postgresql':'rw' 'postgres:18'My docker run command for Redis: docker run -d --name='netbox-redis' --net='netbox_net' --pids-limit 2048 -e TZ="Europe/London" -e HOST_OS="Unraid" -e HOST_HOSTNAME="tower" -e HOST_CONTAINERNAME="netbox-redis" -l net.unraid.docker.managed=dockerman -l net.unraid.docker.icon='https://raw.githubusercontent.com/juusujanar/unraid-templates/master/img/Redis-logo.png' 'redis'Does anyone have any idea how I can fix the recursive media folder? Everything seems to be working fine, but I’m sure if I try and add images things will go wrong.

I’ve tried this, but as the images show in my original post, ALL dockers ramp up.

I do monitor my containers with cAdvisor, prior to the CPU runaway, the monitoring does not show any tell tail signs, but because cAdvisor is itself a container, as soon as the issue hits cAdvisor goes unresponsive and does not provide any data.

That is a guess, can I have a more targeted approach please.

There must be a more targeted approach available than this. With the number of dockers I have and the frequency of the lockups this will take 1.5 to 2 years.

Hello, I've been an Unraid user for quite a few years. Recently I've had issues with the CPU running away and pegging at 100% with system load ramping up to over 2000%. I have the server being monitored with Prometheus and Grafana. Last night 24/03/2025 @ 00:16:00 the CPU ran away. This prevents all dockers from running and makes it very hard to SSH into the machine. I noticed this at about 05:45 ish and managed to SSH in. I have learnt from experience that if I restart docker with /etc/rc.d/rc.docker restart the system usually recovers. Image below showing the CPU and RAM usage during this episode. Diagnostic files are also attached. Any help in diagnosing this issue would be very much appreciated. tower-diagnostics-20250324-2139.zip

Hello, I have a Fujitsu S1300. the drivers are not included as part of the sane package. I have the driver and can get sane working in Linux Mint. I need to figure out how to get the driver into the Docker image. Could anyone give me some directions please. Thank you.

That is great, thanks for your help.

Hi, I'm back from holiday now, I have power cycled the machine and I am still getting my parity drive as "disabled". I have moved the drives about in the slots and the parity drive is still disabled. The SMART check does not show any errors. I am on Unraid 6.12.2. How do I enable my parity drive again. diagnostics attached: tower-diagnostics-20230805-2034.zip

Ok, thanks. What does that mean. Possible hardware fault. How do I get the Parity back online? Do I need to do a hard power reset?

I’ve rebooted, Parity is still disabled, see attached image. I can’t do a complete power cycle for a week or so.

Hi, diagnostics zip file attached. I’m now going to reboot to see what happens. tower-diagnostics-20230726-0831.zip