Home > Tutorials > Run service health checks

Running service health checks

Learn how to run service health checks in Cloud Identity Plane (CIP). This document describes internal service health check APIs intended to be used by developers, for example, to write a monitoring script that reports the system status. Health checks are available only for services mentioned in this document.

Prerequisites

  • CIP is up and running.

  • Access to service configuration data (for example a docker-compose file) or information about the service domain from other sources.

api-gateway health check

Send a request to the /status endpoint of api-gateway. Use the service name and port configured for your deployment.

curl -v http://api-gateway-host:4050/status

The response can be either 200 OK if all services are up:

{
    "kafka": "UP",
    "hazelcast": "UP",
    "acp": "UP",
    "customerService": "UP",
    "authz": "UP"
}

or 503 Service Unavailable if one or more services are down:

{
    "kafka": "UP",
    "hazelcast": "UP",
    "acp": "DOWN",
    "customerService": "UP",
    "authz": "UP"
}

All of the following components are checked by default:

Component Description
kafka Kafka message broker used by the service
hazelcast Storage for token information, authentication and authorization sessions, authentication events shared data caches, and more
acp Checks if ACP is alive. This check should only be used when CIP is used as an ACP component. Disable this check in your api-gateway configuration otherwise.
customerService Service used to track and store information regarding organizations
authz Service used to track and store authorization information

You can disable the health check on any components from the above list. For more information, read Configuring health checks for api-gateway.

authz-service health check

Send a request to the /status endpoint of authz_service. Use the service name and port configured for your deployment.

curl -v http://authz_service-host:9050/status

The response can be either 200 OK if all services are up:

{
    "kafka": "UP",
    "hazelcast": "UP"
}

or 503 Service Unavailable if one or more services are down:

{
    "kafka": "UP",
    "hazelcast": "DOWN"
}

All of the following components are checked by default:

Component Description
kafka Kafka message broker used by the service
hazelcast Storage for token information, authentication and authorization sessions, authentication events, shared data caches, and more

You can disable the health check on any components from the above list. For more information, read Configuring health checks for authz_service.

devices-service health check

Send a request to the /status endpoint of devices-service. Use the service name and port configured for your deployment.

curl -v http://devices-service-host:8090/status

The response can be either 200 OK if all services are up:

{
"deviceStorage":"UP"
}

or 503 Service Unavailable if one or more services are down:

{
"deviceStorage":"DOWN"
}

All of the following components are checked by default:

Component Description
deviceStorage Data store holding device data

You can disable the health check on any components from the above list. For more information, read Configuring health checks for devices-service.

iam-services health check

Send a request to the /status endpoint of iam-services. Use the service name and port configured for your deployment.

curl -v http://iam-services-host:7900/status

The response can be either 200 OK if all services are up:

{
    "userStorage": "UP",
    "sessionStorage": "UP",
    "kafka": "UP",
    "acp": "UP",
    "deviceService": "UP",
    "permissionService": "UP"
}

or 503 Service Unavailable if one or more services are down:

{
    "userStorage": "UP",
    "sessionStorage": "UP",
    "kafka": "UP",
    "acp": "UP",
    "deviceService": "DOWN",
    "permissionService": "UP"
}

All of the following components are checked by default:

Component Description
userStorage Data store holding user data.
sessionStorage Data store holding session information.
kafka Kafka message broker used by the service
acp Checks if ACP is alive. This check should only be used when CIP is used as an ACP component. Disable this check in your iam-services configuration otherwise.
deviceService Service used to track and store device information.

You can disable the health check on any components from the above list.. For more information, read Configuring health checks for iam-services.

permission-service health check

Send a request to the /status endpoint of permission-service. Use the service name and port configured for your deployment.

curl -v http://permission-service-host:6969/status

The response can be either 200 OK if all services are up:

{
    "permissionStorage":"UP"
}

or 503 Service Unavailable if one or more services are down:

{
    "permissionStorage":"DOWN"
}

All of the following components are checked by default:

Component Description
permissionStorage Data store holding permission data

You can disable the health check on any components from the above list. For more information, read Configuring health checks for permission-service.