ZenML · Arazzo Workflow

ZenML Trace Deployment Runs

Version 1.0.0

Select a pipeline deployment, resolve its pipeline, and read the latest run it produced.

1 workflow 1 source API 1 provider
View Spec View on GitHub AIMachine LearningMLOpsLLMOpsPipelinesOpen SourcePythonArazzoWorkflows

Provider

zenml

Workflows

trace-deployment-runs
From a deployment, read its pipeline and the latest run it produced.
Lists deployments to select one, reads its pipeline, and lists runs for that pipeline to surface the latest execution.
3 steps inputs: accessToken outputs: deploymentId, latestRunId, pipelineName
1
listDeployments
listDeployments
List pipeline deployments and select the first one to trace.
2
readPipeline
getPipeline
Read the deployment's associated pipeline for naming context.
3
listDeploymentRuns
listPipelineRuns
List the runs scoped to the deployment's pipeline, newest first, to surface the most recent execution.

Source API Descriptions

Arazzo Workflow Specification

zenml-trace-deployment-runs-workflow.yml Raw ↑
arazzo: 1.0.1
info:
  title: ZenML Trace Deployment Runs
  summary: Select a pipeline deployment, resolve its pipeline, and read the latest run it produced.
  description: >-
    Traces a pipeline deployment through to the runs it has produced. The
    workflow lists deployments to select one, reads its associated pipeline for
    context, and lists the runs scoped to that pipeline to surface the most
    recent execution. Every step spells out its request inline, including the
    bearer Authorization header, so the flow can be read and executed without
    opening the underlying OpenAPI description.
  version: 1.0.0
sourceDescriptions:
- name: zenmlApi
  url: ../openapi/zenml-openapi.yml
  type: openapi
workflows:
- workflowId: trace-deployment-runs
  summary: From a deployment, read its pipeline and the latest run it produced.
  description: >-
    Lists deployments to select one, reads its pipeline, and lists runs for
    that pipeline to surface the latest execution.
  inputs:
    type: object
    required:
    - accessToken
    properties:
      accessToken:
        type: string
        description: ZenML JWT access token obtained from the login endpoint.
  steps:
  - stepId: listDeployments
    description: >-
      List pipeline deployments and select the first one to trace.
    operationId: listDeployments
    parameters:
    - name: Authorization
      in: header
      value: Bearer $inputs.accessToken
    - name: page
      in: query
      value: 1
    - name: size
      in: query
      value: 20
    successCriteria:
    - condition: $statusCode == 200
    outputs:
      deploymentId: $response.body#/items/0/id
      pipelineId: $response.body#/items/0/pipeline_id
  - stepId: readPipeline
    description: >-
      Read the deployment's associated pipeline for naming context.
    operationId: getPipeline
    parameters:
    - name: Authorization
      in: header
      value: Bearer $inputs.accessToken
    - name: pipeline_id
      in: path
      value: $steps.listDeployments.outputs.pipelineId
    successCriteria:
    - condition: $statusCode == 200
    outputs:
      pipelineName: $response.body#/name
  - stepId: listDeploymentRuns
    description: >-
      List the runs scoped to the deployment's pipeline, newest first, to
      surface the most recent execution.
    operationId: listPipelineRuns
    parameters:
    - name: Authorization
      in: header
      value: Bearer $inputs.accessToken
    - name: pipeline_id
      in: query
      value: $steps.listDeployments.outputs.pipelineId
    - name: page
      in: query
      value: 1
    - name: size
      in: query
      value: 1
    successCriteria:
    - condition: $statusCode == 200
    outputs:
      latestRunId: $response.body#/items/0/id
      runCount: $response.body#/total
  outputs:
    deploymentId: $steps.listDeployments.outputs.deploymentId
    pipelineName: $steps.readPipeline.outputs.pipelineName
    latestRunId: $steps.listDeploymentRuns.outputs.latestRunId