Integration · Data streaming
Apache Flink + OrchestrAI
Catalog exported 2026-09-02 · Apache Flink website
Submit, list, and cancel Apache Flink jobs from chat with cancellation gated by confirmation.
OrchestrAI exposes 3 Apache Flink operations: 1 is low-risk (read-only or low-impact), and 2 create or modify resources and run only after you confirm the plan. 1 of them also carries a step-level approval gate.
What teams use it for
Streaming teams ask OrchestrAI to submit a Flink job to the cluster, list what is running, and cancel a job that is lagging behind its source. Listing jobs is read-only, submitting is medium risk and runs on request, and cancelling is confirmed because it stops in-flight processing. Savepoints, checkpoints, job restarts, and metrics have no operation yet, so those still go through the Flink REST API or dashboard.
Every Apache Flink operation, with its risk level
| Operation | What it does | Risk | Step-level approval |
|---|---|---|---|
List Flink Jobs |
List Flink jobs | Low risk | No |
Cancel Flink Job |
Cancel Flink job | Creates resources | Yes |
Submit Flink Job |
Submit Flink job | Creates resources | No |
Risk tiers come from the catalog: low is read-only or low-impact, medium creates resources and is reversible, high modifies existing resources, destructive may lose data. Every plan that creates or changes resources is shown with its cost estimate and waits for your confirmation. Operations marked with a step-level approval pause again on their own step. Destructive operations require a typed risk phrase.
What you connect
A Apache Flink credential (stored as flink).
Connected-service tokens are envelope-encrypted with a per-record key wrapped by a cloud KMS.
Prompts that work
- List all running Flink jobs and their uptime
- Submit the clickstream-enrichment jar to the Flink cluster with parallelism 8
- Cancel the Flink job named sessionization-v2
Before anything runs
Every mutation shows its plan, cost estimate, and blast radius, then waits for your confirmation. Destructive operations require a typed risk phrase. Credentials are minted per run through OIDC federation and discarded afterward; nothing you create here is invisible later, because every resource lands in the desired-state ledger where drift is detected and can be converged. Details on the security page.
Frequently asked questions
- Can OrchestrAI take a Flink savepoint before cancelling a job?
- No. There is no savepoint operation. Cancel stops the job and waits for confirmation, so trigger a savepoint yourself first if you need one.
- How many Flink operations does OrchestrAI support?
- Three: list jobs, submit job, and cancel job. Listing is low risk, submit is medium, and cancel is medium with confirmation.
- How does OrchestrAI authenticate to Apache Flink?
- You add a Apache Flink credential once in the connections screen. It is envelope-encrypted with a per-record key wrapped by a cloud KMS and is only decrypted inside the run that needs it.
Related integrations
Try it on your own account
Connect your cloud read-only and see your resources, drift, and costs before anything runs. $5 minimum to start. Unused credits refunded in your first 14 days.
Unused credits refunded in your first 14 days.