Skip to main content

Platform overview

Platform overview​

For how Styrmin compares to Argo CD, Flux, and manual Helm, see Why Styrmin?.

What you can do with Styrmin​

After you install Styrmin on a Kubernetes cluster and load a driver for an application, you can:

  1. Deploy an application by choosing a driver, an environment, and a version. The driver sets the Kubernetes-specific configuration (storage classes, ingress, chart version), so you set driver parameters instead of chart values.
  2. Back up, upgrade, and restore with the procedures that the driver defines. For example, the driver contains the backup steps for Neo4j and the upgrade steps for Infrahub.
  3. Manage an environment as one unit. Group applications into environments. Back up every application in an environment with one operation. Clone an environment, including its production data, to test an upgrade. Applications in a cloned environment connect only to each other.

You run day-to-day operations from the UI, the CLI, or the API. kubectl still works. Styrmin does not change how applications expose their APIs or accept traffic from external systems.

Components​

   ┌────────────┐   GraphQL   ┌──────────┐    recorded intent     ┌────────┐
   │  CLI / UI  │ ──────────► │  Server  │ ─────────────────────► │ Agent  │
   │   / SDK    │             │ (Python) │   Prefect flows +      │ (in    │
   └────────────┘             └────┬─────┘   StyrminDeployment    │ cluster│
                                   │              resources       │   )    │
                                   ▼                              └───┬────┘
                              PostgreSQL                              │
                                                                      ▼
                                                                Kubernetes
                                                                resources
  • The server owns the database and the public GraphQL API. It records what should run.
  • The agent runs inside the target cluster. It reads the recorded intent (Prefect flows and StyrminDeployment resources) and changes the cluster to match. There is one agent per cluster.
  • The status reporter, inside the agent, sends the observed state of pods and deployments to the server, so the API and the UI show what runs in the cluster.
  • The CLI (styrminctl), the SDK (styrmin-sdk), and the frontend are clients of the same GraphQL API.

For the principles behind this design, see the Styrmin Constitution.

Core terms​

The GraphQL schema, the SDK, the CLI, the UI, and these docs use the terms below. For the full list, see Core concepts and the Glossary.

Application driver​

An application driver is a versioned specification of how Styrmin deploys, configures, backs up, and upgrades one application. Each application has one driver, and Styrmin deploys only applications that have a driver.

See What is an application driver?.

Cluster​

A cluster is a Kubernetes cluster that Styrmin manages. You set the cluster-level settings (storage classes, ingress, …) once, and every application deployed on the cluster uses them.

Styrmin supports one cluster at a time. Support for several clusters is on the roadmap.

Environment​

An environment is a logical deployment target in one cluster, such as dev, staging, prod, or team-a. An environment groups applications that share a lifecycle and a backup storage location. Backup, upgrade, and clone operations can apply to a whole environment.

Cluster ─┬─ Environment "dev"      ─── Deployment "infrahub-dev"
         │                          └─ Deployment "semaphore-dev"
         └─ Environment "staging"   ─── Deployment "infrahub-staging"

Deployment​

A deployment is one application deployed in one environment.

Actions and hooks​

A driver attaches Python actions to lifecycle hooks. There are two hooks, setup and upgrade, and each has three modes: pre, core, and post. Actions run in Prefect flows on the agent and can call built-in primitives (stop_components, start_components, execute_commands, upgrade_deployment_version, …). With actions, a driver defines an upgrade procedure (stop workers, run migrations, restart, verify) as a sequence that Styrmin runs the same way on every upgrade.

Next steps​