.tar archive and load it into your local Docker image store with docker load. This approach works well for development, testing, and production environments where container orchestration is managed outside of Kubernetes.
Load the Container Image
The Console provides a pre-built container image as a.tar archive (maverics-orchestrator.tar). Download it from the Download Orchestrator Software modal inside any Deployment in the Console.
Load the image into your local Docker image store:
maverics-orchestrator. Confirm the image was loaded successfully:
Run the Container
Start the Orchestrator container with port mappings for HTTP traffic and TLS, volume mounts for configuration and certificates, and any required environment variables:-p 8080:8080— HTTP traffic port-p 9443:9443— HTTPS/TLS traffic port (health endpoint default)-v— Mounts for configuration and TLS certificates-e— Environment variables for runtime configuration
Volume Mounts
The following volume mounts are recommended for production deployments:
Mount configuration and certificate directories as read-only where possible to follow the principle of least privilege:
Environment Variables
Environment variables can be passed to the container individually with-e flags or in bulk using an environment file:
Docker Compose Example
Adocker-compose.yml for running the Orchestrator with persistent configuration and TLS:
Stable Deployment Identity
The Orchestrator includes asoid field in every log entry for identifying and correlating logs by instance. By default, each container receives a unique soid because containers have ephemeral machine identities — this is the expected behavior for horizontally scaled deployments where each container is a distinct instance.
For singleton deployment scenarios where you want the same soid to persist across container restarts (for example, a single Orchestrator container that gets recreated during updates), mount a stable /etc/machine-id file into the container:
soid field is used.
In-Place Config Reload
The Orchestrator applies configuration changes in place with zero downtime when it detects a config change. Three settings enable this in Docker:1. Enable reload polling
PassMAVERICS_RELOAD_CONFIG=true to the container:
2. Ensure a writable temp directory
The Orchestrator creates a Unix domain socket under$TMPDIR to coordinate the
handoff. In standard Docker containers /tmp is writable, so no additional
configuration is needed.
If you run the container with --read-only, mount a writable tmpfs and point
TMPDIR to it:
3. Mount a stable machine identity
The Orchestrator uses/etc/machine-id to derive the handoff socket path. If it
computes a different path across the reload (which happens when the file is missing
and it falls back to a random UUID), the reload applies the new config but
in-flight state (sessions, caches) is lost.
Mount the host’s /etc/machine-id into the container. This is the same mount
shown in Stable Deployment Identity — it serves
both log correlation and state-preserving reloads:
Docker Compose example
A completedocker-compose.yml with all three settings:
If your config file is mounted read-only from a host directory and you update it
in place, the Orchestrator detects the change within the polling interval (default
30s) and performs the reload automatically — no
docker restart required.Related Pages
Installation Overview
System requirements, download options, CLI flags, and environment variables
Configuration
Configure the Orchestrator after installation
Getting Started
End-to-end quick-start guide