dagster-cross-job-to-orchestra

Dagster Cross-Job Triggering -> Orchestra

Overview

Dagster cross-job patterns — @run_status_sensor launching a downstream job, @asset_sensor on an upstream asset, and cross-code-location asset dependencies — all have Orchestra equivalents. Orchestra provides two mechanisms:

1. TRIGGER_PIPELINE task — an explicit pipeline step that triggers another pipeline and waits for it to complete. Use when one pipeline must trigger another mid-flow and wait.

2. trigger_events: block — event-driven triggering at the pipeline root. When an upstream pipeline completes, this pipeline starts automatically. This is the cleaner equivalent of @run_status_sensor / @asset_sensor.

---

@run_status_sensor (SUCCESS, launching a job) -> trigger_events: (preferred)

# Dagster
@run_status_sensor(
    run_status=DagsterRunStatus.SUCCESS,
    monitored_jobs=[nightly_elt],
    request_job=daily_report,
)
def trigger_report_on_elt_success(context):
    return RunRequest(run_config={"ops": {"build": {"config": {"env": "prod"}}}})

# Orchestra — daily-report pipeline
version: v1
name: daily-report

trigger_events:
  - type: pipeline
    pipeline_id: "uuid-of-nightly-elt-pipeline"
    statuses: [SUCCEEDED, WARNING]
    run_inputs:
      env: prod

pipeline:
  ...

TriggerEventModel fields:

Field	Required	Notes
type	yes	Always pipeline
pipeline_id	yes	UUID of the upstream pipeline, or "*" for any
statuses	no	Default [SUCCEEDED, WARNING]
run_inputs	no	Inputs passed to the triggered run

Multiple entries = OR logic (fires when any upstream completes).

---

Explicit launch-and-wait -> ORCHESTRA + TRIGGER_PIPELINE

When a pipeline must trigger another and wait before continuing:

task-001:
  integration: ORCHESTRA
  integration_job: TRIGGER_PIPELINE
  name: trigger_reporting
  parameters:
    pipeline_id: "uuid-of-daily-reporting-pipeline"
    run_inputs:
      env: prod
    branch: null
  depends_on: []
  condition: null
  tags: []