parlel/emr

A zero-dependency, in-process fake of AWS EMR (Elastic MapReduce). Speaks the AWS JSON 1.1 wire protocol, so the real @aws-sdk/client-emr works against it unchanged.


Port	`4709`
Protocol	AWS JSON 1.1 (`X-Amz-Target: ElasticMapReduce.<Op>`)
Health	`GET /_parlel/health`
Reset	`POST /_parlel/reset`

Default connection

AWS_ENDPOINT_URL=http://127.0.0.1:4709
AWS_ACCESS_KEY_ID=parlel
AWS_SECRET_ACCESS_KEY=parlel
AWS_REGION=us-east-1

Any SigV4 credentials are accepted (auth is not verified).

Supported operations

Category	Operations
Clusters	`RunJobFlow`, `ListClusters`, `DescribeCluster`, `TerminateJobFlows`
Steps	`AddJobFlowSteps`, `ListSteps`, `DescribeStep`

Generated ids: j-… (clusters / job flows), s-… (steps).

SDK usage example

import { EMRClient, RunJobFlowCommand, ListStepsCommand } from "@aws-sdk/client-emr";

const emr = new EMRClient({
  endpoint: "http://127.0.0.1:4709",
  region: "us-east-1",
  credentials: { accessKeyId: "parlel", secretAccessKey: "parlel" },
});

const r = await emr.send(new RunJobFlowCommand({
  Name: "etl-cluster",
  ReleaseLabel: "emr-7.1.0",
  Instances: { InstanceCount: 3, KeepJobFlowAliveWhenNoSteps: true },
  Applications: [{ Name: "Spark" }],
  Steps: [{ Name: "ingest", HadoopJarStep: { Jar: "command-runner.jar", Args: ["spark-submit", "job.py"] } }],
}));
const steps = await emr.send(new ListStepsCommand({ ClusterId: r.JobFlowId }));
console.log(steps.Steps[0].Status.State); // "COMPLETED"

Access via MCP / preview URL

When running inside a Daytona sandbox, this HTTP service is exposed at an automatically-provisioned preview URL. Point the SDK endpoint at that URL and add the x-daytona-preview-token header on requests.

Surface coverage

This emulator faithfully replicates the API surface most application code and agents exercise. Anything below the supported lines is either an intentional design choice for a fast, zero-cost local emulator (✓ By design) or a candidate for a future release (⟳ Roadmap) — never a silent inaccuracy.

Legend: ✅ fully supported · ◐ accepted (stored, not strictly enforced) · ✓ by design · ⟳ on the roadmap.

Area	Limitation
Execution	Nothing runs — clusters land in `WAITING` and steps are immediately `COMPLETED`.
Instances	Instance groups/fleets are recorded opaquely; no real EC2 provisioning.
Step types	All steps are treated as Hadoop JAR steps; logs and stderr/stdout are not captured.
Auto-scaling	Managed scaling, instance fleets, and bootstrap actions are not modeled.
Auth	SigV4 is accepted but never validated.
Persistence	In-memory; lost on restart/reset.

Configuration — `test.env`

Copy these into your test.env (used by the bridge sidecar flow). Tokens are Parlel's seeded test credentials — any non-empty value is accepted by the emulator, so you rarely need to change them. Swap in real credentials only when pointing at the live service in prod.env.

AWS_ACCESS_KEY_ID=parlel
AWS_SECRET_ACCESS_KEY=parlel
AWS_REGION=us-east-1
AWS_ENDPOINT_URL=http://parlel-bridge:4709