Hi Ansible community proxmox maintainers, we are working on a new certified ansible.proxmox collection.
First, I want to thank you for your hard work maintaining the community.proxmox collection over the years. Your efforts have provided a critical bridge, and the collection has been an invaluable resource for users automating their Proxmox environments.
I am reaching out to give you an early heads-up that Red Hat Engineering is currently developing and certifying a new, standalone ansible.proxmox collection.
I want to be completely transparent right upfront: our goal is absolutely not to step on your toes or push the community project out. We are building this specifically to provide official, SLA-backed support for our enterprise Ansible Automation Platform (AAP) customers who require a certified path for Proxmox automation.
Because Red Hat is taking on this official maintenance burden, we want to coordinate with you to ensure a smooth transition for existing users and avoid fragmenting the ecosystem. We would love to collaborate on this handoff in whatever way makes the most sense for you. A few potential paths:
Graceful Deprecation & Redirection: If you agree that the Red Hat maintained collection should be the path forward, our engineering team can submit the PRs to your repository to set up the module routing stubs and deprecation warnings. This allows existing playbooks to seamlessly redirect to the ansible.proxmox namespace without breaking user environments.
Feature Parity & Certification Requests: Please let us know if there are specific modules or features from your collection that you would like to see certified. We want to make sure we are prioritizing the automation pieces that the community relies on most.
Open Invitation to Contribute: We would be incredibly grateful for any code, feedback, or guidance you would like to contribute to the new repository. Your domain expertise here is unmatched, and we would welcome you as contributors to the new upstream project.
Our priority is making sure this is a positive transition that respects the foundation you’ve built, while giving the community an open, vendor-supported tool moving forward.
Let us know how you’d like to proceed. If you have time for a quick call to align on the details, we would be happy to set that up, or we can simply map out the next steps asynchronously here.
Thanks for the heads-up! It’s good to see that Proxmox’ getting more traction up to the point it’s worth considering a certified, Red Hat supported collection!
I’m not the sole maintainer (I am the longest, but @IamLunchbox@PendaGTP have been way more productive then I have in making the improvements happen ), but I appreciate the offer and I am willing to have a serious look at it.
It would require quite a bit of a gap analysis, as I expect some modules in the collection right now may or may not be suitable for a RH-supported collection directly. Not because they offer non-standard PVE functionality, but some of them (community.proxmox.proxmox for example) provide a myriad of functions that would (hindsight 20/20) be better off being split up in multiple modules.
What might be a useful path forward would be:
Check with the other maintainers and members of the forum if they agree with merging (see poll below)
Inventory modules that either have (long) outstanding refactors that would fit in the RH collection. This could potentially mean that the modules can be deprecated and removed instead of refactored.
Check for modules that do not fit in the RH collection and keep those in the current setup
My answer generally would be yes, but my biggest concern would be maintaining feature parity. I’ll have to look at all the things I use (its probably the standard things though) to give a full answer, but as long as there is enough in the new collection, it will be fine.
Hi Steve,
thank you for reaching out to us. I think this is a good path forward. Our codebase has grown organically and a refactor is urgently needed, while there is no maintainer with enough time budget to do so.
Nevertheless, as far as I can tell there has already been some effort in your module since the beginning of october and the module does not mention its development status or any limitations. This puts your public collection in direct competition to ours and I wouldve preferred, if Red Hat Engineering woudlve reached out to us before it publishes the collection and bypasses / replaces our efforts. In other OSS communities this could be considered a hostile fork - I personally don’t think it is, since the refactoring and certification needs are crucial.
In the end I suppose most users will prefer the official collection and I believe rightly so. That makes it even more crucial to talk to the collection maintainers, which gets superseeded, early on. Reaching out after the competion / replacement was published and asking if and how we want to do a transition certainly has a taste to it.
Regarding your request:
Graceful Deprecation & Redirection: I think this is a good idea! If possible, it would be great there is transparency about the feature parity and limitations of the replacing module when we redirect.
Certification Requests: I think the most interesting modules will be vm management itself and inventory mangement. I, personally, think managing the PVE firewall is important as well and a complex undertaking, but I don’t know how many users actually utilize the proxmox firewall.
I am more then happy to further discuss the transition and look forward to work together in that matter