Aliyun Exporter

Some metrics are not available via the standard CloudMonitor API. The exporter uses 'Special Projects' to handle these.

One such project is rds_performance. These projects use different APIs and do not consume your CloudMonitor API invocation quota, but they can be slower. It is recommended to run special projects in a standalone exporter instance to avoid impacting standard metric collection.

The exporter provides self-monitoring metrics:

• cloudmonitor_request_latency_seconds: Latency of CloudMonitor API calls.
• cloudmonitor_failed_request_latency_seconds: Latency of failed CloudMonitor API calls.
• aliyun_{project}_{metric}_up: A boolean metric indicating if a specific Cloud Monitor metric was successfully scraped.

Some metrics are not exposed via the standard Cloud Monitor API. To maintain configuration consistency, use the special project rds_performance to fetch detailed RDS performance data.

metrics:
  rds_performance:
  - name: <RDS_METRIC_NAME>

The exporter hosts a simple web interface to help you discover available CloudMonitor metrics. After launching, visit http://localhost:9525 to browse:

• host:port/: Lists all available monitor projects.
• host:port/projects/{project}: Shows metrics meta for a specific project.
• host:port/yaml/{project}: Provides a configuration YAML for the project's metrics.

To deploy the full stack (Exporter, Prometheus, Alertmanager, Grafana, and DingTalk integration), set the following environment variables before running docker-compose up -d:

• PROMETHEUS_HOST: Your public IP or hostname.
• DINGTALK_TOKEN: Your DingTalk robot token.
• ALIYUN_REGION: Your Alibaba Cloud region.
• ALIYUN_ACCESS_ID: Your Alibaba Cloud Access Key ID.
• ALIYUN_ACCESS_SECRET: Your Alibaba Cloud Access Key Secret.

export PROMETHEUS_HOST=YOUR_PUBLIC_IP_OR_HOSTNAME
export DINGTALK_TOKEN=YOUR_DINGTALK_ROBOT_TOEKN
export ALIYUN_REGION=YOUR_REGION
export ALIYUN_ACCESS_ID=YOUR_ID
export ALIYUN_ACCESS_SECRET=YOUR_SECRET
docker-compose up -d

To deploy the complete stack (Exporter, Prometheus, Alertmanager, Grafana, and DingTalk webhook), set the following environment variables and run docker-compose up -d from the ./docker-compose directory.

Required Environment Variables:

• PROMETHEUS_HOST: Public IP or hostname of Prometheus (used for alert links).
• DINGTALK_TOKEN: DingTalk robot token for notifications.
• ALIYUN_REGION: Your Alibaba Cloud region.
• ALIYUN_ACCESS_ID: Your Alibaba Cloud Access Key ID.
• ALIYUN_ACCESS_SECRET: Your Alibaba Cloud Access Key Secret.

export PROMETHEUS_HOST=YOUR_PUBLIC_IP_OR_HOSTNAME
export DINGTALK_TOKEN=YOUR_DINGTALK_ROBOT_TOEKN
export ALIYUN_REGION=YOUR_REGION
export ALIYUN_ACCESS_ID=YOUR_ID
export ALIYUN_ACCESS_SECRET=YOUR_SECRET
docker-compose up -d

You can launch a complete monitoring stack including Aliyun-Exporter, Prometheus, Grafana, and Alertmanager using the provided docker-compose setup. This requires Docker 1.17+.

1. Clone the repository.
2. Navigate to the docker-compose directory.
3. Run the stack with your Alibaba Cloud credentials as environment variables.

Once running, you can access Grafana at http://localhost:3000 (default credentials: admin:admin).

git clone git@github.com:aylei/aliyun-exporter.git
cd docker-compose
ALIYUN_ACCESS_ID=YOUR_ACCESS_ID ALIYUN_ACCESS_SECRET=YOUR_ACCESS_KEY docker-compose up

The exporter is a Python application and requires Python 3.5 or higher. Install it using pip3.

pip3 install aliyun-exporter

You can run the exporter as a container. An external configuration file must be mounted into the container.

docker pull aylei/aliyun-exporter:0.3.1

docker run -p 9525:9525 -v $(pwd)/aliyun-exporter.yml:$(pwd)/aliyun-exporter.yml aylei/aliyun-exporter:0.3.1 -c $(pwd)/aliyun-exporter.yml

The exporter is configured using a YAML file. Key configuration sections include:

• rate_limit: Request rate limit per second (default: 10).
• credential: Alibaba Cloud credentials (access_key_id, access_key_secret, and region_id which defaults to cn-hangzhou).
• metrics: A required list of CloudMonitor projects and their specific metrics. For each metric, you can specify name, rename, period (query period in seconds, default 60), and measure (e.g., Average, default Average).
• info_metrics: A list of projects to include instance information (e.g., ecs, rds, redis).

rate_limit: 5
credential:
  access_key_id: <YOUR_ACCESS_KEY_ID>
  access_key_secret: <YOUR_ACCESS_KEY_SECRET>
  region_id: <REGION_ID>

metrics:
  acs_cdn:
  - name: QPS
    rename: qps
    period: 60
    measure: Average

info_metrics:
  - ecs
  - rds
  - redis

After installation, run the exporter by specifying a configuration file.

• Default port: 9525
• Default config location: ./aliyun-exporter.yml

Metrics are exposed at http://localhost:9525/metrics.

aliyun-exporter -p 9525 -c aliyun-exporter.yml

When running the aliyun-exporter service using the provided Docker Compose setup, you must provide the following environment variables to authenticate with Alibaba Cloud and configure connectivity:

• ALIYUN_ACCESS_ID: Your Alibaba Cloud Access Key ID.
• ALIYUN_ACCESS_SECRET: Your Alibaba Cloud Access Key Secret.
• ALIYUN_REGION: The Alibaba Cloud region (e.g., cn-hangzhou).
• PROMETHEUS_HOST: The host address used for the Prometheus external URL (defaults to localhost).
• DINGTALK_TOKEN: The access token for the DingTalk robot webhook used by the webhook-dingtalk service.

# Required environment variables
ALIYUN_ACCESS_ID=YOUR_ACCESS_ID
ALIYUN_ACCESS_SECRET=YOUR_ACCESS_SECRET
ALIYUN_REGION=cn-hangzhou
PROMETHEUS_HOST=localhost
DINGTALK_TOKEN=YOUR_DINGTALK_ROBOT_TOKEN