Skip to content

README.md

Latest commit

聽

History

History
82 lines (63 loc) 路 3.1 KB

README.md

File metadata and controls

82 lines (63 loc) 路 3.1 KB
Raw

Deployment infrastructure

This directory contains the supported production deployment surface for CCAM. The complete operator guide is ../DEPLOYMENT.md.

Supported topology

CCAM uses SQLite and supports one active dashboard writer per persistent volume. The deployment assets enforce one replica and a Recreate strategy. HPA, active-active replicas, blue-green, and canary deployments are intentionally not shipped while SQLite remains the persistence backend.

Directory map

Path Purpose
helm/agent-monitor/ Helm 4-compatible chart with safety schema, digest support, retained PVC, Ingress/Gateway API, MCP, NetworkPolicy, and ServiceMonitor
kubernetes/base/ Restricted-PSS Kustomize base with one Recreate-managed writer
kubernetes/overlays/ Dev, staging, and production resources and namespaces
kubernetes/components/ Optional MCP, monitoring, Gateway API, and CSI snapshot layers
nginx/ Rootless Nginx edge with WebSocket support and default-blocked hooks, metrics, and MCP routes
secrets/ Ignored local secret-file contract for Compose
terraform/ Provider-neutral deployment to an existing Kubernetes cluster through the validated Helm chart
scripts/ Validation, deploy, backup, restore, rollback, health, and teardown operations

Validate

npm run deploy:validate

The gate checks Dockerfiles, Compose profiles, Nginx, Helm, Kustomize, Terraform, dependency audits, authorship headers, and the one-writer invariant.

Container stack

umask 077
openssl rand -hex 32 > deployments/secrets/dashboard-token
openssl rand -hex 32 > deployments/secrets/hook-token
openssl rand -hex 32 > deployments/secrets/mcp-token
openssl rand -base64 32 > deployments/secrets/grafana-admin-password
npm run docker:full:up

The full stack starts dashboard, authenticated MCP, rootless Nginx, Prometheus, and Grafana. Every host port binds loopback by default.

Kubernetes

Create agent-monitor-secrets with dashboard-token, hook-token, and mcp-token, then use Helm or Kustomize. Prefer immutable image tags or digests.

REGISTRY="ghcr.io/$(gh repo view --json owner -q .owner.login)"
IMAGE_TAG="$(git rev-parse --short HEAD)"
helm upgrade --install agent-monitor deployments/helm/agent-monitor \
  --namespace agent-monitor-production \
  --create-namespace \
  --values deployments/helm/agent-monitor/values-production.yaml \
  --set image.registry= \
  --set image.repository=${REGISTRY}/claude-code-agent-monitor \
  --set image.tag=${IMAGE_TAG} \
  --atomic --wait --timeout 10m

Operations

./deployments/scripts/db-backup.sh --env production \
  --namespace agent-monitor-production --output ./backups

./deployments/scripts/deploy.sh --env production --method helm \
  --registry "${REGISTRY}" --image claude-code-agent-monitor \
  --tag "${IMAGE_TAG}" --skip-build

./deployments/scripts/rollback.sh --env production --method helm \
  --namespace agent-monitor-production

Production deploy, restore, rollback, and teardown flows back up first and fail closed when the backup cannot be verified.