Managed Service for Prometheus:Create a data delivery task

Prerequisites

Before you begin, make sure that you have:

• A Prometheus instance for one of the following monitoring scenarios:

  • ACK cluster monitoring

  • Alibaba Cloud service monitoring

  • ECS instance monitoring

• EventBridge activated and authorized in the EventBridge console. For more information, see Activate EventBridge and grant permissions

• A target service to receive the delivered data: an ApsaraMQ for Kafka instance, a MaxCompute instance, or another Prometheus instance. For setup instructions, see ApsaraMQ for Kafka overview or Import monitoring data from a Prometheus instance to MaxCompute

Data delivery depends on EventBridge, which became commercially available on June 3, 2025. For billing details, see EventBridge billing overview.

Create a delivery task

1. Log on to the ARMS console.

2. In the left navigation pane, choose Managed Service for Prometheus > Data Delivery.

3. In the top navigation bar, select a region, and then click Create Task.

4. In the dialog box, set Task Name and Task Description, and then click OK.

5. On the Edit Task page, configure the data source and event target:

   1. Click + Add Data Source, configure the following parameters, and then click Determine.

      Parameter	Description	Example
      Prometheus Instance	The Prometheus instance whose data you want to deliver.	c78cb8273c02*****
      Data Filtering	Label-based filters that select which metrics to deliver. You can switch between Blacklist Mechanism and Whitelist Mechanism modes. Use regular expressions and separate multiple conditions with line breaks. All conditions are evaluated with a logical AND -- data is delivered only when every condition is met.	__name__=AliyunEcs_CPUUtilization|AliyunEcs_memory_usedutilization regionId=cn-hangzhou id=i-2ze0mxp.*
      Data Labeling	Custom labels added to delivered metric data. Separate multiple labels with line breaks.	deliver_test_key1=ssss deliver_test_key2=yyyy

   2. Click Add Target and configure the event target. Use the default settings in the Retry section unless your scenario requires otherwise.

6. Click OK, and then click Save.

7. After saving, the task automatically starts in the Running state. Data arrives at the target with a delay of 2 to 5 minutes.

Data format

Managed Service for Prometheus converts metric data to JSON before delivery for compatibility with other Alibaba Cloud services.

Prometheus format (source):

http_requests_total{method="POST", handler="/api/tracks"} 1027 1609459200000

JSON format (delivered):

{
  "__name__": "http_requests_total",
  "method": "POST",
  "handler": "/api/tracks",
  "value": 1027,
  "timestamp": 1609459200000
}

Each metric label becomes a JSON field. The value and timestamp fields are extracted from the Prometheus data point.

Limitations

• Real-time data only. Only data generated after you create a delivery task can be exported. Historical data delivery is not supported.

• Supported instance types. Not all Prometheus instance types support data delivery.

  Instance type	Supported
  Prometheus for Alibaba Cloud services	Yes, except for free instances with names starting with cloud-product-prometheus
  Prometheus for container services	Yes
  Prometheus for Flink Serverless	Yes
  Prometheus for Kubernetes	Yes
  General-purpose Prometheus instance	Yes, except for instances whose data is reported through OpenTelemetry endpoints

• Cross-VPC delivery. If the Prometheus instance and the target reside in different virtual private clouds (VPCs), add the vSwitch CIDR block of the target VPC to the Prometheus instance whitelist. Otherwise, data delivery may fail due to network connectivity issues. To find the vSwitch CIDR block, go to the vSwitch page in the VPC console.