Velero is a living project, and a good operator knows which way the wind is blowing. This episode covers the Velero roadmap: kopia as the default, CSI maturity, and multi-version Kubernetes support, plus the community channels — GitHub velero-io/velero, Slack #velero, the velero.io docs, and CNCF sandbox status.

You've learned a lot of technical material — but the tool you use will change. Understanding the roadmap means understanding where this tool is going, so the architectural decisions you make today don't collide with the project's direction tomorrow. Episode 21 closes the technical discussion by looking ahead and at the people behind the project.
Driving on a new road without a map isn't enough — you also need to know who's building the road, and where they gather when something breaks. This is that "map and network".
The direction has been clear since 1.17: kopia fully replaces restic. The timeline:
The implication for you: if you still hold restic data, the migration (episode 17) is a dedicated task, not a "when I get around to it" activity. Velero also uses the Unified Repository — a single repository abstraction on top of kopia — which becomes the foundation for richer data management ahead.
CSI support has been stable since 1.14 and keeps maturing: VolumeGroupSnapshot API improvements for K8s 1.34+ compatibility, better VolumeSnapshotClass handling, and data mover pod diagnostic features. CSI-based volume backup is expected to become the main path for modern storage workloads.
Velero consistently maintains the 1.18-latest support range, testing against the three most recent K8s releases in each of its releases (1.18 tested on 1.33.7/1.34.1/1.35.0). This means: you can upgrade Velero later than K8s without losing compatibility — but still stick to a supported Velero version.
Sandbox status provides governance guarantees, but it's not a guarantee that "every feature exists". The realistic view: the core (backup/restore/migration) is mature; some areas (e.g. kopia FSB for extreme scenarios) still carry a beta label. Read the beta labels before relying on them in production.
Note
A healthy operator habit: read the release notes of every minor release. A small feature that changes behavior (e.g. the restic deprecation) is cheaper to prevent by reading than to fix after production hits it.
The development hub: source code, issues, discussions, and release notes. You can:
velero bug (episode 16) — the issue template is ready to fill.curl -s https://api.github.com/repos/velero-io/velero/releases/latest | grep tag_nameThe #velero channel on the Kubernetes Slack (slack.k8s.io) is the fastest discussion place: Q&A, release announcements, and help from maintainers. It's the first channel to open when the docs don't answer.
Complete and always-updated docs: getting started, architecture, compatibility matrix, troubleshooting, and file system backup. Make it your homepage when debugging (not random blogs — official docs reflect the latest version).
As a CNCF sandbox project, official project info is listed on the CNCF site — useful for understanding the project's position in the cloud-native ecosystem and making adoption decisions.
Contribution isn't always code. Some high-impact options:
Tip
The most valuable contributions to Velero right now: helping with restic → kopia migration and reporting kopia/CSI bugs with detail. Both align directly with the project roadmap and save other operators from the same traps.
Key takeaways:
velero-io/velero, Slack #velero, the velero.io docs, the CNCF site.In the final episode 22, we summarize the whole journey: Ecosystem, Alternatives & Final Reflection — comparing Velero vs Kasten K10 vs Stash vs CloudNativePG vs etcd backup, when to choose which, a 23-episode recap, and a complete production checklist.