- Metrics service toegevoegd
- Applicatie services starten op, behalve eveai_chat_client - Connectiviteit naar admin / eveai_app niet functioneel
This commit is contained in:
Josako
@@ -1,283 +0,0 @@
|
||||
#!/bin/bash
|
||||
# Deploy All EveAI Dev Services Script
|
||||
# File: deploy-all-services.sh
|
||||
|
||||
set -e
|
||||
|
||||
# Colors voor output
|
||||
RED='\033[0;31m'
|
||||
GREEN='\033[0;32m'
|
||||
YELLOW='\033[1;33m'
|
||||
BLUE='\033[0;34m'
|
||||
NC='\033[0m' # No Color
|
||||
|
||||
# Function voor colored output
|
||||
print_status() {
|
||||
echo -e "${BLUE}[INFO]${NC} $1"
|
||||
}
|
||||
|
||||
print_success() {
|
||||
echo -e "${GREEN}[SUCCESS]${NC} $1"
|
||||
}
|
||||
|
||||
print_warning() {
|
||||
echo -e "${YELLOW}[WARNING]${NC} $1"
|
||||
}
|
||||
|
||||
print_error() {
|
||||
echo -e "${RED}[ERROR]${NC} $1"
|
||||
}
|
||||
|
||||
# Check if kubectl is pointing to the right cluster
|
||||
check_cluster_context() {
|
||||
print_status "Checking cluster context..."
|
||||
|
||||
CURRENT_CONTEXT=$(kubectl config current-context)
|
||||
if [[ "$CURRENT_CONTEXT" != "kind-eveai-dev-cluster" ]]; then
|
||||
print_error "Wrong cluster context: $CURRENT_CONTEXT"
|
||||
print_error "Expected: kind-eveai-dev-cluster"
|
||||
echo "Switch context with: kubectl config use-context kind-eveai-dev-cluster"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
print_success "Using correct cluster context: $CURRENT_CONTEXT"
|
||||
}
|
||||
|
||||
# Wait for pods to be ready
|
||||
wait_for_pods() {
|
||||
local namespace=$1
|
||||
local app_label=$2
|
||||
local timeout=${3:-300}
|
||||
|
||||
print_status "Waiting for $app_label pods to be ready..."
|
||||
|
||||
if kubectl wait --for=condition=Ready pods -l app=$app_label -n $namespace --timeout=${timeout}s; then
|
||||
print_success "$app_label pods are ready"
|
||||
return 0
|
||||
else
|
||||
print_error "$app_label pods failed to become ready within ${timeout}s"
|
||||
return 1
|
||||
fi
|
||||
}
|
||||
|
||||
# Deploy services in correct order
|
||||
deploy_infrastructure() {
|
||||
print_status "Deploying infrastructure services (Redis, MinIO)..."
|
||||
|
||||
if kubectl apply -f redis-minio-services.yaml; then
|
||||
print_success "Infrastructure services deployed"
|
||||
else
|
||||
print_error "Failed to deploy infrastructure services"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
# Wait for infrastructure to be ready
|
||||
wait_for_pods "eveai-dev" "redis" 180
|
||||
wait_for_pods "eveai-dev" "minio" 300
|
||||
}
|
||||
|
||||
deploy_application_services() {
|
||||
print_status "Deploying EveAI application services..."
|
||||
|
||||
if kubectl apply -f eveai-services.yaml; then
|
||||
print_success "Application services deployed"
|
||||
else
|
||||
print_error "Failed to deploy application services"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
# Wait for key services to be ready
|
||||
wait_for_pods "eveai-dev" "eveai-app" 180
|
||||
wait_for_pods "eveai-dev" "eveai-api" 180
|
||||
wait_for_pods "eveai-dev" "eveai-chat-client" 180
|
||||
}
|
||||
|
||||
deploy_static_ingress() {
|
||||
print_status "Deploying static files service and Ingress..."
|
||||
|
||||
# Deploy static files service
|
||||
if kubectl apply -f static-files-service.yaml; then
|
||||
print_success "Static files service deployed"
|
||||
else
|
||||
print_error "Failed to deploy static files service"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
# Deploy Ingress
|
||||
if kubectl apply -f eveai-ingress.yaml; then
|
||||
print_success "Ingress deployed"
|
||||
else
|
||||
print_error "Failed to deploy Ingress"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
# Wait for services to be ready
|
||||
wait_for_pods "eveai-dev" "static-files" 60
|
||||
|
||||
# Wait for Ingress to be ready
|
||||
print_status "Waiting for Ingress to be ready..."
|
||||
kubectl wait --namespace eveai-dev \
|
||||
--for=condition=ready ingress/eveai-ingress \
|
||||
--timeout=120s || print_warning "Ingress might still be starting up"
|
||||
}
|
||||
|
||||
deploy_monitoring_only() {
|
||||
print_status "Deploying monitoring services..."
|
||||
|
||||
if kubectl apply -f monitoring-services.yaml; then
|
||||
print_success "Monitoring services deployed"
|
||||
else
|
||||
print_error "Failed to deploy monitoring services"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
# Wait for monitoring services
|
||||
wait_for_pods "eveai-dev" "flower" 120
|
||||
wait_for_pods "eveai-dev" "prometheus" 180
|
||||
wait_for_pods "eveai-dev" "grafana" 180
|
||||
}
|
||||
|
||||
# Check service status
|
||||
check_services() {
|
||||
print_status "Checking service status..."
|
||||
|
||||
echo ""
|
||||
print_status "Pods status:"
|
||||
kubectl get pods -n eveai-dev
|
||||
|
||||
echo ""
|
||||
print_status "Services status:"
|
||||
kubectl get services -n eveai-dev
|
||||
|
||||
echo ""
|
||||
print_status "Persistent Volume Claims:"
|
||||
kubectl get pvc -n eveai-dev
|
||||
}
|
||||
|
||||
# Test service connectivity via Ingress
|
||||
test_connectivity_ingress() {
|
||||
print_status "Testing Ingress connectivity..."
|
||||
|
||||
# Test Ingress endpoints
|
||||
endpoints=(
|
||||
"http://minty.ask-eve-ai-local.com:3080/admin/"
|
||||
"http://minty.ask-eve-ai-local.com:3080/api/healthz/ready"
|
||||
"http://minty.ask-eve-ai-local.com:3080/chat-client/"
|
||||
"http://minty.ask-eve-ai-local.com:3080/static/"
|
||||
"http://localhost:3009" # MinIO Console (direct)
|
||||
"http://localhost:3010" # Prometheus (direct)
|
||||
"http://localhost:3012" # Grafana (direct)
|
||||
)
|
||||
|
||||
for endpoint in "${endpoints[@]}"; do
|
||||
print_status "Testing $endpoint..."
|
||||
if curl -f -s --max-time 10 "$endpoint" > /dev/null; then
|
||||
print_success "$endpoint is responding via Ingress"
|
||||
else
|
||||
print_warning "$endpoint is not responding (may still be starting up)"
|
||||
fi
|
||||
done
|
||||
}
|
||||
|
||||
# Test service connectivity (legacy function for backward compatibility)
|
||||
test_connectivity() {
|
||||
test_connectivity_ingress
|
||||
}
|
||||
|
||||
# Show connection information for Ingress setup
|
||||
show_connection_info_ingress() {
|
||||
echo ""
|
||||
echo "=================================================="
|
||||
print_success "EveAI Dev Cluster deployed successfully!"
|
||||
echo "=================================================="
|
||||
echo ""
|
||||
echo "🌐 Service URLs (via Ingress):"
|
||||
echo " Main Application:"
|
||||
echo " • Main App: http://minty.ask-eve-ai-local.com:3080/admin/"
|
||||
echo " • API: http://minty.ask-eve-ai-local.com:3080/api/"
|
||||
echo " • Chat Client: http://minty.ask-eve-ai-local.com:3080/chat-client/"
|
||||
echo " • Static Files: http://minty.ask-eve-ai-local.com:3080/static/"
|
||||
echo ""
|
||||
echo " Infrastructure:"
|
||||
echo " • Redis: redis://minty.ask-eve-ai-local.com:3006"
|
||||
echo " • MinIO S3: http://minty.ask-eve-ai-local.com:3008"
|
||||
echo " • MinIO Console: http://minty.ask-eve-ai-local.com:3009"
|
||||
echo ""
|
||||
echo " Monitoring:"
|
||||
echo " • Flower (Celery): http://minty.ask-eve-ai-local.com:3007"
|
||||
echo " • Prometheus: http://minty.ask-eve-ai-local.com:3010"
|
||||
echo " • Grafana: http://minty.ask-eve-ai-local.com:3012"
|
||||
echo ""
|
||||
echo "🔑 Default Credentials:"
|
||||
echo " • MinIO: minioadmin / minioadmin"
|
||||
echo " • Grafana: admin / admin"
|
||||
echo " • Flower: Felucia / Jungles"
|
||||
echo ""
|
||||
echo "🛠️ Management Commands:"
|
||||
echo " • kubectl get all -n eveai-dev"
|
||||
echo " • kubectl get ingress -n eveai-dev"
|
||||
echo " • kubectl logs -f deployment/eveai-app -n eveai-dev"
|
||||
echo " • kubectl describe ingress eveai-ingress -n eveai-dev"
|
||||
echo ""
|
||||
echo "🗂️ Data Persistence:"
|
||||
echo " • Host data path: $HOME/k8s-data/dev/"
|
||||
echo " • Logs path: $HOME/k8s-data/dev/logs/"
|
||||
}
|
||||
|
||||
# Show connection information (legacy function for backward compatibility)
|
||||
show_connection_info() {
|
||||
show_connection_info_ingress
|
||||
}
|
||||
|
||||
# Main execution
|
||||
main() {
|
||||
echo "=================================================="
|
||||
echo "🚀 Deploying EveAI Dev Services to Kind Cluster"
|
||||
echo "=================================================="
|
||||
|
||||
check_cluster_context
|
||||
|
||||
# Deploy in stages
|
||||
deploy_infrastructure
|
||||
print_status "Infrastructure deployment completed, proceeding with applications..."
|
||||
sleep 5
|
||||
|
||||
deploy_application_services
|
||||
print_status "Application deployment completed, proceeding with Nginx and monitoring..."
|
||||
sleep 5
|
||||
|
||||
deploy_static_ingress
|
||||
deploy_monitoring_only
|
||||
print_status "All services deployed, running final checks..."
|
||||
sleep 10
|
||||
|
||||
check_services
|
||||
test_connectivity_ingress
|
||||
show_connection_info_ingress
|
||||
}
|
||||
|
||||
# Check for command line options
|
||||
case "${1:-}" in
|
||||
"infrastructure")
|
||||
check_cluster_context
|
||||
deploy_infrastructure
|
||||
;;
|
||||
"apps")
|
||||
check_cluster_context
|
||||
deploy_application_services
|
||||
;;
|
||||
"monitoring")
|
||||
check_cluster_context
|
||||
deploy_nginx_monitoring
|
||||
;;
|
||||
"status")
|
||||
check_cluster_context
|
||||
check_services
|
||||
;;
|
||||
"test")
|
||||
test_connectivity
|
||||
;;
|
||||
*)
|
||||
main "$@"
|
||||
;;
|
||||
esac
|
||||
@@ -34,6 +34,7 @@ spec:
|
||||
selector:
|
||||
matchLabels:
|
||||
app: eveai-app
|
||||
tier: frontend
|
||||
template:
|
||||
metadata:
|
||||
labels:
|
||||
@@ -119,6 +120,7 @@ spec:
|
||||
selector:
|
||||
matchLabels:
|
||||
app: eveai-api
|
||||
tier: frontend
|
||||
template:
|
||||
metadata:
|
||||
labels:
|
||||
@@ -204,6 +206,7 @@ spec:
|
||||
selector:
|
||||
matchLabels:
|
||||
app: eveai-chat-client
|
||||
tier: frontend
|
||||
template:
|
||||
metadata:
|
||||
labels:
|
||||
@@ -440,6 +443,7 @@ spec:
|
||||
selector:
|
||||
matchLabels:
|
||||
app: eveai-entitlements
|
||||
tier: backend
|
||||
template:
|
||||
metadata:
|
||||
labels:
|
||||
|
||||
19
k8s/dev/ingress-nginx-resources-patch.yaml
Normal file
19
k8s/dev/ingress-nginx-resources-patch.yaml
Normal file
@@ -0,0 +1,19 @@
|
||||
# Ingress-NGINX Controller Resource Patch
|
||||
# File: ingress-nginx-resources-patch.yaml
|
||||
# Purpose: Patch the ingress-nginx-controller deployment with higher resource limits
|
||||
# to prevent pthread_create() failures and worker process crashes
|
||||
#
|
||||
# This is a strategic merge patch that will be applied using:
|
||||
# kubectl patch deployment ingress-nginx-controller -n ingress-nginx --patch-file=<this-file>
|
||||
spec:
|
||||
template:
|
||||
spec:
|
||||
containers:
|
||||
- name: controller
|
||||
resources:
|
||||
requests:
|
||||
cpu: 500m
|
||||
memory: 512Mi
|
||||
limits:
|
||||
cpu: 2000m
|
||||
memory: 2Gi
|
||||
19
k8s/dev/metrics-server-patch.yaml
Normal file
19
k8s/dev/metrics-server-patch.yaml
Normal file
@@ -0,0 +1,19 @@
|
||||
# Metrics Server Patch for Kind Compatibility
|
||||
# File: metrics-server-patch.yaml
|
||||
# Purpose: Patch the metrics-server deployment with Kind-specific configuration
|
||||
# and appropriate resource limits for development environment
|
||||
#
|
||||
# This is a strategic merge patch that will be applied using:
|
||||
# kubectl patch deployment metrics-server -n kube-system --patch-file=<this-file>
|
||||
spec:
|
||||
template:
|
||||
spec:
|
||||
containers:
|
||||
- name: metrics-server
|
||||
args:
|
||||
- --cert-dir=/tmp
|
||||
- --secure-port=10250
|
||||
- --kubelet-preferred-address-types=InternalIP,ExternalIP,Hostname
|
||||
- --kubelet-use-node-status-port
|
||||
- --metric-resolution=15s
|
||||
- --kubelet-insecure-tls
|
||||
@@ -233,6 +233,118 @@ install_ingress_controller() {
|
||||
kubectl get services -n ingress-nginx
|
||||
}
|
||||
|
||||
# Patch Ingress Controller Resources
|
||||
patch_ingress_resources() {
|
||||
print_status "Patching Ingress Controller resources..."
|
||||
|
||||
# Wait a moment for the deployment to be fully created
|
||||
sleep 5
|
||||
|
||||
# Check if patch file exists
|
||||
local patch_file="ingress-nginx-resources-patch.yaml"
|
||||
if [[ ! -f "$patch_file" ]]; then
|
||||
print_error "Patch file not found: $patch_file"
|
||||
return 1
|
||||
fi
|
||||
|
||||
# Patch the ingress-nginx-controller deployment with higher resource limits
|
||||
print_status "Updating resource limits for ingress-nginx-controller using manifest file..."
|
||||
|
||||
# Apply patch with retry logic
|
||||
local max_attempts=5
|
||||
local attempt=1
|
||||
local success=false
|
||||
|
||||
while [ $attempt -le $max_attempts ] && [ "$success" = false ]; do
|
||||
print_status "Attempt $attempt/$max_attempts - patching ingress controller resources..."
|
||||
|
||||
if kubectl patch deployment ingress-nginx-controller -n ingress-nginx --patch-file "$patch_file"; then
|
||||
print_success "Successfully patched ingress-nginx-controller resources"
|
||||
success=true
|
||||
else
|
||||
if [ $attempt -lt $max_attempts ]; then
|
||||
print_warning "Patch attempt $attempt failed, retrying in 5 seconds..."
|
||||
sleep 5
|
||||
attempt=$((attempt + 1))
|
||||
else
|
||||
print_error "Failed to patch ingress-nginx-controller resources after $max_attempts attempts"
|
||||
return 1
|
||||
fi
|
||||
fi
|
||||
done
|
||||
|
||||
# Wait for rollout to complete
|
||||
print_status "Waiting for ingress controller rollout to complete..."
|
||||
kubectl rollout status deployment/ingress-nginx-controller -n ingress-nginx --timeout=300s
|
||||
|
||||
if [ $? -eq 0 ]; then
|
||||
print_success "Ingress Controller resource patch completed successfully"
|
||||
|
||||
# Verify the new resource settings
|
||||
print_status "Verifying new resource settings..."
|
||||
kubectl describe deployment ingress-nginx-controller -n ingress-nginx | grep -A 10 "Limits:\|Requests:" || true
|
||||
else
|
||||
print_error "Ingress Controller rollout failed"
|
||||
return 1
|
||||
fi
|
||||
}
|
||||
|
||||
# Install Metrics Server
|
||||
install_metrics_server() {
|
||||
print_status "Installing Metrics Server..."
|
||||
|
||||
# Apply metrics server with Kind-specific configuration
|
||||
kubectl apply -f https://github.com/kubernetes-sigs/metrics-server/releases/latest/download/components.yaml
|
||||
|
||||
# Check if patch file exists
|
||||
local patch_file="metrics-server-patch.yaml"
|
||||
if [[ ! -f "$patch_file" ]]; then
|
||||
print_error "Patch file not found: $patch_file"
|
||||
return 1
|
||||
fi
|
||||
|
||||
# Patch metrics server for Kind (disable TLS verification)
|
||||
print_status "Patching Metrics Server for Kind compatibility using manifest file..."
|
||||
|
||||
# Wait for metrics server deployment to exist
|
||||
local max_wait=30
|
||||
local wait_count=0
|
||||
while ! kubectl get deployment metrics-server -n kube-system &> /dev/null; do
|
||||
if [ $wait_count -ge $max_wait ]; then
|
||||
print_error "Metrics server deployment not found after waiting"
|
||||
return 1
|
||||
fi
|
||||
sleep 2
|
||||
wait_count=$((wait_count + 1))
|
||||
done
|
||||
|
||||
# Apply the patch
|
||||
if kubectl patch deployment metrics-server -n kube-system --patch-file "$patch_file"; then
|
||||
print_success "Successfully patched metrics-server configuration"
|
||||
else
|
||||
print_warning "Failed to patch metrics-server, but continuing..."
|
||||
fi
|
||||
|
||||
# Wait for metrics server to be ready
|
||||
print_status "Waiting for Metrics Server to be ready..."
|
||||
kubectl wait --for=condition=available deployment/metrics-server -n kube-system --timeout=300s
|
||||
|
||||
if [ $? -eq 0 ]; then
|
||||
print_success "Metrics Server installed and ready"
|
||||
|
||||
# Test metrics server
|
||||
print_status "Testing metrics server..."
|
||||
sleep 10 # Give metrics server time to collect initial metrics
|
||||
if kubectl top nodes &> /dev/null; then
|
||||
print_success "Metrics Server is working correctly"
|
||||
else
|
||||
print_warning "Metrics Server installed but may need more time to collect metrics"
|
||||
fi
|
||||
else
|
||||
print_warning "Metrics Server installation completed but readiness check failed"
|
||||
fi
|
||||
}
|
||||
|
||||
# Apply Kubernetes manifests
|
||||
apply_manifests() {
|
||||
print_status "Applying Kubernetes manifests..."
|
||||
@@ -351,6 +463,8 @@ main() {
|
||||
create_cluster
|
||||
verify_cri_status
|
||||
install_ingress_controller
|
||||
patch_ingress_resources
|
||||
install_metrics_server
|
||||
apply_manifests
|
||||
configure_registry_certificates
|
||||
verify_cluster
|
||||
|
||||
Reference in New Issue
Block a user