kafka-devops

What This Skill Does

The kafka-devops skill acts as a specialized Site Reliability Engineer (SRE) for Apache Kafka clusters. It leverages deep domain knowledge to assist with infrastructure as code, cluster configuration, performance tuning, and incident management. Whether you are deploying on Kubernetes using Strimzi, managing traditional virtual machines, or configuring Managed Kafka services, this agent provides actionable insights. It excels at diagnosing consumer lag, optimizing broker partitioning strategies, automating CI/CD pipelines for topic management, and establishing robust monitoring alerting rules.

Installation

To integrate this skill into your OpenClaw environment, execute the following command in your terminal or terminal-integrated agent console:

clawhub install openclaw/skills/skills/anton-abyzov/sw-kafka-devops

Ensure that your OpenClaw agent has the necessary permissions to access your cloud environment or Kubernetes configuration files if you intend for it to perform direct infrastructure modifications.

Use Cases

• Cluster Sizing & Capacity Planning: Use the agent to calculate partition counts based on throughput requirements and retention policies.
• CI/CD Integration: Automatically validate and apply topic configurations, ACLs, and schema registry updates via GitOps workflows.
• Incident Response: Analyze diagnostic logs to identify bottlenecks in ISR (In-Sync Replicas), troubleshoot under-replicated partitions, or debug consumer group rebalancing issues.
• Performance Tuning: Gain recommendations on JVM heap settings, disk I/O optimization for log segments, and network throughput adjustments.

Example Prompts

1. "Analyze my current Kafka cluster configuration and recommend partition counts for a high-throughput topic expecting 500MB/s ingestion with a replication factor of 3."
2. "Draft a Terraform module for a Strimzi KafkaCluster resource that implements tiered storage and enables JMX exporter metrics."
3. "I am seeing frequent rebalances in my consumer group. Write a script to analyze consumer lag and provide a summary of the worst-performing partitions."

Tips & Limitations

• Context is Key: Always provide relevant configuration snippets (like server.properties) when asking for specific tuning advice.
• Safety First: When using the agent for configuration changes, always request a dry-run or a Terraform plan output before applying changes to production clusters.
• Limitations: The agent provides advisory information and code generation. It does not replace the need for physical monitoring tools like Prometheus or Confluent Control Center. Ensure all suggested configuration changes are validated in a non-production staging environment first.