homelab/migrate-to-gitops.sh

#!/usr/bin/env bash
#
# Incremental GitOps Migration - Live Cluster
# Migrates existing cluster to new bootstrap-local + GitOps structure
#
set -euo pipefail

GREEN='\033[0;32m'
YELLOW='\033[1;33m'
RED='\033[0;31m'
NC='\033[0m'

log() { echo -e "${GREEN}[$(date +'%H:%M:%S')]${NC} $*"; }
warn() { echo -e "${YELLOW}[$(date +'%H:%M:%S')]${NC} $*"; }
error() { echo -e "${RED}[$(date +'%H:%M:%S')]${NC} $*"; }

SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"

log "========================================="
log "GitOps Migration - Live Cluster"
log "========================================="
echo ""

# Step 1: Verify prerequisites
log "Step 1: Verifying prerequisites..."
kubectl cluster-info > /dev/null || { error "kubectl not configured"; exit 1; }
command -v argocd > /dev/null || warn "ArgoCD CLI not found - will use kubectl only"

# Step 2: Update data-schemas path (critical - removes ddb-cluster duplication)
log "Step 2: Updating data-schemas Application path..."
log "  Current: k8s/data (includes ddb-cluster - DUPLICATION)"
log "  New:     k8s/data/schemas (schemas only - no duplication)"

if kubectl get application data-schemas -n argocd &>/dev/null; then
    kubectl patch application data-schemas -n argocd --type=json -p='[
      {
        "op": "replace",
        "path": "/spec/source/path",
        "value": "k8s/data/schemas"
      },
      {
        "op": "replace",
        "path": "/metadata/annotations/argocd.argoproj.io~1sync-wave",
        "value": "6"
      }
    ]'
    log "  ✓ data-schemas updated"
else
    warn "  data-schemas Application not found - will be created by homelab-root"
fi

# Step 3: Apply Longhorn 3-node HA config (already working, but ensure it's in git)
log "Step 3: Verifying Longhorn 3-node HA configuration..."
if [[ -d "k8s/infrastructure/longhorn" ]]; then
    kubectl apply -k k8s/infrastructure/longhorn/ || warn "Longhorn config apply failed (may already be applied)"
    log "  ✓ Longhorn 3-node HA config applied"
else
    warn "  Longhorn config not found - skipping"
fi

# Step 4: Update homelab-root if needed
log "Step 4: Ensuring app-of-apps root is up to date..."
kubectl apply -k k8s/argocd/root/
log "  ✓ homelab-root updated"

# Step 5: Sync all applications in wave order
log "Step 5: Syncing all applications..."
if command -v argocd &>/dev/null; then
    log "  Using ArgoCD CLI for sync..."
    argocd app sync homelab-root --prune || warn "homelab-root sync failed - check manually"
    
    # Wait a bit for child apps to be created
    sleep 5
    
    # Sync all apps
    argocd app sync --all --timeout 600 || warn "Some apps may need manual intervention"
else
    log "  ArgoCD CLI not found - triggering sync via kubectl..."
    # Refresh all apps
    kubectl get applications -n argocd -o name | while read app; do
        kubectl patch $app -n argocd --type=merge -p='{"operation":{"initiatedBy":{"username":"kubectl"},"sync":{"revision":"HEAD"}}}'
    done
fi

# Step 6: Wait for critical apps
log "Step 6: Waiting for critical applications..."
log "  Waiting for cert-manager..."
kubectl wait --for=condition=available --timeout=300s deployment/cert-manager -n cert-manager 2>/dev/null || warn "cert-manager not ready"

log "  Waiting for ingress-nginx..."
kubectl wait --for=condition=available --timeout=300s deployment/ingress-nginx-controller -n ingress-nginx 2>/dev/null || warn "ingress-nginx not ready"

log "  Waiting for DDB cluster..."
kubectl wait --for=jsonpath='{.status.phase}'='Cluster in healthy state' --timeout=300s cluster/ddb-cluster -n ddb 2>/dev/null || warn "DDB not ready"

# Step 7: Verify final state
log "Step 7: Verifying final state..."
echo ""

log "==> ArgoCD Applications:"
kubectl get applications -n argocd --no-headers | head -20

echo ""
log "==> Storage verification:"
NODE_COUNT=$(kubectl get nodes.longhorn.io -n longhorn-system --no-headers 2>/dev/null | wc -l | tr -d ' ')
VOLUME_REPLICAS=$(kubectl get volumes.longhorn.io -n longhorn-system -o jsonpath='{.items[0].spec.numberOfReplicas}' 2>/dev/null || echo "0")
log "  Longhorn nodes: $NODE_COUNT/3"
log "  Volume replicas: $VOLUME_REPLICAS (should be 3)"

echo ""
log "==> DDB cluster:"
kubectl get cluster -n ddb 2>/dev/null || echo "Not found"

echo ""
log "==> Forgejo:"
kubectl get deployment,svc -n cicd | grep forgejo || echo "Not found"

echo ""
log "========================================="
log "✅ Migration Complete!"
log "========================================="
echo ""
echo "Next steps:"
echo "  1. Verify all apps are Synced + Healthy:"
echo "     kubectl get applications -n argocd"
echo "     argocd app list"
echo ""
echo "  2. Access services:"
echo "     ArgoCD:    https://argocd.riotpiao.com"
echo "     Forgejo:   https://forgejo.riotpiao.com"
echo "     Grafana:   https://grafana.riotpiao.com"
echo ""
echo "  3. Update CLAUDE.md to reflect 3-node HA storage"
echo "     (all volumes now have 3 replicas across 3 nodes)"
echo ""