Is anyone here actually running kubernetes on their LXC containers?

[esi_y]

Well for example, if we compare to OpenShift or OpenStack they are huge projects kind a working on a different approach to the same problem PVE is.

They don't work on the same problem at all. Different approach is e.g. k8s. It's scalable approach which PVE is not.

In my point of view some things like Multi Cluster integration and HA between Geo locations can be done with K8S for example, but then everything comes as a performance problem with VMs and etc. Then why to even use PVE? Why not just a Pure K8S batemetal?

Precisely because PVE is for the small-scale. For those that build large scale will do k8s bare metal. That is, if your architecture is k8s ready.

BTW if you check the latest feature request in this regard (multi geo location) is something requested 5 to 6 years ago. It has been a long time ago And people are still working in a beta release

Because it does not suit well at all the original philosophy which became a baggage. Every other solution has some sort of monitor node, except PVE wanted to make the nodes equal, relay to each other, etc. Then the limitation to realistic number of nodes per cluster. And pretend it's "multi-geo location" when it will in reality be multi-cluster single control plane (and relaying). It scales badly and it becomes like all the other solutions (which have an advantage with the more traditional architecture).



I will leave the rest ... I guess it was for @Tekuno-Kage to react on. But the community is not the steering committee. In fact, I did not even see any sort of statement on governance from Proxmox. So the forum is like pitching playground. The developer mailing list looks like overprotective parenting group at times. Nothing too radical, as in ... not the slightest even.

 

[Tekuno-Kage]

@dietmar , @tom , @fabian , @aaron​

Hi guys, I reached out to you because I already interacted with you on other treats. Some of us, like @tempacc346235 and myself, benefit from it. The suggestion gets some detractors, too, like @navigator. Can you bring some guidance?

Also, we can start a new thread if it seems better. I looked on the forum to see if someone mentions Kata Containers anywhere else but found no reference.

Does the proposed component for improving Proxmox have merit, or should it be ignored? In this case, there are some merits in how the community could assist.

Regards,

 

[esi_y]

@Tekuno-Kage I don't think you will get the desired reaction by referencing me in this one. I have often different ideas no how to do things here and my style of pitching it is considered somehow undesirable by PVE staff. As in, I am often saying "this or that is silly way to do it" and people hear I call them silly, which I do not. If anything, it will give the post of yours a bad vibe during considering it. You may want to open a new thread with the name-dropping.

But I liked your find!

 

[fabian]

we aren't currently looking at integrating yet another container management stack or runtime. we did evaluate "microvms" a while back (which is a similar concept to katacontainers, but within Qemu), and that might still be interesting to fully spec out.