What will happen to vSphere Integrated Containers?

[vitaprimo]

Hi all,

I've been trying for days to deploy Kubernetes on vSphere but I keep failing--I mean--I keep failing getting a deployment where I feel confident enough to deploy from scratch without following some guide like I can do with every other thing in the infrastructure. It's just too complicated and, since I haven't run Kubernetes anywhere ever, to me the goal is the same as vSphere Integrated Containers' but with more steps. I know it's more than that but I only need an easy way to manage containers without having to think where to stick 'em and optionally can show status on vCenter/Fusion/ESXi.

I'd like to keep using VIC but on its last release, around the time Kubernetes on vSphere was released, things seem to just have stopped plus it always felt like "in the works" to begin with, so I'm unsure if it's a good idea or not. I can't find anything that mentions both VIC and Kubernetes (vSphere 7) anywhere to find some sort of progression/pivot there.

Was VIC abandoned? I know it's may be kept alive by the open source community but it might as well flocked away to Kubernetes because it's the hot new thing of the same interest. Was there a statement or something like that? I don't really follow this and I miss a lot of stuff--for instance, this year I learned I had included access to the newest Log Insight and I was still using 4-something.

Thanks! :smileygrin:

[daphnissov]

To directly answer your question first, this is a futures state which we can't answer here. And if we did know, we probably could say in a public venue. That said, my *unofficial thought* is that VIC is probably not going to be developed much further in the future. This is because ultimately there is very little value in running "naked" containers directly without some orchestration system. And at this point, there are so many solid ways of just running containers that it doesn't make sense to keep developing one very niche solution. Kubernetes is a layer that spans this so it handles those containers. So your real goal, once you've learned about containers themselves with something like Docker Desktop, should be to learn K8s. If you're having trouble with the K8s aspect and want assistance, just ask here and we can assist. But we need to know your goals, what you have to work with, and where *specifically* you're falling down.

[vitaprimo]

Yeah, it makes sense and it sucks.

I appreciate the help but I really want to solve (or not) problems on my own. However, I do want to thank you for the advice, I had just sort of given up but this all got me motivate and tried again and got pretty close until I ran into outdated mismatching versions. It was fun and I learned some new tricks. I connected the nodes among themselves using Cockpit and accidentally stumbled into another possible orchestration solution.

I also learned using the vSphere integration would lock me from mundane things like renaming a datastore, shutting down vCenter, who knows what would happen is vSAN misbehaves, and then there's the specific version matching. There's way to many variables that ironically aren't variable (dynamic)--unlike basically anything else on vSphere, these havet to be pre-set, never changed after the fact because a little change somewhere efs up things massively. It's like deploying a Root CA, but once again…with more steps.

I'll just have to make do with vSphere-detached Kubernetes when I get there or *aaS from Red Hat's OpenShift, or maybe it has vSphere integration has matured by then.

Thanks!

[magnusviri]

I kind of wish something was said public about this. I just got interested in containers and am trying to learn everything and I thought VIC was a viable solution. But I can't get my VCH installed (some config weirdness) and in the process of searching for an answer I finally discover this thread and I eventually found the contributions to VIC is pretty much dying out.

https://github.com/vmware/vic/graphs/contributors

I could've saved a lot of time if I had known VIC is basically done. 😞