Pulsar Engine Entry
Use this as the primary create-time entry for Pulsar. Tier-1 Pulsar never falls back to kubeblocks-engine-generic.
Cold-Start Contract
- Run kubeblocks-preflight whenever the environment is not already profiled.
- Treat engine-create-matrix as the create-time truth for topology, version strategy, sizing, validation, next hops, and forbidden routes.
- Use observability-capability-matrix before promising monitoring readiness.
- Keep Pulsar as a broker-plus-storage platform path, not as a generic streaming placeholder.
Topology Selection
- Default topology:
basic basic: default path with brokers, bookies, and ZooKeeper, suitable for most first deployments.enhanced: use when the rollout explicitly needs proxy and bookies-recovery components.external-zookeeper: use only when the platform already has a trusted ZooKeeper service reference and the user explicitly wants to reuse it.- If the user cannot justify extra components or external coordination, stay on
basic.
ServiceVersion / Version Strategy
- Strategy: use the stable addon default from current examples unless the user explicitly pins a serviceVersion
- Current addon evidence surfaces Pulsar
2.11.2and3.0.2. - If the user asks for "latest", stay on an addon-backed supported line rather than a raw upstream tag.
Preflight Interpretation
storage_class: required before apply because brokers and bookies persist to PVC-backed storage.volume_binding_mode: confirm storage placement before multi-component rollout.addon_readiness: Pulsar addon must be installed first.dependency_clusters: if the plan reuses an external ZooKeeper path, validate those service references before apply.
Sizing Profiles
demo:basicwith small broker and bookie footprints for evaluation.production:basicorenhancedwith explicit storage, monitoring, and component-count decisions made up front.- Treat bookie count and broker count as real durability and throughput decisions, not just cosmetic replica numbers.
Minimal Create Path
apiVersion: apps.kubeblocks.io/v1
kind: Cluster
metadata:
name: <cluster>
namespace: <ns>
spec:
terminationPolicy: Delete
- Do not leave this skill to read raw addon examples before drafting the manifest.
- Set
clusterDef: pulsarfor every Pulsar create path in this wave. basic: render the addon topology astopology: pulsar-basic-cluster, keepname: broker,name: bookies, andname: zookeeper, and size broker/bookie storage directly.enhanced: render the addon topology astopology: pulsar-enhanced-cluster, keep thebasiccomponents, and addname: proxyplusname: bookies-recovery.external-zookeeper: keepclusterDef: pulsar, use external service references for ZooKeeper, and do not leave both a localname: zookeeperquorum and an external ZooKeeper contract in the same manifest.- Validate with
kubectl apply --dry-run=server -f <pulsar-cluster.yaml>. - Apply with
kubectl apply -f <pulsar-cluster.yaml>. - Watch
kubectl get cluster <name> -n <ns> -wuntil the phase isRunning.
Connection and Validation
- Supported connection methods:
in-cluster service,port-forward,nodeport-or-loadbalancer - First validation step:
kubectl get cluster <name> -n <ns>and wait forRunning. - Default validation: verify broker or proxy service readiness and use Pulsar admin or client tooling before handoff.
- If the rollout uses proxy, validate clients through the proxy path, not only broker-local connectivity.
Next Hops
- Day-2 operations currently route to
kubeblocks-op-lifecycle,kubeblocks-op-horizontal-scale,kubeblocks-op-reconfigure, andkubeblocks-observability-router. - Treat credential and TLS work as engine-specific validation rather than assuming kubeblocks-manage-accounts or kubeblocks-configure-tls are default Pulsar paths.
- Do not send this path to kubeblocks-rebuild-replica; Pulsar recovery should stay on troubleshoot or engine-specific repair workflows instead.
- Route broken service references, broker startup failures, or bookie issues to kubeblocks-troubleshoot.
Forbidden Routes
- Never route Pulsar create through
kubeblocks-engine-generic,kubeblocks-family-streaming,kubeblocks-engine-kafka, orkubeblocks-engine-rabbitmq. - If the request is really Kafka-style log streaming or queue-style RabbitMQ semantics, switch engines before apply.
Evidence Anchors
- Use the evidence anchors below only as optional secondary evidence and parity checks after the manifest is already drafted here. A cold-start runtime should not need these files, but a stronger agent may inspect them when available.
- Current addon evidence:
examples/pulsar/cluster-basic.yaml,examples/pulsar/cluster-enhanced.yaml,examples/pulsar/cluster-service-refer.yaml.