Struct PrometheusServer 

pub struct PrometheusServer { /* private fields */ }

Manager for a Prometheus metrics server instance.

This struct handles the lifecycle of a Prometheus server, which can either run as a Docker container or as an embedded server within the application process. It provides methods for starting, stopping, and monitoring the server.

Implementations

impl PrometheusServer

pub fn new( config: PrometheusServerConfig, metrics_registry: Option<Arc<Registry>>, enhanced_metrics_provider: Arc<EnhancedMetricsProvider>, ) -> Result<Self>

Creates a new Prometheus server manager with the provided configuration and metrics registry.

This constructor prepares the Prometheus server infrastructure but does not start the server. The actual server (Docker container or embedded) is started when start() is called.

Parameters

• config - The configuration settings for the Prometheus server
• metrics_registry - Optional Prometheus registry for the embedded server mode
• enhanced_metrics_provider - Provider that generates metrics data

Errors

Returns an error if the Docker manager connection fails to initialize

pub async fn create_docker_container(&self) -> Result<()>

Creates and configures a new Docker container for the Prometheus server.

Sets up a new container with appropriate port mappings, volume mounts, and configuration based on the server settings. This does not start the container, only creates it.

Parameters

• network - Optional Docker network to attach the container to

Errors

Returns an error if the Docker API fails to create the container

Panics

Panics if mutex locks cannot be acquired

pub fn registry(&self) -> Option<Arc<Registry>>

Returns a reference to the metrics registry used by the embedded Prometheus server.

This registry contains all the metrics that are exposed by the embedded Prometheus server. Returns None if no registry was provided during initialization.

pub fn container_id(&self) -> Option<String>

Returns the Docker container ID if the server is running in Docker mode.

This identifier can be used to reference the specific Prometheus container for management operations via Docker API. The container ID is only available when the server is configured to run in Docker mode and after the container has been created.

Returns

• Some(String) - Container ID if the server is using Docker and a container has been created
• None - If the server is not using Docker, the container hasn’t been created yet, or if there was an error acquiring the lock on the container ID

pub fn is_docker_based(&self) -> bool

Checks if this server is configured to use Docker.

The Prometheus server can operate in two modes:

1. Docker mode: Runs Prometheus in a separate Docker container, providing isolation and easier management but requiring Docker to be available
2. Embedded mode: Runs a Prometheus-compatible metrics HTTP server directly within the application process, with no external dependencies

Returns

• true - If the server is configured to use Docker
• false - If the server is using the embedded mode

Trait Implementations

impl Clone for PrometheusServer

fn clone(&self) -> PrometheusServer

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Drop for PrometheusServer

fn drop(&mut self)

Executes the destructor for this type.

impl ServerManager for PrometheusServer

async fn start( &self, network: Option<&str>, bind_ip: Option<String>, ) -> Result<()>

Starts the Prometheus server in either Docker or embedded mode.

• Docker – Creates (or reuses) a container, mounts configuration/data volumes, optionally connects it to a network, and waits for the container health-check.
• Embedded – Binds an in-process HTTP server to host:port and serves the /metrics endpoint backed by the supplied registry.

The call blocks until the server is ready.

Parameters

• network – Optional Docker network name (Docker mode only).

Errors

Returns an error if the container or server fails to start, health-check fails, the port is already in use, or required configuration is missing.

• network - Optional Docker network name to connect the container to. Only used in Docker mode. When provided, the container will be connected to this network to allow service discovery between services (e.g., Prometheus to scrape metrics from other containers).

Errors

Returns an error if:

• Docker container creation or startup fails (permission issues, Docker daemon not running, image not found)
• Port binding fails for the embedded server (port already in use or insufficient permissions)
• Server startup times out or fails for any other reason
• Configuration file generation fails or specified paths are invalid (Docker mode only)
• Health check fails for Docker container
• No metrics registry is provided for embedded server mode

async fn stop(&self) -> Result<()>

Stops the running Prometheus server and performs necessary cleanup.

This method safely terminates the Prometheus server instance based on its operational mode:

Docker Mode

When running as a Docker container:

• Retrieves the container ID from internal state
• Stops the Docker container using the Docker API
• Removes the container to free up resources
• Clears the internal container ID reference
• Handles cases where container is already stopped gracefully

Embedded Mode

When running as an embedded server:

• Acquires lock on the server instance
• Triggers a graceful shutdown of the HTTP server
• Waits for in-flight requests to complete
• Releases resources associated with the server

The method is idempotent and safe to call multiple times, even if the server is not running.

Errors

Returns an error if:

• Docker API encounters errors when stopping or removing the container
• Communication with Docker daemon fails
• Container removal fails due to permission issues

fn url(&self) -> String

Returns the fully qualified URL where the Prometheus server can be accessed.

Constructs a URL using the configured host and port values in the format: http://{host}:{port}

This URL can be used to:

• Access the Prometheus web UI for manual query and visualization
• Configure Grafana datasources programmatically
• Set up health checks or monitoring of the Prometheus instance
• Access the Prometheus HTTP API for programmatic queries

The URL is valid regardless of whether the server is running in Docker mode or embedded mode, as both modes expose the same interface on the configured host:port combination.

async fn is_running(&self) -> Result<bool>

Checks if the Prometheus server is currently running.

For Docker-based servers, checks the container status. For embedded servers, checks if the server instance exists.

Errors

Returns an error if checking the server status fails or if the Docker API reports an error.

async fn wait_until_ready(&self, timeout_secs: u64) -> Result<()>

Waits until the Prometheus server is ready to accept connections.

Periodically checks the server status until it’s ready or until the timeout expires. For Docker-based servers, performs HTTP health checks. For embedded servers, verifies the server is bound and responding.

Parameters

• timeout_secs - Maximum time to wait in seconds

Errors

Returns an error if the server fails to become ready within the timeout period or if health checks fail.