Monitoring LXD VMs

We’d been able to monitor LXD containers using Netdata and wanted to know if there’s a way to monitor the VM’s as well since LXD VM’s use QEMU.

Saw this:

https://community.netdata.cloud/t/need-some-help-with-cgroups-configuration-for-qemu-vms/1167/3?u=philip

But in the /sys/fs/cgroup tree I didn’t see any qemu slices. Does this mean it’s not possible? Or do the VM’s groups reside in another location??

Hello!

Try this it might help
https://learn.netdata.cloud/docs/agent/collectors/cgroups.plugin

Regards.

@Ibragim_Ganizade

Have seen it but where do the lxd’s VM cgroups reside?

I have allowed qemu in the search pattern but Netdata doesn’t show the VM’s, containers are visible though.

Now I will try to deploy LXD VM on a test environment and check how Netdata works

Regards.

Yeah, @Ibragim_Ganizade. Thank you!

Very interesting

On a freshly installed Ubuntu, I started the container and vm

NetData only shows the container

I think out of the box NetData will not show the virtual machine

I think I need to somehow configure NetData

I am looking further

Regards.

For containers, there are these “lxc.payload.container_name” directories where the Netdata reads the information from. For VMs I do not know where they exist - if at all they exist.

The URL I shared, this one:

https://community.netdata.cloud/t/need-some-help-with-cgroups-configuration-for-qemu-vms/1167/3?u=philip

That person said he’s using Proxmox VMs and he could see the VM metrics but they are grouped together, how to ungroup them is the question he had.

However, if Proxmox VMs are visible, then LXD VM should be visible too, as LXD also uses QEMU for creating the VMs. That’s what I so far know. Please add to it anything you know, or correct me if I said something wrong.

Anyways, thank you for looking into this!! Much appreciated!

1 Like

Thank you for the idea

Proxmox uses KVM VMs

And I decided to install Virt-Manager and deploy VM there

Netdata Saw my virtual machine and container

But still can’t see LXD VM

I think there is a possibility that Netdata does not support LXD VM (But this is not certain)

I hope there is a solution to this problem because I also use Netdata + LXD (It would be nice to have monitoring for the VM)

Regards.

Ah, I see.

Yes, it would be nice to have Netdata monitor LXD VMs as well.

Have you looked at LXD’s metric exporter?

https://linuxcontainers.org/lxd/docs/master/metrics/#metrics

Will have a look at that. But it would be nice to have Netdata monitor VMs as well as we currently use it to monitor LXD containers.

What will be cgroup name or the location of the VMs will be?

I’m not sure what you mean. VMs don’t have cgroup names?

What I mean is why I’m not seeing anything related to LXD VMs in cgroup information? Some people see even the qemu metrics in netdata. But even when LXD uses qemu why am I not able to see them in Netdata? That’s out of your hands I think. But where do the VM cgroups reside?

LXD doesn’t use cgroups for VMs. It runs the qemu process directly.

Oh, I see. So there’s no way of monitoring the VM with Netdata?

I dont know anything about Netdata I’m afraid.
But you’d not likely get much info out of a cgroup that just wrapped a qemu process (maybe CPU and memory usage of the process itself).

Does netdata have support for scarping openmetrics endpoints?

I don’t know. I’ll have to ask the netdata team.

Hi all!

I think NetData should add support for LXD VM

It would not be bad, I think a lot of people need it
How will the news please let me know

Regards.

Surely, will do that.

We’ll probably have a call with the netdata team as we are facing some other issues with it, so they said a real-time conversation may be better.

This one here: https://community.netdata.cloud/t/lxc-containers-stats-are-not-shown/3616

That’s a long post, if you know anything about that issue, please let me know.

Yeah, will bring up this VM thing in the conversation and let’s see what they say.

1 Like

Hello, Andy here from Netdata.

I think pointing Netdata Prometheus collector at the metrics endpoint should for sure be an easy way to pull the metrics from the metrics endpoint into Netdata

https://learn.netdata.cloud/docs/agent/collectors/go.d.plugin/modules/prometheus

I need to learn a bit more about LXD VM’s and how may or may not be different than normal VM’s but just wanted to mention the /metrics approach as one potential quick solution.