Concepts

CCattler takes a fundamentally different approach to container orchestration. Instead of managing mutable API objects (Pods, Deployments, ReplicaSets), CCattler operates on facts — immutable observations about what the world should look like and what it actually looks like.

Core ideas

Facts, not objects

The cluster stores key-value facts in etcd, organized into three namespaces:

• desired/ — what the world should look like (written by users and policy controllers)
• observed/ — what the world actually looks like (written by node agents)
• placement/ — scheduling decisions (written by the scheduler)

The gap between desired/ and observed/ is the work. Controllers close that gap continuously.

No YAML

CCattler uses a purpose-built domain language instead of YAML. You declare intent — the system figures out the implementation:

service web {
    image nginx:1.28
    instances 3
    expose 8080
}

This compiles to facts: service(name="web", image="nginx:1.28", instances=3), exposes(service="web", port=8080).

Everything communicates through state

No component calls another component directly. The scheduler writes a placement fact; the node agent notices it. The agent starts a container and writes an observed state fact; the endpoint controller notices it. This makes the system naturally resilient — any component can crash and restart without breaking the others.

Kubernetes comparison

If you know Kubernetes, here's how CCattler maps:

Kubernetes	CCattler	Key difference
Cluster	Fact Store + Agents	No explicit cluster object — nodes register by writing facts
Pod	Instance + Group	An instance is one container; groups share network/volumes
Deployment / ReplicaSet	Service fact + Instance controller	desired=5, actual=3 → create 2 more. No object chain
Control Plane	Runner + Controllers + Store	Controllers are stateless rule engines, not resource managers
Reconciliation loop	facts → controllers → changes → agents → observations → repeat	Single loop for all capabilities
Taints/Tolerations	restrict / accept	Human-readable placement constraints
Affinity/Anti-affinity	require / prefer	Direct label matching, no complex expressions

Learn more

• Facts & State — the fact-based state model
• Reconciliation — how the loop works
• Controllers — rule engines that close the gap
• Networking — service discovery, VIPs, DNS
• Security — zero-trust security model
• Ecosystem Comparison — where CCattler fits among orchestrators and management layers
• Kubernetes Comparison — how CCattler replaces the K8s production tooling ecosystem