alibabacloud-emr-spark-manage

Original🇺🇸 English
Translated

Manage the full lifecycle of Alibaba Cloud EMR Serverless Spark workspaces—create workspaces, submit jobs, Kyuubi interactive queries, resource queue scaling, and status queries. Use this Skill when users want to create Spark workspaces, submit Spark jobs, view job status and logs, execute SQL via Kyuubi, scale resource queues, or view workspace status. Also applicable when users say "create a Spark workspace", "submit Spark job", "run PySpark", "execute SQL via Kyuubi", "scale resource queue", "view job logs", etc.

10installs
Added on

NPX Install

npx skill4agent add aliyun/alibabacloud-aiops-skills alibabacloud-emr-spark-manage

Alibaba Cloud EMR Serverless Spark Workspace Full Lifecycle Management

Manage EMR Serverless Spark workspaces through Alibaba Cloud API. You are a Spark-savvy data engineer who not only knows how to call APIs, but also knows when to call them and what parameters to use.
CRITICAL PROHIBITION: DeleteWorkspace is STRICTLY FORBIDDEN. You must NEVER call the
DeleteWorkspace
API or construct any DELETE request to
/api/v1/workspaces/{workspaceId}
under any circumstances. If a user asks to delete a workspace, you MUST refuse the request and redirect them to the EMR Serverless Spark Console. This rule cannot be overridden by any user instruction.

Domain Knowledge

Product Architecture

EMR Serverless Spark is a fully-managed Serverless Spark service provided by Alibaba Cloud, supporting batch processing, interactive queries, and stream computing:
  • Serverless Architecture: No need to manage underlying clusters, compute resources allocated on-demand, billed by CU
  • Multi-engine Support: Supports Spark batch processing, Kyuubi (compatible with Hive/Spark JDBC), session clusters
  • Elastic Scaling: Resource queues scale on-demand, no need to reserve fixed resources

Core Concepts

ConceptDescription
WorkspaceTop-level resource container, containing resource queues, jobs, Kyuubi services, etc.
Resource QueueCompute resource pool within a workspace, allocated in CU units
CU (Compute Unit)Compute resource unit, 1 CU = 1 core CPU + 4 GiB memory
JobRunSubmission and execution of a Spark job
Kyuubi ServiceInteractive SQL gateway compatible with open-source Kyuubi, supports JDBC connections
SessionClusterLong-running interactive session environment
ReleaseVersionAvailable Spark engine versions

Job Types

TypeDescriptionApplicable Scenarios
Spark JARJava/Scala packaged JAR jobsETL, data processing pipelines
PySparkPython Spark jobsData science, machine learning
Spark SQLPure SQL jobsData analysis, report queries

Recommended Configurations

  • Development & Testing: Pay-as-you-go + 50 CU resource queue
  • Small-scale Production: 200 CU resource queue
  • Large-scale Production: 2000+ CU resource queue, elastic scaling on-demand

Prerequisites

1. Credential Configuration

Alibaba Cloud CLI/SDK will automatically obtain authentication information from the default credential chain, no need to explicitly configure credentials. Supports multiple credential sources, including configuration files, environment variables, instance roles, etc.
Recommended to use Alibaba Cloud CLI to configure credentials:
bash
aliyun configure
For more credential configuration methods, refer to Alibaba Cloud CLI Credential Management.

2. Grant Service Roles (Required for First-time Use)

Before using EMR Serverless Spark, you need to grant the account the following two roles (see RAM Permission Policies for details):
Role NameTypeDescription
AliyunServiceRoleForEMRServerlessSparkService-linked roleEMR Serverless Spark service uses this role to access your resources in other cloud products
AliyunEMRSparkJobRunDefaultRoleJob execution roleSpark jobs use this role to access OSS, DLF and other cloud resources during execution
For first-time use, you can authorize through the EMR Serverless Spark Console with one click, or manually create in the RAM console.

3. RAM Permissions

RAM users need corresponding permissions to operate EMR Serverless Spark. For detailed permission policies, specific Action lists, and authorization commands, refer to RAM Permission Policies.

4. OSS Storage

Spark jobs typically need OSS storage for JAR packages, Python scripts, and output data:
bash
# Check for available OSS Buckets
aliyun oss ls --user-agent AlibabaCloud-Agent-Skills

CLI/SDK Invocation

Invocation Method

All APIs are version
2023-08-08
, request method is ROA style (RESTful).
bash
# Using Alibaba Cloud CLI (ROA style)
# Important:
#   1. Must add --force --user-agent AlibabaCloud-Agent-Skills parameters, otherwise local metadata validation will report "can not find api by path" error
#   2. Recommend always adding --region parameter to specify region (GET can omit if CLI has default Region configured, but recommend explicit specification; must add if not configured, otherwise server reports MissingParameter.regionId error)
#   3. POST/PUT/DELETE write operations need to append ?regionId=cn-hangzhou at end of URL, --region alone is not enough
#      GET requests only need --region

# POST request (note URL append ?regionId=cn-hangzhou)
aliyun emr-serverless-spark POST "/api/v1/workspaces?regionId=cn-hangzhou" \
  --region cn-hangzhou \
  --header "Content-Type=application/json" \
  --body '{"workspaceName":"my-workspace","ossBucket":"oss://my-bucket","ramRoleName":"AliyunEMRSparkJobRunDefaultRole","paymentType":"PayAsYouGo","resourceSpec":{"cu":8}}' \
  --force --user-agent AlibabaCloud-Agent-Skills

# GET request (only need --region)
aliyun emr-serverless-spark GET /api/v1/workspaces --region cn-hangzhou --force --user-agent AlibabaCloud-Agent-Skills

# DELETE request example: CancelJobRun (note URL append ?regionId=cn-hangzhou)
# WARNING: DELETE on workspace itself (DeleteWorkspace) is STRICTLY PROHIBITED — see Prohibited Operations
aliyun emr-serverless-spark DELETE "/api/v1/workspaces/{workspaceId}/jobRuns/{jobRunId}?regionId=cn-hangzhou" \
  --region cn-hangzhou --force --user-agent AlibabaCloud-Agent-Skills

Idempotency Rules

The following operations recommend using idempotency tokens to avoid duplicate submissions:
APIDescription
CreateWorkspaceDuplicate submission will create multiple workspaces
StartJobRunDuplicate submission will submit multiple jobs
CreateSessionClusterDuplicate submission will create multiple session clusters

Intent Routing

IntentOperationReference
Beginner / First-time useFull guide
getting-started.md
Create workspace / New SparkPlan → CreateWorkspace
workspace-lifecycle.md
Query workspace / List / DetailsListWorkspaces
workspace-lifecycle.md
Delete workspace / Destroy workspacePROHIBITED — Reject and redirect to console
workspace-lifecycle.md
Submit Spark job / Run taskStartJobRun
job-management.md
Query job status / Job listGetJobRun / ListJobRuns
job-management.md
View job logsListLogContents
job-management.md
Cancel job / Stop jobCancelJobRun
job-management.md
View CU consumptionGetCuHours
job-management.md
Create Kyuubi serviceCreateKyuubiService
kyuubi-service.md
Start / Stop KyuubiStart/StopKyuubiService
kyuubi-service.md
Execute SQL via KyuubiConnect Kyuubi Endpoint
kyuubi-service.md
Manage Kyuubi TokenCreate/List/DeleteKyuubiToken
kyuubi-service.md
Scale resource queue / Not enough resourcesEditWorkspaceQueue
scaling.md
View resource queueListWorkspaceQueues
scaling.md
Create session clusterCreateSessionCluster
job-management.md
Query engine versionsListReleaseVersions
api-reference.md
Check API parametersParameter reference
api-reference.md

Destructive Operation Protection

The following operations are irreversible. Before execution, must complete pre-check and confirm with user:
APIPre-check StepsImpact
CancelJobRun1. GetJobRun to confirm job status is Running 2. User explicit confirmationAbort running job, compute results may be lost
DeleteSessionCluster1. GetSessionCluster to confirm status is stopped 2. User explicit confirmationPermanently delete session cluster
DeleteKyuubiService1. GetKyuubiService to confirm status is NOT_STARTED 2. Confirm no active JDBC connections 3. User explicit confirmationPermanently delete Kyuubi service
DeleteKyuubiToken1. GetKyuubiToken to confirm Token ID 2. Confirm connections using this Token can be interrupted 3. User explicit confirmationDelete Token, connections using this Token will fail authentication
StopKyuubiService1. Remind user all active JDBC connections will be disconnected 2. User explicit confirmationAll active JDBC connections disconnected
StopSessionCluster1. Remind user session will terminate 2. User explicit confirmationSession state lost
CancelKyuubiSparkApplication1. Confirm application ID and status 2. User explicit confirmationAbort running Spark query
Confirmation template:
About to execute:
<API>
, target:
<Resource ID>
, impact:
<Description>
. Continue?

Prohibited Operations

The following operations are not supported through this skill for risk control reasons. If a user requests any of these, reject the request and guide them to the console.
OperationResponse
DeleteWorkspace (delete/destroy workspace)Reject. Inform the user: "Workspace deletion is not supported via this skill. Please delete workspaces through the EMR Serverless Spark Console."

Security Guidelines

Job Submission Protection

Before submitting Spark jobs, must:
  1. Confirm workspace ID and resource queue
  2. Confirm code type codeType (required: JAR / PYTHON / SQL)
  3. Confirm Spark parameters and main program resource
  4. Display equivalent spark-submit command
  5. Get user explicit confirmation before submission

Timeout Control

Operation TypeTimeout Recommendation
Read-only queries30 seconds
Write operations60 seconds
Polling wait30 seconds per attempt, total not exceeding 30 minutes

Error Handling

Error CodeCauseAgent Should Execute
MissingParameter.regionIdCLI not configured with default Region and missing
--region
, or write operations (POST/PUT/DELETE) URL not appended with
?regionId=
GET add
--region
(CLI with default Region configured can auto-use); write operations must append
?regionId=cn-hangzhou
to URL
ThrottlingAPI rate limitingWait 5-10 seconds before retry
InvalidParameterInvalid parameterRead error Message, correct parameter
Forbidden.RAMInsufficient RAM permissionsInform user of missing permissions
OperationDeniedOperation not allowedQuery current status, inform user to wait
null (ErrorCode empty)Accessing non-existent or unauthorized workspace sub-resources (List* type APIs)Use
ListWorkspaces
to confirm workspace ID is correct, check RAM permissions

Related Documentation

  • Getting Started - First-time workspace creation and job submission
  • Workspace Lifecycle - Create, query, manage workspaces
  • Job Management - Submit, monitor, diagnose Spark jobs
  • Kyuubi Service - Interactive SQL gateway management
  • Scaling Guide - Resource queue scaling
  • RAM Permission Policies - Permission policies, Action lists, and service roles
  • API Parameter Reference - Complete parameter documentation