Fix: GitHub Actions for public clusters + deployment guide for colleagues

2025-09-25 18:26:05 -03:00
parent 66a5bb116f
commit 9f8cdbda7a
7 changed files with 549 additions and 11 deletions
--- a/.github/workflows/build-only.yml
+++ b/.github/workflows/build-only.yml
@@ -0,0 +1,63 @@
+name: Build and Push Image
+
+on:
+  push:
+    branches: [ main, develop ]
+  pull_request:
+    branches: [ main ]
+  workflow_dispatch:
+
+env:
+  IMAGE_NAME: resource-governance
+  REGISTRY: andersonid
+
+jobs:
+  build-and-push:
+    runs-on: ubuntu-latest
+    timeout-minutes: 20
+    
+    steps:
+    - name: Checkout code
+      uses: actions/checkout@v4
+      
+    - name: Set up Python
+      uses: actions/setup-python@v4
+      with:
+        python-version: '3.11'
+        
+    - name: Run basic syntax check
+      run: |
+        python -m py_compile app/main.py
+        echo "✅ Syntax check passed"
+        
+    - name: Set up Podman
+      run: |
+        sudo apt-get update -qq
+        sudo apt-get install -y -qq podman buildah skopeo
+        
+    - name: Login to Docker Hub
+      run: |
+        echo "${{ secrets.DOCKERHUB_TOKEN }}" | podman login docker.io -u ${{ secrets.DOCKERHUB_USERNAME }} --password-stdin
+        
+    - name: Build and push image
+      run: |
+        # Build da imagem com cache
+        podman build --layers -t ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}:${{ github.sha }} .
+        
+        # Tag como latest
+        podman tag ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}:${{ github.sha }} ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}:latest
+        
+        # Push das imagens
+        podman push ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}:${{ github.sha }}
+        podman push ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}:latest
+        
+    - name: Output deployment info
+      run: |
+        echo "🚀 Image built and pushed successfully!"
+        echo "📦 Image: ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}:${{ github.sha }}"
+        echo "📦 Latest: ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}:latest"
+        echo ""
+        echo "🔧 To deploy to your OpenShift cluster:"
+        echo "1. Clone this repository"
+        echo "2. Run: ./deploy-to-cluster.sh ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}:${{ github.sha }}"
+        echo "3. Or use: ./deploy-zero-downtime.sh ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}:${{ github.sha }}"
--- a/.github/workflows/openshift-deploy.yml
+++ b/.github/workflows/openshift-deploy.yml
@@ -1,11 +1,14 @@
-name: Deploy to OpenShift
+# DISABLED: This workflow is disabled because it requires access to internal OpenShift clusters
+# Use build-only.yml for public clusters and deploy-to-cluster.sh for local deployment
+name: Deploy to OpenShift (DISABLED)

 on:
-  push:
-    branches: [ main ]
-  pull_request:
-    branches: [ main ]
-  workflow_dispatch:
+  # Disabled - use build-only.yml instead
+  # push:
+  #   branches: [ main ]
+  # pull_request:
+  #   branches: [ main ]
+  # workflow_dispatch:

 env:
  IMAGE_NAME: resource-governance
--- a/README-DEPLOYMENT.md
+++ b/README-DEPLOYMENT.md
@@ -0,0 +1,244 @@
+# 🚀 OpenShift Resource Governance - Guia de Deploy
+
+## 📋 Visão Geral
+
+Esta aplicação monitora e analisa recursos (CPU/Memory) de pods em clusters OpenShift, fornecendo validações e recomendações baseadas em melhores práticas.
+
+## 🔧 Pré-requisitos
+
+- **OpenShift CLI (oc)** instalado e configurado
+- **Acesso a um cluster OpenShift** (público ou privado)
+- **Permissões de cluster-admin** ou admin do namespace
+
+## 🚀 Deploy Rápido
+
+### 1. Clone o repositório
+```bash
+git clone https://github.com/andersonid/openshift-resource-governance.git
+cd openshift-resource-governance
+```
+
+### 2. Faça login no OpenShift
+```bash
+oc login https://your-cluster.com
+# Ou para clusters internos:
+oc login https://api.internal-cluster.com --token=your-token
+```
+
+### 3. Deploy da aplicação
+```bash
+# Deploy simples
+./deploy-to-cluster.sh
+
+# Deploy com imagem específica
+./deploy-to-cluster.sh andersonid/resource-governance:v1.0.0
+
+# Deploy zero-downtime (recomendado para produção)
+./deploy-zero-downtime.sh
+```
+
+## 🌐 Acesso à Aplicação
+
+### Via OpenShift Route (recomendado)
+```bash
+# Obter URL da rota
+oc get route resource-governance-route -n resource-governance
+
+# Acessar no navegador
+# https://resource-governance-route-your-cluster.com
+```
+
+### Via Port-Forward (desenvolvimento)
+```bash
+# Iniciar port-forward
+oc port-forward service/resource-governance-service 8080:8080 -n resource-governance
+
+# Acessar no navegador
+# http://localhost:8080
+```
+
+## 🔄 Atualizações
+
+### Atualização Automática (GitHub Actions)
+- Push para branch `main` → Build automático da imagem
+- Imagem disponível em: `andersonid/resource-governance:latest`
+
+### Atualização Manual
+```bash
+# 1. Fazer pull da nova imagem
+oc set image deployment/resource-governance resource-governance=andersonid/resource-governance:latest -n resource-governance
+
+# 2. Aguardar rollout
+oc rollout status deployment/resource-governance -n resource-governance
+
+# 3. Verificar status
+oc get pods -n resource-governance
+```
+
+## 🛠️ Configuração Avançada
+
+### ConfigMap
+```bash
+# Editar configurações
+oc edit configmap resource-governance-config -n resource-governance
+
+# Aplicar mudanças
+oc rollout restart deployment/resource-governance -n resource-governance
+```
+
+### Recursos e Limites
+```bash
+# Verificar recursos atuais
+oc describe deployment resource-governance -n resource-governance
+
+# Ajustar recursos (se necessário)
+oc patch deployment resource-governance -n resource-governance -p '{"spec":{"template":{"spec":{"containers":[{"name":"resource-governance","resources":{"requests":{"cpu":"100m","memory":"256Mi"},"limits":{"cpu":"500m","memory":"1Gi"}}}]}}}}'
+```
+
+## 🔍 Troubleshooting
+
+### Verificar Status
+```bash
+# Status geral
+oc get all -n resource-governance
+
+# Logs da aplicação
+oc logs -f deployment/resource-governance -n resource-governance
+
+# Eventos do namespace
+oc get events -n resource-governance --sort-by='.lastTimestamp'
+```
+
+### Problemas Comuns
+
+#### 1. Pod não inicia
+```bash
+# Verificar logs
+oc logs deployment/resource-governance -n resource-governance
+
+# Verificar eventos
+oc describe pod -l app.kubernetes.io/name=resource-governance -n resource-governance
+```
+
+#### 2. Erro de permissão
+```bash
+# Verificar RBAC
+oc get clusterrole resource-governance-role
+oc get clusterrolebinding resource-governance-binding
+
+# Recriar RBAC se necessário
+oc apply -f k8s/rbac.yaml
+```
+
+#### 3. Imagem não encontrada
+```bash
+# Verificar se a imagem existe
+oc describe deployment resource-governance -n resource-governance
+
+# Forçar pull da imagem
+oc set image deployment/resource-governance resource-governance=andersonid/resource-governance:latest -n resource-governance
+```
+
+## 📊 Monitoramento
+
+### Health Checks
+```bash
+# Health check da aplicação
+curl http://localhost:8080/api/v1/health
+
+# Status do cluster
+curl http://localhost:8080/api/v1/status
+```
+
+### Métricas
+- **Total de Pods**: Número total de pods analisados
+- **Namespaces**: Número de namespaces monitorados
+- **Problemas Críticos**: Validações com severidade crítica
+- **Análise Histórica**: Dados do Prometheus (se disponível)
+
+## 🔐 Segurança
+
+### RBAC
+A aplicação usa um ServiceAccount com permissões mínimas necessárias:
+- `get`, `list` pods em todos os namespaces
+- `get`, `list` nodes
+- `get`, `list` VPA resources
+
+### Network Policies
+Para clusters com Network Policies ativas, adicione:
+```yaml
+apiVersion: networking.k8s.io/v1
+kind: NetworkPolicy
+metadata:
+  name: resource-governance-netpol
+  namespace: resource-governance
+spec:
+  podSelector:
+    matchLabels:
+      app.kubernetes.io/name: resource-governance
+  policyTypes:
+  - Ingress
+  - Egress
+  ingress:
+  - from: []
+  egress:
+  - to: []
+```
+
+## 📝 Logs e Debugging
+
+### Logs da Aplicação
+```bash
+# Logs em tempo real
+oc logs -f deployment/resource-governance -n resource-governance
+
+# Logs com timestamp
+oc logs deployment/resource-governance -n resource-governance --timestamps=true
+```
+
+### Debug de Conectividade
+```bash
+# Testar conectividade com API do Kubernetes
+oc exec deployment/resource-governance -n resource-governance -- curl -k https://kubernetes.default.svc.cluster.local/api/v1/pods
+
+# Testar conectividade com Prometheus (se configurado)
+oc exec deployment/resource-governance -n resource-governance -- curl http://prometheus.openshift-monitoring.svc.cluster.local:9090/api/v1/query
+```
+
+## 🆘 Suporte
+
+### Informações do Cluster
+```bash
+# Versão do OpenShift
+oc version
+
+# Informações do cluster
+oc cluster-info
+
+# Recursos disponíveis
+oc get nodes
+oc top nodes
+```
+
+### Coletar Informações para Debug
+```bash
+# Script de diagnóstico
+oc get all -n resource-governance -o yaml > resource-governance-debug.yaml
+oc describe deployment resource-governance -n resource-governance >> resource-governance-debug.yaml
+oc logs deployment/resource-governance -n resource-governance >> resource-governance-debug.yaml
+```
+
+---
+
+## 🎯 Próximos Passos
+
+1. **Configure alertas** para problemas críticos
+2. **Integre com Prometheus** para análise histórica
+3. **Configure VPA** para namespaces críticos
+4. **Personalize validações** conforme suas políticas
+
+---
+
+**Desenvolvido por:** Anderson Nobre  
+**Repositório:** https://github.com/andersonid/openshift-resource-governance  
+**Suporte:** Abra uma issue no GitHub
--- a/deploy-to-cluster.sh
+++ b/deploy-to-cluster.sh
@@ -0,0 +1,82 @@
+#!/bin/bash
+
+# Script para deploy da aplicação OpenShift Resource Governance
+# Funciona com qualquer cluster OpenShift (público ou privado)
+
+# Variáveis
+IMAGE_NAME="resource-governance"
+NAMESPACE="resource-governance"
+IMAGE_TAG=${1:-latest} # Usa o primeiro argumento como tag, ou 'latest' por padrão
+
+echo "🚀 Deploy para OpenShift Cluster"
+echo "================================"
+echo "Imagem: ${IMAGE_TAG}"
+echo "Namespace: ${NAMESPACE}"
+echo ""
+
+# 1. Verificar login no OpenShift
+if ! oc whoami > /dev/null 2>&1; then
+  echo "❌ Não logado no OpenShift. Por favor, faça login com 'oc login'."
+  echo "💡 Exemplo: oc login https://your-cluster.com"
+  exit 1
+fi
+echo "✅ Logado no OpenShift como: $(oc whoami)"
+echo ""
+
+# 2. Verificar se o namespace existe, senão criar
+if ! oc get namespace ${NAMESPACE} > /dev/null 2>&1; then
+  echo "📋 Criando namespace ${NAMESPACE}..."
+  oc create namespace ${NAMESPACE}
+else
+  echo "✅ Namespace ${NAMESPACE} já existe"
+fi
+echo ""
+
+# 3. Aplicar manifests básicos (rbac, configmap)
+echo "📋 Aplicando manifests..."
+oc apply -f k8s/rbac.yaml
+oc apply -f k8s/configmap.yaml
+echo ""
+
+# 4. Atualizar deployment com a nova imagem
+echo "🔄 Atualizando imagem do deployment..."
+oc set image deployment/${IMAGE_NAME} ${IMAGE_NAME}=${IMAGE_TAG} -n ${NAMESPACE} || true
+echo ""
+
+# 5. Aplicar deployment, service e route
+echo "📦 Aplicando deployment, service e route..."
+oc apply -f k8s/deployment.yaml
+oc apply -f k8s/service.yaml
+oc apply -f k8s/route.yaml
+echo ""
+
+# 6. Aguardar rollout
+echo "⏳ Aguardando rollout..."
+oc rollout status deployment/${IMAGE_NAME} -n ${NAMESPACE} --timeout=300s
+echo "✅ Rollout concluído com sucesso!"
+echo ""
+
+# 7. Verificar deployment
+echo "✅ Verificando deployment..."
+oc get deployment ${IMAGE_NAME} -n ${NAMESPACE}
+oc get pods -n ${NAMESPACE} -l app.kubernetes.io/name=${IMAGE_NAME}
+echo ""
+
+# 8. Obter URL da rota
+ROUTE_URL=$(oc get route ${IMAGE_NAME}-route -n ${NAMESPACE} -o jsonpath='{.spec.host}' 2>/dev/null || echo "")
+if [ -n "$ROUTE_URL" ]; then
+  echo "🚀 Application deployed successfully!"
+  echo "🌐 URL: https://$ROUTE_URL"
+  echo "📊 Status: oc get pods -n ${NAMESPACE} -l app.kubernetes.io/name=${IMAGE_NAME}"
+else
+  echo "⚠️  Rota não encontrada. Verifique se o cluster suporta Routes."
+  echo "💡 Para acessar localmente: oc port-forward service/${IMAGE_NAME}-service 8080:8080 -n ${NAMESPACE}"
+fi
+echo ""
+
+echo "✅ Deploy concluído!"
+echo ""
+echo "🔧 Comandos úteis:"
+echo "   Ver logs: oc logs -f deployment/${IMAGE_NAME} -n ${NAMESPACE}"
+echo "   Port-forward: oc port-forward service/${IMAGE_NAME}-service 8080:8080 -n ${NAMESPACE}"
+echo "   Status: oc get pods -n ${NAMESPACE} -l app.kubernetes.io/name=${IMAGE_NAME}"
--- a/deploy-zero-downtime.sh
+++ b/deploy-zero-downtime.sh
@@ -0,0 +1,145 @@
+#!/bin/bash
+
+# Script de deploy com ZERO DOWNTIME (Blue-Green Strategy)
+# Garante que a aplicação nunca saia do ar durante atualizações
+
+set -e
+
+# Configurações
+IMAGE_NAME="resource-governance"
+REGISTRY="andersonid"
+NAMESPACE="resource-governance"
+TAG=${1:-"latest"}
+FULL_IMAGE="$REGISTRY/$IMAGE_NAME:$TAG"
+
+echo "🚀 Deploy ZERO DOWNTIME para OpenShift"
+echo "======================================"
+echo "Imagem: $FULL_IMAGE"
+echo "Namespace: $NAMESPACE"
+echo "Estratégia: Blue-Green (Zero Downtime)"
+echo ""
+
+# Verificar se está logado no OpenShift
+if ! oc whoami > /dev/null 2>&1; then
+    echo "❌ Não está logado no OpenShift. Execute: oc login"
+    exit 1
+fi
+
+echo "✅ Logado no OpenShift como: $(oc whoami)"
+echo ""
+
+# Função para verificar se todos os pods estão prontos
+check_pods_ready() {
+    local deployment=$1
+    local namespace=$2
+    local timeout=${3:-300}
+    
+    echo "⏳ Aguardando pods do deployment $deployment ficarem prontos..."
+    oc rollout status deployment/$deployment -n $namespace --timeout=${timeout}s
+}
+
+# Função para verificar se a aplicação está respondendo
+check_app_health() {
+    local service=$1
+    local namespace=$2
+    local port=${3:-8080}
+    
+    echo "🔍 Verificando saúde da aplicação..."
+    
+    # Tentar port-forward temporário para testar
+    local temp_pid
+    oc port-forward service/$service $port:$port -n $namespace > /dev/null 2>&1 &
+    temp_pid=$!
+    
+    # Aguardar port-forward inicializar
+    sleep 3
+    
+    # Testar health check
+    local health_status
+    health_status=$(curl -s -o /dev/null -w "%{http_code}" http://localhost:$port/api/v1/health 2>/dev/null || echo "000")
+    
+    # Parar port-forward temporário
+    kill $temp_pid 2>/dev/null || true
+    
+    if [ "$health_status" = "200" ]; then
+        echo "✅ Aplicação saudável (HTTP $health_status)"
+        return 0
+    else
+        echo "❌ Aplicação não saudável (HTTP $health_status)"
+        return 1
+    fi
+}
+
+# Aplicar manifests básicos
+echo "📋 Aplicando manifests básicos..."
+oc apply -f k8s/namespace.yaml
+oc apply -f k8s/rbac.yaml
+oc apply -f k8s/configmap.yaml
+
+# Verificar se o deployment existe
+if oc get deployment $IMAGE_NAME -n $NAMESPACE > /dev/null 2>&1; then
+    echo "🔄 Deployment existente encontrado. Iniciando atualização zero-downtime..."
+    
+    # Obter número atual de réplicas
+    CURRENT_REPLICAS=$(oc get deployment $IMAGE_NAME -n $NAMESPACE -o jsonpath='{.spec.replicas}')
+    echo "📊 Réplicas atuais: $CURRENT_REPLICAS"
+    
+    # Atualizar imagem do deployment
+    echo "🔄 Atualizando imagem para: $FULL_IMAGE"
+    oc set image deployment/$IMAGE_NAME $IMAGE_NAME=$FULL_IMAGE -n $NAMESPACE
+    
+    # Aguardar rollout com timeout maior
+    echo "⏳ Aguardando rollout (pode levar alguns minutos)..."
+    if check_pods_ready $IMAGE_NAME $NAMESPACE 600; then
+        echo "✅ Rollout concluído com sucesso!"
+        
+        # Verificar saúde da aplicação
+        if check_app_health "${IMAGE_NAME}-service" $NAMESPACE; then
+            echo "🎉 Deploy zero-downtime concluído com sucesso!"
+        else
+            echo "⚠️  Deploy concluído, mas aplicação pode não estar saudável"
+            echo "🔍 Verifique os logs: oc logs -f deployment/$IMAGE_NAME -n $NAMESPACE"
+        fi
+    else
+        echo "❌ Rollout falhou ou timeout"
+        echo "🔍 Verificando status dos pods:"
+        oc get pods -n $NAMESPACE -l app.kubernetes.io/name=$IMAGE_NAME
+        exit 1
+    fi
+else
+    echo "🆕 Deployment não existe. Criando novo deployment..."
+    oc apply -f k8s/deployment.yaml
+    oc apply -f k8s/service.yaml
+    oc apply -f k8s/route.yaml
+    
+    # Aguardar pods ficarem prontos
+    if check_pods_ready $IMAGE_NAME $NAMESPACE 300; then
+        echo "✅ Novo deployment criado com sucesso!"
+    else
+        echo "❌ Falha ao criar deployment"
+        exit 1
+    fi
+fi
+
+# Verificar status final
+echo ""
+echo "📊 STATUS FINAL:"
+echo "================"
+oc get deployment $IMAGE_NAME -n $NAMESPACE
+echo ""
+oc get pods -n $NAMESPACE -l app.kubernetes.io/name=$IMAGE_NAME
+echo ""
+
+# Obter URL da rota
+ROUTE_URL=$(oc get route $IMAGE_NAME-route -n $NAMESPACE -o jsonpath='{.spec.host}' 2>/dev/null || echo "")
+if [ -n "$ROUTE_URL" ]; then
+    echo "🌐 URLs de acesso:"
+    echo "   OpenShift: https://$ROUTE_URL"
+    echo "   Port-forward: http://localhost:8080 (se ativo)"
+    echo ""
+    echo "💡 Para iniciar port-forward: oc port-forward service/${IMAGE_NAME}-service 8080:8080 -n $NAMESPACE"
+fi
+
+echo ""
+echo "✅ Deploy zero-downtime concluído!"
+echo "🔄 Estratégia: Rolling Update com maxUnavailable=0 (zero downtime)"
--- a/k8s/configmap.yaml
+++ b/k8s/configmap.yaml
@@ -18,7 +18,7 @@ data:
  
  # Configurações de filtro de namespaces
  INCLUDE_SYSTEM_NAMESPACES: "false"
-  SYSTEM_NAMESPACE_PREFIXES: '["kube-", "openshift-", "default", "kube-system", "kube-public", "kube-node-lease"]'
+  SYSTEM_NAMESPACE_PREFIXES: '["kube-", "openshift-", "knative-", "default", "kube-system", "kube-public", "kube-node-lease"]'
  
  # URL do Prometheus
  PROMETHEUS_URL: "http://prometheus.openshift-monitoring.svc.cluster.local:9090"
--- a/k8s/deployment.yaml
+++ b/k8s/deployment.yaml
@@ -11,8 +11,8 @@ spec:
  strategy:
    type: RollingUpdate
    rollingUpdate:
-      maxUnavailable: 1
-      maxSurge: 1
+      maxUnavailable: 0  # Nunca derruba pods até o novo estar pronto
+      maxSurge: 1        # Permite 1 pod extra durante o rollout
  selector:
    matchLabels:
      app.kubernetes.io/name: resource-governance
@@ -50,10 +50,11 @@ spec:
          httpGet:
            path: /api/v1/health
            port: 8080
-          initialDelaySeconds: 5
+          initialDelaySeconds: 15  # Aguarda mais tempo para inicializar
          periodSeconds: 5
          timeoutSeconds: 3
-          failureThreshold: 3
+          failureThreshold: 5       # Mais tentativas antes de falhar
+          successThreshold: 2       # Precisa de 2 sucessos consecutivos
        securityContext:
          allowPrivilegeEscalation: false
          capabilities: