Quick question here, I have 2 Cache folders, Can I put 2 different Cache folders in the Docker config? If so, how do I do that, what would be the second container path?

You can do anything, could be "/mnt/cache2" inside docker container. Next you need to add same Path to monitoring (zabbix webui)

Hello,

I am having a couple issues and any direction would be great in my Zabbix learning path.

 

When Zabbix reaches out to the agent-2 it comes back with "Cannot fetch data: Get "http://1.28/system/df": dial unix /var/run/docker.sock: connect: permission denied."

 I am running the container with Privileged turned on. 

 

From the server side I am assuming the user zabbix is requesting

 grep zabbix /etc/passwd
zabbix:x:985:985:Zabbix Monitoring System:/var/lib/zabbix:/sbin/nologin

 

I am not sure where to look to get this resolved.

 

---------------------------

 

My other issue that I think is unrelated but ill throw it up here in case anyone has any clues. Any metric with "unraid" in it returns an unknown. Exmaple:  "Unknown metric unraid.disks.totals"

 

I am trying to figure out why it is not working and what to change it to if that is the issue.

 

 

I am sure I am missing information. Any guidance or suggestions would be greatly appreciated. 

 

Thanks in advance for any help!

On 10/1/2024 at 5:43 PM, Gobananas said:

When Zabbix reaches out to the agent-2 it comes back with "Cannot fetch data: Get "http://1.28/system/df": dial unix /var/run/docker.sock: connect: permission denied."

 I am running the container with Privileged turned on. 

 

Try creating a new path variable inside the Docker container configuration - path in container and on host is /var/run/docker.sock. Save it and then do the "add group", "add to groups file" thing, restart container. After I added /var/run/docker.sock to my container config this started working!!

On 8/19/2021 at 5:10 PM, jimmy898 said:

With [PLUGIN] CA APPLICATION AUTO UPDATE I've added a script that runs after the update at container start. It should automate the permissions fix.

Core commands are:

 

docker exec --user root zabbix-agent2 sh -c "addgroup -g 281 docker"
docker exec --user root zabbix-agent2 sh -c "sed -i "s/docker:x:281:/docker:x:281:zabbix/g" /etc/group"

 

@jimmy898 can you share the script / how you got this to work?

 

Is this as simple as:

 

#!/usr/bin/bash
docker exec --user root zabbix-agent2 sh -c "addgroup -g 281 docker"
docker exec --user root zabbix-agent2 sh -c "sed -i "s/docker:x:281:/docker:x:281:zabbix/g" /etc/group"

 

Edited October 25, 20241 yr by autumnwalker

On 9/30/2022 at 11:34 AM, TurboStreetCar said:

I will try passing /proc/net/dev to the container at the /host/ path, but im not sure how to "reroute" the net key to that new location. i assume it would be somewhere in the conf file? Only issue with that is updates would break that modification. 
 

 

my unraid also is using br0 for the IP address, and it does not show up in Zabbix.

It seems I am facing the same challenge.

 

Were you able to solve it in the meanwhile? How?

 

3 hours ago, murkus said:

 

my unraid also is using br0 for the IP address, and it does not show up in Zabbix.

It seems I am facing the same challenge.

 

Were you able to solve it in the meanwhile? How?

 

 

I found how to fix this. The reason that the container does not see the host interfaces is that the container runs in network bridge mode. I don't know whether that is the default for this container as it comes from @VRx, I may have switched to from network host mode to network bridge mode myself at some point.

 

If the container runs in network host mode it does see all interfaces (the actual interface name filtering can be changed in the LLD macros).

 

@VRx Just in case the container does not come set to host network mode by edfault, this may be a welcome change for the next release.

 

I had the same problem with the '/mnt/array' mount on Zabbix. I did a 'df -h' and see that the filesystem is 'shfs', 

 

nas:/var/lib/zabbix$ df -h
Filesystem      Size  Used Avail Use% Mounted on
/dev/loop2       60G   23G   37G  39% /
tmpfs            64M     0   64M   0% /dev
shm              64M     0   64M   0% /dev/shm
/dev/sdc1       466G   75G  391G  17% /mnt/cache
shfs             73T   11T   63T  14% /mnt/array
/dev/loop2       60G   23G   37G  39% /etc/hosts
tmpfs           128M  2.1M  126M   2% /run/docker.sock

 

 

so, in Zabbix I clone all items for the '/mnt/cache' (which works), as suggested by @epiphanyplx, replacing 'cache' for 'array' and 'btrfs' for 'shfs', and it seems to work.

 

I don't know on others versions, but in Unraid 7.0, '/mnt/user' is a  'shfs' filesystem.

 

Hope it helps.

Edited February 10, 20251 yr by archibaldo

On 2/10/2025 at 5:25 PM, archibaldo said:

so, in Zabbix I clone all items for the '/mnt/cache' (which works), as suggested by @epiphanyplx, replacing 'cache' for 'array' and 'btrfs' for 'shfs', and it seems to work.

Thanks, note that this doesn't help with the missing NICs.

I am trying to get Zabbix to monitor my GPU via nvidia-smi (https://github.com/zabbix/community-templates/tree/main/Server_Hardware/Other/template_nvidia-smi_integration/7.0). I've successfully mapped /etc/zabbix/zabbix_agentd.d as a volume and I created my .conf there for UserParameters. I have those working, but I get an error of Value "sh: nvidia-smi: not found" in Zabbix.

I suspect this is due to permissions and the container cannot execute nvidia-smi on the server. Any thoughts about how to do this?