Oh i have spent years looking.
ProxMox is probably the closest option, but has no multi-clustering
support. The clusters are more or less isolated from each other, and
would need another layer if you needed the ability to migrate between
them.
Also been looking at ProxMox for ages. We were using OpenVZ in
compliance-heavy production so a nice GUI and a VM option seemed ideal. But
SWsoft/Parallels/Virtuzzo and ProxMox were competing not collaborating and diverging with
distinct hypervisors and IaaS containers, which seems a silly tribal squarrel in face of
today's cloud invasion. Never thought LXC might do better than OpenVZ (or I might
return to Xen from KVM). Redhat fought both IaaS containers with nothing but VMs and only
got saved by (PaaS) Docker, which they then tried to smother with podman and Kubernetes.
But Proxmox is not HCI or only via DIY.
XCP-ng, cool. No spice support. No UI for managing clustered storage
that is open source.
true and the most attractive option seems a paid upgrade (and
not ready yet)
Don't think I'd miss SPICE or that it has much of a future.
But nothing HCI has ever deployed so quickly and easily, including vSphere as the supposed
market leader.
Harvester, probably the closest / newest contender. Needs a lot more
attention / work.
It looks great, thanks for the tip. No idea if they survive the
next three months.
I am adding the link here, because that name needs to be searched right ;-)
https://harvesterhci.io/
OpenNebula, more like a DIY AWS than anything else, but was functional
last i played with it.
Lots of material, but is it actually open source? And do
they have a free tier? They seem to have everything... which makes me more suspicious than
happy these days.
Has anyone actually played with OpenShift virtualization (replaces RHV)?
Wonder if OKD supports it with a similar model?
The oVirt team had hinted at an
integrated container VM solution when I started with oVirt. Obviously the pods and
etc-daemons need to run somewhere and VMs or IaaS containers like OpenVZ or LXC would be a
good start. That never materialized and evidently they chose to abandon IaaS and HCI
completely.
But there is plenty of workloads still out there, that are more comfortable with a IaaS
abstraction and more concerned with scale-in than scale-out. Or which live at the real
edge, in the field, on tracks or roads or in the middle of an ocean.
I don't quite see how OpenShift replaces RHV, especially not at the [real] edge. The
software industry may be transitioning towards cloudy application models, but it's
note quite all there yet.
My impression is that Redhat is very carefully avoiding a CentOS repeat for every other
open source project they do. Their upstream variants seem far more beta in OKD and Kubvirt
than oVirt ever was.
"No Free Lunch" seems to have been chiseled into the three letters of their new
owners for almost a century now.