I’m not saying you can’t run production on it, after all RHV was based on it. What I’m saying is that it’s a run at your own risk because you can’t buy support. Some companies are okay with taking on that risk, others aren’t.
You should be automating that stuff anyway. Make a template that has the drivers installed, or write up some Ansible that does the install for you. It’s annoying, but it shouldn’t be a problem once you have an automation pipeline in place.
Well our red hat account manager said
Well, they’re wrong. It has been clearly communicated that OpenShift Virt is not a replacement for RHV. It does not do the enterprise things like having concepts of dataceners, etc… It is to help people migrate to containerization.
That’s just sales being sales. Just because you can use a hammer to drive a screw doesn’t mean it’s the right tool for the job. Which is clearly the conclusion you already came to.
red hat virt is being deprecated soon, to be replaced by ovirt which is def not prod ready
Not exactly. oVirt is the upstream open source project that Red Hat Virtualization was based on. It is never and will never be “production ready.” Because they don’t sell support for it.
Red Hat does not have a replacement for RHV. You can use OpenShift Virtualization to make virtual machines, but it’s not designed to be a replacement for RHV or a competitor to VMware. It is designed to be a stepping stone for people looking to containerize their workloads or keep that one legacy app around that can’t be containerized. You might be better served by OpenStack, but that’s an entire cloud orchestration tool, and the next version will require OpenShift to host the control plane.
Nutanix is probably the best competitor to VMware. Proxmox is another solution that would be great for small to medium businesses. SUSE’s Rancher team is working on Harvester, which could become something that competes. Honestly, there isn’t a lot of competition in this market because it’s not where growth is. People are moving to containers and kubernetes. Even VMware knows it and they are playing catch up with Tanzu.
Source: I’m a consultant for Red Hat.
That’s a poor excuse. If something is secret or higher it has a different TLD. The SIPRnet uses .smil for example. There are also tools at the boundaries that don’t allow going from SIPR to NIPR unless they meet specific criteria. Basically you can only leak those secrets accidentally if they were already on a system they shouldn’t have been on.
And you spread that server load by selecting different servers. While what you’re saying is technically true, in a practical sense if everyone picked a more local server that would be one way to achieve what you’re saying.
Ansible is my specialty. It’s kind of hard to get in the mindset of declarative after all the years of writing imperative scripts. Once you do, it’s amazing.