Imported from timstewart-dynatrace/Best-Practice-Notebooks (
CLOUD - Cloud Provider Integrations/AGENTS.md). Install upstream withnpx skills add timstewart-dynatrace/Best-Practice-Notebooks --skill CLOUD - Cloud Provider Integrations. Copyright stays with the author.
AGENTS.md — CLOUD: Cloud Provider Integrations
Per-series routing for AI agents. Repo-wide rules: ../AGENTS.md. Humans: see README.md.
9 notebooks on integrating Dynatrace with AWS, Azure, and GCP: the Clouds app vs classic ActiveGate polling, per-provider connection and authentication setup, EKS and serverless deep dives, CloudWatch and GCP Pub/Sub log ingestion, and multi-cloud observability patterns.
Routing table
Read only the file(s) matching the question. All paths are under markdown/.
| When the question is about… | Read |
|---|---|
| Clouds app vs classic ActiveGate polling, cloud entity model, cloud metrics vs host metrics, multi-cloud strategy planning | -[CLOUD]-01-cloud-integration-fundamentals.md |
| Connecting AWS: IAM role via CloudFormation, key-based auth removal (1.267), enabling services, CloudWatch metric delays, Metric Streams via Firehose, API cost control | -[CLOUD]-02-aws-integration.md |
| EKS clusters: DynaKube on EKS, node groups vs Fargate, IRSA, Container Insights comparison, namespace cost allocation | -[CLOUD]-03-aws-eks-monitoring.md |
Lambda and serverless: cold starts, throttles, FaaS invocation metrics (dt.service.faas_invoke.count), Service Detection v2 (GA), API Gateway, Step Functions, DynamoDB, end-to-end serverless tracing |
-[CLOUD]-04-aws-lambda-serverless.md |
| Connecting Azure: Azure Native Dynatrace Service, Entra ID app registration with a federated identity credential and Monitoring Reader, supported services, AKS, resource group mapping | -[CLOUD]-05-azure-integration.md |
Connecting GCP: service account auth, push-based Helm/Pub/Sub integration on GKE, Pub/Sub log-forwarder sizing and horizontal scaling, supported services, GKE monitoring, Cloud Run OneAgent coverage caveats (Java/Node.js only, build-time credentials, Gen1 metric gaps), manual traceparent propagation across Pub/Sub |
-[CLOUD]-06-gcp-integration.md |
| Forwarding cloud logs: Amazon Data Firehose setup, deprecated Lambda forwarder migration, S3 ingestion, source-vs-Dynatrace filtering, OpenPipeline processing, log cost | -[CLOUD]-07-cloudwatch-log-ingestion.md |
| Running on more than one cloud: cross-cloud comparison queries, naming conventions, unified dashboards and alerting, cost comparison, governance | -[CLOUD]-08-multi-cloud-patterns.md |
| Consolidated checklist: connection/auth choices, service tiering, log ingestion defaults, alert windows for delayed cloud metrics | -[CLOUD]-99-best-practice-summary.md |
If more than three rows match, start with -[CLOUD]-99-best-practice-summary.md
and follow its pointers.
Related series
- Operator/DynaKube depth behind the EKS/AKS/GKE notebooks:
../K8S - Kubernetes Monitoring/ - Parsing and routing forwarded cloud logs after ingestion:
../OPLOGS - OpenPipeline Logs/ - Collector-based ingest as an alternative cloud data path:
../OTEL - OpenTelemetry Integration/ - Controlling cloud-integration ingest cost:
../FINOPS - Cost Management & FinOps/
Rules
- Read-only; markdown only (see repo-root AGENTS.md for the full format table).
- Filenames contain literal brackets and a leading dash — quote paths in shell.
- Prefer
smartscapeNodesquery forms when quoting;fetch dt.entity.*variants shown in these notebooks are deprecated alternatives. - Cite by notebook ID (e.g. "CLOUD-02") and mention that the matching JSON in
notebooks/can be imported into a Dynatrace tenant for interactive use.
