skills.rest
sakicodes
sakicodesCOMMUNITY
Explore all skills by this author →
GitHub

developing-applications-on-managed-service-for-apache-flink

Guides development, deployment, and operations of Apache Flink applications on Amazon Managed Service for Apache Flink.

Updated Jul 1, 2026
Search Tags:#aws#stream processing#iceberg#cdc#kinesis#apache flink#managed service for apache flink
One-click install
npx skills add https://github.com/sakicodes/BuildFestHackathon26 --skill developing-applications-on-managed-service-for-apache-flink-sakicodes
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: developing-applications-on-managed-service-for-apache-flink
Source: https://github.com/sakicodes/BuildFestHackathon26/tree/main/.agents/skills/developing-applications-on-managed-service-for-apache-flink
Command: npx skills add https://github.com/sakicodes/BuildFestHackathon26 --skill developing-applications-on-managed-service-for-apache-flink-sakicodes

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Generic Flink knowledge leads to incorrect answers for Amazon Managed Service for Apache Flink (MSF) because the service has unique constraints: the KPU resource model, prohibited checkpoint and parallelism configuration in application code, the kinesisanalyticsv2 vs kinesisanalytics identifier split across CLI, IAM, and CloudWatch, two-phase IaC deployments, and snapshot lifecycle rules. This Skill supplies the MSF-specific guidance needed to answer and build correctly. ## Core Features & Use Cases - MSF-Aware Development Guidance: Enforces correct patterns such as no checkpoint or parallelism config in app code, operator UID discipline, and Secrets Manager-based CDC credentials. - Connector and Integration References: Covers Kinesis (polling and EFO), Kafka/MSK, Iceberg sinks and tuning, and CDC sources for MySQL, PostgreSQL, Oracle, SQL Server, and MongoDB with version-correct Maven coordinates. - Operations and Troubleshooting: Provides runbooks for application lifecycle, safe deploys with rollback paths, KPU sizing, checkpoint tuning, restart-loop diagnosis, monitoring, and Flink 1.x to 2.x migration. - Use Case: A developer asks how to deploy a Flink CDC pipeline from Aurora MySQL to Iceberg on MSF. The Skill loads the CDC connector guide, Iceberg guide, dependency management, and IaC references to produce correct code, pom.xml, IAM policies, and a two-phase deployment plan. ## Quick Start Ask how to build, deploy, size, or troubleshoot an Apache Flink application on Amazon Managed Service for Apache Flink and the relevant reference guides will be loaded before answering.

Frequently Asked Questions about developing-applications-on-managed-service-for-apache-flink

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I deploy code updates to a Managed Service for Apache Flink application?▼

Upload the new JAR to a versioned S3 key, fetch the current ApplicationVersionId, then call update-application pointing at the new key. From RUNNING state the app auto-restarts; from READY state you must call start-application afterward. Always pair deploy steps with rollback options such as RollbackApplication or restore from a pre-deploy snapshot.

How do I configure a Flink CDC source for MySQL on Amazon MSF?▼

Use the incremental MySqlSource from the org.apache.flink.cdc.connectors.mysql.source package with env.fromSource, since the legacy SourceFunction API is removed in Flink 2.x. Fetch database credentials from Secrets Manager at job startup, enable binlog in ROW format, and assign a unique server-id range matching your parallelism.

What is the difference between kinesisanalytics and kinesisanalyticsv2 in AWS?▼

The kinesisanalyticsv2 identifier is used only for the CLI and SDK API calls, while IAM action prefixes, Service Quotas, CloudWatch namespaces, and the trust principal all use kinesisanalytics without the v2. Mixing these up is the most common cause of permission and AssumeRole failures with MSF.

Why is my MSF application stuck in a restart loop?▼

Restart loops usually stem from customer code bugs, permission issues, snapshot incompatibility after a version upgrade, or parallelism exceeding maxParallelism. Diagnose by calling ListApplicationOperations and DescribeApplicationOperation to read statusDescription, then check CloudWatch Logs for runtime errors before retrying or rolling back.

Can I configure checkpoint intervals and parallelism in Flink application code on MSF?▼

No, MSF manages checkpoint intervals and parallelism at the service level through KPU configuration, and setting them in application code is prohibited. Configure CheckpointInterval via UpdateApplication with ConfigurationType CUSTOM, and only set operator-level parallelism when business logic specifically requires it.

Which Flink CDC version works with Flink 2.2 on MSF?▼

Flink CDC 3.6.x is required for Flink 2.2, using the per-Flink-version Maven coordinate 3.6.0-2.2. The unsuffixed 3.6.0 artifact does not exist on Maven Central, and versions 3.5.x and earlier only support Flink 1.20.

Related Skills

Explore More Software Engineering Skills→
psmfdpsmfd

aws-expert

Generates AWS infrastructure recommendations covering IAM, S3, Route 53, VPC, EKS, ECR, ECS, Beanstalk, and MSK.

Official
Advanced
heaptracetechnologyheaptracetechnology
13

Set Up CloudWatch Monitoring

Deploy and configure CloudWatch monitoring stacks for multi-service AWS environments.

Community
Advanced
richardnromanrichardnroman

developing-applications-on-managed-service-for-apache-flink

Guides development, deployment, and operation of Flink applications on Amazon Managed Service for Apache Flink.

Community
Advanced
teamniteoteamniteo
5

skill-creator

Develop, evaluate, and refine custom AI skills with iterative testing and benchmarking.

Official
Advanced