Apache NiFi™ on the
Stackable Data Platform

The supported Apache NiFi versions – covering both 1.x and 2.x lines – are listed on the Stackable documentation’s supported-versions page, updated each release. Pick the version that fits your existing workflows by setting the image in your NifiCluster, or use a custom image for experimental setups.
Click here for the current list.

Scaling is handled declaratively by adjusting the replica count per role-group in the NifiCluster CRD. The operator automatically provisions new pods, updates configurations and ensures that services remain consistent during scale-up or scale-down events. This accommodates higher data volumes or more concurrent flows without downtime for the rest of the cluster (except for NiFi 1.x upgrades, which may require a restart).

NiFi 2.x supports rolling upgrades, with nodes updated sequentially without stopping the entire cluster – the operator handles pod restarts, configuration updates and flow-state migration, minimizing disruption to running workloads. NiFi 1.x requires a cluster restart for updates, as leader election and internal state management necessitate full cluster cycling.

Security is layered: SBOM verification and signed container images guarantee trustworthy, auditable deployments. TLS encryption secures communication between nodes and clients. Integration with OIDC/LDAP lets organizations enforce enterprise authentication standards, and optional OPA policies enable fine-grained, policy-based access control. Together, this makes NiFi deployments secure and suitable for regulated environments.

Yes. Stackable provides SBOM (Software Bill of Materials) verification, signed container images and support for air-gapped deployments, ensuring that clusters can operate securely without external internet access. This is crucial for highly regulated industries or environments with strict network isolation policies.

The operator has been tested on major managed and self-hosted Kubernetes platforms: EKS, AKS, GKE, OpenShift, IONOS and K3s. This flexibility lets you deploy NiFi reliably across cloud providers or on-premise infrastructure, while still benefiting from declarative management and operator automation.

Connections are configured through NiFi processors in the NiFi web UI. The operator simplifies deploying clusters with preconfigured connectors for Kafka, S3/MinIO, databases and other systems. This makes it easy to integrate NiFi into an existing data ecosystem without writing custom code – while keeping central control over your flows.

Absolutely. The operator allows you to override almost all NiFi configuration settings via the CRD, including JVM options, cluster properties, processor defaults and security settings. This makes it possible to tailor deployments to specific performance, compliance or operational requirements.

Yes. NiFi exposes Prometheus metrics for all cluster components. These can be consumed by existing monitoring and alerting systems to track processor throughput, queue sizes, JVM performance and overall cluster health. Stackable also documents best practices for integrating metrics with Grafana dashboards for real-time observability.

Yes. You can define multiple NifiCluster CRDs in the same Kubernetes namespace or across namespaces. Each cluster runs independently, allowing isolation between environments (e.g. dev, staging, production) or multi-tenant architectures while using the same operator.