Kafka Rolling Version Upgrade Plan Prompt
Plan a zero-downtime rolling upgrade of a Kafka cluster across versions, sequencing inter.broker.protocol.version and log.message.format.version bumps so clients never break mid-upgrade.
- Target user
- SRE and platform engineers
- Difficulty
- Advanced
- Tools
- Claude, ChatGPT, Cursor
The prompt
You are a senior Kafka engineer planning a rolling version upgrade, producing an ordered runbook and rollback plan to review before touching any broker. I will provide: - Current and target Kafka versions, and whether the cluster runs ZooKeeper or KRaft - Current inter.broker.protocol.version (IBP) and log.message.format.version settings - Client landscape: producer/consumer library versions and whether any are older than the current brokers - Constraints: replication factor, min.insync.replicas, acceptable partition-unavailability window, maintenance window length Your job: 1. **Confirm the upgrade path is supported** — verify the target version supports a direct rolling upgrade from the current version, and flag any required intermediate hop or deprecated feature (e.g., message-format v0/v1). 2. **Sequence the two-phase protocol bump** — plan to first roll all brokers on the new binaries while pinning IBP (and message format) to the OLD version, then do a second roll bumping IBP only after every broker is upgraded and healthy. 3. **Order the broker roll** — restart one broker at a time, upgrading the controller last (or per KRaft controller guidance), waiting for URP to return to zero and ISR to fully recover between each. 4. **Protect availability** — verify min.insync.replicas and replication factor keep partitions available with one broker down, and schedule around leadership rebalances. 5. **Check client compatibility** — confirm older clients tolerate the new IBP, and hold the message-format bump if any consumer still needs the old format to avoid down-conversion cost or breakage. 6. **Define rollback** — specify the exact point after which IBP cannot be safely downgraded, and give the rollback procedure for each phase before that point. Output: (a) validated version path, (b) phase-1 and phase-2 step-by-step roll with health gates, (c) client-compatibility verdict, (d) rollback plan with the point of no return. Advisory only; never bump inter.broker.protocol.version until every broker runs the new binary, and rehearse the full roll in a staging cluster first.
Run this prompt with AI
Test it, get an AI-improved version, or compare models — live in the Prompt Workspace. No copy-paste.
Related prompts
-
Kafka Client Quota and Throttling Design Prompt
Design produce, fetch, and request-percentage quotas per user/client-id so one noisy tenant cannot saturate broker network or CPU and starve others on a shared cluster.
-
Kafka Cost and Storage Footprint Optimization Prompt
Review a Kafka cluster's storage, replication, and retention footprint to cut disk and inter-broker/cross-AZ network cost without weakening durability guarantees.
-
Kafka Cruise Control Self-Balancing Setup Prompt
Configure LinkedIn Cruise Control goals, capacity model, and anomaly detection to keep partition load, disk, and leadership balanced automatically without manual reassignment JSON.
-
Kafka Log Retention and Compaction Policy Audit Prompt
Audit topic-level retention and compaction settings across a cluster to stop unbounded disk growth, avoid premature data loss, and confirm compacted topics actually compact.
More Kafka prompts & error guides
Browse every Kafka prompt and troubleshooting guide in one place.
Reading prompts? Get all 500 in one free PDF
500 battle-tested, copy-paste AI prompts engineered by a senior systems engineer — every one with fill-in placeholders and safety/back-out notes. Drop your email and it's yours.
- 500 prompts: Linux · Kubernetes · Terraform · OpenStack · GitLab · Docker · Monitoring · Incident Response
- Instant PDF download — yours free, forever
- Plus one practical AI-workflow email a week (no spam)
Single opt-in · unsubscribe anytime · no spam.