Has Anyone Run KubeBlocks in Production Alongside or Instead of CloudNativePG?
Asked by
MellowCedar42
Is anyone using KubeBlocks for production workloads, particularly PostgreSQL? I'm trying to understand how it compares with CloudNativePG in practice, especially for upgrades, backups, and long-term reliability. KubeBlocks is appealing because it supports several database types that we may need, while using CloudNativePG would mean adding a separate operator just for PostgreSQL. I'd especially appreciate firsthand experience with whether its PostgreSQL support has been stable beyond simply getting the pods running, and whether its Database and DatabaseRole resources are useful in real deployments.
Answers
The biggest reason to consider KubeBlocks is its broader database coverage. If you already need it for other engines, having PostgreSQL support in the same management model could simplify the platform. The tradeoff is that CloudNativePG is much more focused, so it may offer a deeper and more mature PostgreSQL experience. I’d compare the actual features and recovery procedures rather than choosing based only on the number of supported databases.
Reply to this answer
I haven’t used KubeBlocks in production. My evaluation got as far as creating the database pods, but I didn’t gain enough confidence in the operational side to proceed. The ability to manage Database and DatabaseRole resources is appealing, though I’d want to see more real-world reports about upgrades, backups, failover, and day-two maintenance before relying on it for important PostgreSQL workloads.
Reply to this answer
CloudNativePG has been very reliable for us, so that’s the baseline I’d compare against. The main things I’d want to validate with KubeBlocks are upgrade behavior, backup and restore workflows, failure recovery, and how quickly issues get fixed. Supporting multiple database engines is a real advantage, but I’d be cautious about switching PostgreSQL workloads without testing those operational paths thoroughly.
Reply to this answer
Add your answer
Answers are reviewed before publication.
MellowCedar42: That’s exactly what I’m trying to figure out. KubeBlocks fits some of our other database needs, so using one operator would be attractive if its PostgreSQL upgrades and backups are genuinely dependable.