Forge Container services is now in Preview, and therefore fully supported. However, it remains under active development and may be subject to shorter deprecation windows. Preview features are suitable for early adopters in production environments.

We release preview features so partners and developers can study, test, and integrate them prior to General Availability (GA). For more details, see Forge EAP, Preview, and GA.

Service Health monitoring provides real-time visibility into the health and performance of your Forge Container services. You can view these metrics in the developer console to monitor resource usage and service availability.

Service Health metrics are available only for apps that use Forge Container services. These metrics are not available for standard Forge function-based apps.

View Service Health metrics

To view Service Health metrics for your container services:

Access the developer console.
Select the Forge app that uses Forge Container services.
Select Metrics in the left menu.
Select Service Health in the left menu.

The Service Health page displays a table of your container services with the following information:

Column	Description
Service Instance	The name of the container service.
Status	Indicates whether the service is receiving production traffic. Possible values are `Available` and `Unavailable`.
Region	The region where the service is deployed.
Instances	The number of currently running instances out of the maximum configured instances.
Avg. CPU	The average CPU usage across all container instances within the service, shown as a percentage.
Avg. Memory	The average memory usage across all container instances within the service, shown as a percentage.
Deployed at	The date and time the service was last deployed.
Action	Link to view the logs for the service.

The Service Health page shows the current state of your services at the time the page loads. The page does not update automatically. To see the latest metrics, use the Refresh button or reload the page.

Filter metrics

Use the filters at the top of the Service Health page to narrow the displayed data:

Environment: Select a single environment to view metrics for. Only one environment can be selected at a time.
Region: Select one or more regions to filter by. By default, all regions are selected. Use Select all and Clear selections to manage your selections. Clearing all selections defaults back to showing all regions.
Reset to default: Resets the environment filter to production and the region filter to all regions.

The service table supports pagination with Previous and Next controls. The default page size is 10 services.

Available metrics

The following Service Health metrics are available for Forge Container services:

Metric	Type	Description
`container_cpu_usage_percentage`	Gauge	Container CPU usage as a percentage of the container's CPU limit.
`container_memory_usage_percentage`	Gauge	Container memory working-set usage as a percentage of the container's memory limit. Working set is the memory used for out-of-memory (OOM) decisions.
`service_instance_count`	Gauge	The number of running instances for a container service in a given region.
`container_uptime_seconds`	Gauge	The number of seconds a container has been running since it last started.
`container_status_restarts_total`	Cumulative Counter	The total number of times a container has restarted. This counter resets to zero when the pod is replaced.

Export container metrics

You can export Service Health metrics using the App metrics API. To export container metrics, include the following values in the query.filters.metrics array of your API request:

CONTAINER_CPU_USAGE_PERCENTAGE
CONTAINER_MEMORY_USAGE_PERCENTAGE
SERVICE_INSTANCE_COUNT
CONTAINER_UPTIME_SECONDS
CONTAINER_STATUS_RESTARTS_TOTAL

For details on setting up metrics export, see Export app metrics.

Tags and dimensions

The following tags and dimensions are available with container metrics when using the App metrics API:

Tag	Description	Applicable metrics
`appId`	The unique identifier for the Forge app.	All container metrics
`environmentId`	The environment UUID.	All container metrics
`serviceKey`	The service key as defined in the manifest.	All container metrics
`region`	The region where the service instance is running.	All container metrics
`cluster_uid`	The cluster where the service instance is running.	All container metrics
`container`	The container name.	`container_cpu_usage_percentage`, `container_memory_usage_percentage`, `container_uptime_seconds`, `container_status_restarts_total`
`pod`	The pod identifier for the container instance.	`container_cpu_usage_percentage`, `container_memory_usage_percentage`, `container_uptime_seconds`, `container_status_restarts_total`