Use awscurl to query with Prometheus-compatible APIs

API requests for Amazon Managed Service for Prometheus must be signed with SigV4. You can use awscurl to simplify the querying process.

To install awscurl, you need to have Python 3 and pip package manager installed.

On a Linux based instance, the following command installs awscurl.

$ pip3 install awscurl

On a macOS machine, the following command installs awscurl.

$ brew install awscurl

The following example is a sample awscurl query. Replace the Region, Workspace-id and QUERY inputs with appropriate values for your use case:

# Define the Prometheus query endpoint URL. This can be found in the Amazon Managed Service for Prometheus console page 
# under the respective workspace. 

$ export AMP_QUERY_ENDPOINT=https://aps-workspaces.Region.amazonaws.com/workspaces/Workspace-id/api/v1/query

# credentials are infered from the default profile
$ awscurl -X POST --region Region \
                  --service aps "${AMP_QUERY_ENDPOINT}" -d 'query=QUERY'  --header 'Content-Type: application/x-www-form-urlencoded'

Note

Your query string must be url encoded.

For a query like query=up, you could get results such as:

{
  "status": "success",
  "data": {
    "resultType": "vector",
    "result": [
      {
        "metric": {
          "__name__": "up",
          "instance": "localhost:9090",
          "job": "prometheus",
          "monitor": "monitor"
        },
        "value": [
          1652452637.636,
          "1"
        ]
      },
    ]
  }
}

In order for awscurl to sign the provided requests, you will need to pass the valid credentials in one of the following ways:

• Provide the access key ID and the Secret key for the IAM role. You can find the access key and the secret key for the role in the https://console.aws.amazon.com/iam/.

  For example:

  $ export AMP_QUERY_ENDPOINT=https://aps-workspaces.Region.amazonaws.com/workspaces/Workspace_id/api/v1/query
  
  $ awscurl -X POST --region <Region> \
                    --access_key <ACCESS_KEY> \
                    --secret_key <SECRET_KEY> \
                    --service aps "$AMP_QUERY_ENDPOINT?query=<QUERY>"

• Reference the configuration files stored in the .aws/credentials and /aws/config file. You can also choose to specify the name of the profile to be used. If unspecified, the default file will be used. For example:

  $ export AMP_QUERY_ENDPOINT=https://aps-workspaces.<Region>.amazonaws.com/workspaces/<Workspace_ID>/api/v1/query
  $ awscurl -X POST --region <Region> \
                    --profile <PROFILE_NAME> 
                    --service aps "$AMP_QUERY_ENDPOINT?query=<QUERY>"

• Use the instance profile associated with the EC2 instance.

Executing query requests using awscurl container

When installing a different version of Python and the associated dependencies is not feasible, a container can be used to package the awscurl application and its dependencies. The following example uses a Docker runtime to deploy awscurl, but any OCI compliant runtime and image will work.

$ docker pull okigan/awscurl
$ export AMP_QUERY_ENDPOINT=https://aps-workspaces.Region.amazonaws.com/workspaces/Workspace_id/api/v1/query
$ docker run --rm -it okigan/awscurl --access_key $AWS_ACCESS_KEY_ID  --secret_key $AWS_SECRET_ACCESS_KEY \ --region Region --service aps "$AMP_QUERY_ENDPOINT?query=QUERY"