AWS AmazonCloudWatch documentation change

Service: AmazonCloudWatch · 2026-07-01 · Documentation low

File: AmazonCloudWatch/latest/monitoring/Creating-pipelines.md

Summary

Added comprehensive documentation for metrics pipelines including creation procedures, CLI commands, configuration examples, and supported API operations.

Security assessment

The changes introduce operational documentation for a new metrics pipeline feature without security-specific content. No security vulnerabilities or security features are mentioned in the added content.

Diff

diff --git a/AmazonCloudWatch/latest/monitoring/Creating-pipelines.md b/AmazonCloudWatch/latest/monitoring/Creating-pipelines.md
index d032288d3..389a769b6 100644
--- a//AmazonCloudWatch/latest/monitoring/Creating-pipelines.md
+++ b//AmazonCloudWatch/latest/monitoring/Creating-pipelines.md
@@ -6,0 +7,2 @@
+Creating a logs pipelineCreating a metrics pipeline
+
@@ -9 +11,5 @@
-The pipeline configuration wizard guides you through creating your data pipeline.
+CloudWatch pipelines supports creating pipelines for both logs and metrics data. The following sections describe how to create each type of pipeline.
+
+## Creating a logs pipeline
+
+The pipeline configuration wizard guides you through creating your logs data pipeline.
@@ -13 +19 @@ The pipeline configuration wizard guides you through creating your data pipeline
-  2. Under **Destination** , specify the destination details. CloudWatch Logs is the default destination. 
+  2. Under **Destination** , specify the destination details. For logs pipelines, CloudWatch Logs is the only supported destination. For metrics pipelines, CloudWatch Metrics is the only supported destination.
@@ -15 +21 @@ The pipeline configuration wizard guides you through creating your data pipeline
-  3. Under **Processor** , add the desired processors and parsers. A parser is a required first step for certain data types. You can perform custom parsing using processors like Grok or CSV. Processors that are not supported by the data type are disabled.
+  3. Under **Processor** , add the desired processors and parsers. A parser is a required first step for certain data types. You can perform custom parsing using processors like Grok or CSV. You can use the lookup processor to enrich log events with data from an Amazon CloudWatch Logs lookup table. Processors that are not supported by the data type are disabled.
@@ -35,0 +42,72 @@ Pipeline processor configurations are logged in AWS CloudTrail events for auditi
+## Creating a metrics pipeline
+
+You can create a metrics pipeline by using the CloudWatch console or the AWS CLI. With a metrics pipeline, you can process CloudWatch metrics data, apply transformations, and forward the processed metrics to CloudWatch.
+
+### To create a metrics pipeline (console)
+
+Use the following procedure to create a metrics pipeline in the CloudWatch console.
+
+  1. Open the CloudWatch console and in the navigation pane, choose **Ingestion** , **Pipelines**.
+
+  2. Choose **Create pipeline**.
+
+  3. For the source type, select **CloudWatch Metrics (OTel)**.
+
+  4. Configure the selection criteria to specify which metrics the pipeline processes. Selection criteria use OTel attribute paths such as `resource.attributes`, `metric.name`, and `datapoint.attributes`. All criteria use AND semantics (all must match).
+
+  5. Under **Processor** , add up to 20 processors to transform your metrics data.
+
+  6. Under **Review and create** , review the pipeline configuration and choose **Create pipeline**.
+
+
+
+
+### To create a metrics pipeline (AWS CLI)
+
+Use the `create-telemetry-pipeline` command to create a metrics pipeline from the AWS CLI.
+    
+    
+    aws observabilityadmin create-telemetry-pipeline \
+        --pipeline-configuration-body file://config.yaml
+
+The following example shows a YAML configuration file that defines a metrics pipeline with a source, processor, and sink.
+    
+    
+    pipeline:
+      source:
+        cloudwatch_metrics:
+          format: otlp
+          selection_criteria:
+            - match_all:
+                - 'resource.attributes["service.name"] == "payment-service"'
+                - 'metric.name == "http.server.request.duration"'
+      processor:
+        - add_attributes:
+            attributes:
+              - key: resource.attributes["team"]
+                value: "platform-engineering"
+      sink:
+        - cloudwatch_metrics: {}
+
+When the pipeline is created, it returns an ARN in the following format:
+    
+    
+    arn:aws:observabilityadmin:region:accountId:telemetry-pipeline/pipelineID
+
+###### Note
+
+CloudWatch Metrics is the only supported destination for metrics pipelines.
+
+### Supported API operations
+
+The following table lists the API operations that you can use to manage metrics pipelines.
+
+Operation | Description  
+---|---  
+`CreateTelemetryPipeline` | Creates a new metrics pipeline.  
+`GetTelemetryPipeline` | Retrieves details about a specific metrics pipeline.  
+`ListTelemetryPipelines` | Lists all metrics pipelines in your account.  
+`DeleteTelemetryPipeline` | Deletes a metrics pipeline.  
+`ValidateTelemetryPipelineConfiguration` | Validates a pipeline configuration before creation.  
+`UpdateTelemetryPipeline` | Updates an existing metrics pipeline configuration.  
+