ai-fixer

Appearance

Kubernetes 部署 ​

使用 Helm Chart 在 Kubernetes 集群中部署 ai-fixer。

重要：ai-fixer 只负责应用本身，PostgreSQL 和 Redis 作为独立的基础设施服务，请提前准备好连接信息。

前置要求 ​

  • Kubernetes 1.24+
  • Helm 3.0+
  • kubectl 已配置
  • PostgreSQL 14+ 已部署（外部服务）
  • Redis 6.0+ 已部署（外部服务）

快速部署 ​

1. 克隆项目 ​

bash
git clone https://github.com/FLM210/ai-fixer.git
cd ai-fixer/deploy/helm/k8s-fixer

2. 创建命名空间 ​

bash
kubectl create namespace ai-fixer

3. 创建 Secret ​

bash
kubectl create secret generic ai-fixer-secrets \
  --from-literal=database-url='postgresql+asyncpg://user:password@your-postgres-host:5432/fixer' \
  --from-literal=redis-url='redis://your-redis-host:6379/0' \
  --from-literal=llm-api-key='sk-xxxxxxxx' \
  -n ai-fixer

4. 配置 values.yaml ​

bash
cp values.yaml my-values.yaml

编辑 my-values.yaml：

yaml
# 镜像配置
image:
  repository: hahtangtang/ai-fixer
  tag: latest

# 副本数
replicaCount: 1

# 服务配置
service:
  type: ClusterIP
  port: 8080

# 环境变量
env:
  - name: DATABASE_URL
    valueFrom:
      secretKeyRef:
        name: ai-fixer-secrets
        key: database-url
  - name: REDIS_URL
    valueFrom:
      secretKeyRef:
        name: ai-fixer-secrets
        key: redis-url
  - name: LLM_API_KEY
    valueFrom:
      secretKeyRef:
        name: ai-fixer-secrets
        key: llm-api-key

# 配置
config:
  LLM_PROVIDER: "anthropic"
  LLM_MODEL: "claude-3-5-sonnet-20241022"
  LOG_LEVEL: "info"

# 不需要创建 Secret（已手动创建）
secrets:
  create: false

5. 部署 ​

bash
helm install ai-fixer . -n ai-fixer -f my-values.yaml

6. 验证部署 ​

bash
# 查看 Pod 状态
kubectl get pods -n ai-fixer

# 查看服务
kubectl get svc -n ai-fixer

# 查看日志
kubectl logs -f deployment/ai-fixer -n ai-fixer

详细配置 ​

values.yaml 完整配置 ​

yaml
# 镜像配置
image:
  repository: hahtangtang/ai-fixer
  tag: latest
  pullPolicy: IfNotPresent

# 副本数
replicaCount: 1

# 服务配置
service:
  type: ClusterIP
  port: 8080

# Ingress 配置
ingress:
  enabled: true
  className: nginx
  annotations:
    cert-manager.io/cluster-issuer: letsencrypt-prod
  hosts:
    - host: ai-fixer.your-domain.com
      paths:
        - path: /
          pathType: Prefix
  tls:
    - secretName: ai-fixer-tls
      hosts:
        - ai-fixer.your-domain.com

# 环境变量
env:
  - name: DATABASE_URL
    valueFrom:
      secretKeyRef:
        name: ai-fixer-secrets
        key: database-url
  - name: REDIS_URL
    valueFrom:
      secretKeyRef:
        name: ai-fixer-secrets
        key: redis-url
  - name: LLM_API_KEY
    valueFrom:
      secretKeyRef:
        name: ai-fixer-secrets
        key: llm-api-key

# 配置
config:
  LLM_PROVIDER: "anthropic"
  LLM_MODEL: "claude-3-5-sonnet-20241022"
  LOG_LEVEL: "info"

# 资源限制
resources:
  requests:
    cpu: 500m
    memory: 512Mi
  limits:
    cpu: 2000m
    memory: 2Gi

# 健康检查
livenessProbe:
  httpGet:
    path: /healthz
    port: 8080
  initialDelaySeconds: 30
  periodSeconds: 10

readinessProbe:
  httpGet:
    path: /healthz
    port: 8080
  initialDelaySeconds: 5
  periodSeconds: 5

# 自动扩缩容
autoscaling:
  enabled: true
  minReplicas: 1
  maxReplicas: 5
  targetCPUUtilizationPercentage: 80

# RBAC
rbac:
  create: true

# ServiceAccount
serviceAccount:
  create: true
  name: ai-fixer

数据库迁移 Job ​

Helm Chart 包含数据库迁移 Job（首次部署或升级时运行）：

yaml
# templates/migrate-job.yaml
apiVersion: batch/v1
kind: Job
metadata:
  name: {{ include "ai-fixer.fullname" . }}-migrate
spec:
  template:
    spec:
      containers:
        - name: migrate
          image: "{{ .Values.image.repository }}:{{ .Values.image.tag }}"
          command: ["alembic", "upgrade", "head"]
          env:
            - name: DATABASE_URL
              valueFrom:
                secretKeyRef:
                  name: ai-fixer-secrets
                  key: database-url
      restartPolicy: OnFailure

手动运行迁移：

bash
kubectl create job --from=cronjob/ai-fixer-migrate ai-fixer-migrate-manual -n ai-fixer

清理 CronJob ​

定期清理超时的工作流：

yaml
# templates/cleanup-cronjob.yaml
apiVersion: batch/v1
kind: CronJob
metadata:
  name: {{ include "ai-fixer.fullname" . }}-cleanup
spec:
  schedule: "0 */1 * * *"
  jobTemplate:
    spec:
      template:
        spec:
          containers:
            - name: cleanup
              image: "{{ .Values.image.repository }}:{{ .Values.image.tag }}"
              command: ["python", "-m", "app.utils.cleanup"]

RBAC 配置 ​

ServiceAccount ​

yaml
apiVersion: v1
kind: ServiceAccount
metadata:
  name: ai-fixer
  namespace: ai-fixer

ClusterRole ​

yaml
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
  name: ai-fixer
rules:
  - apiGroups: [""]
    resources: ["pods", "services", "endpoints", "events"]
    verbs: ["get", "list", "watch"]
  - apiGroups: ["apps"]
    resources: ["deployments", "replicasets"]
    verbs: ["get", "list", "watch", "update", "patch"]
  - apiGroups: [""]
    resources: ["pods/log"]
    verbs: ["get"]

ClusterRoleBinding ​

yaml
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
  name: ai-fixer
roleRef:
  apiGroup: rbac.authorization.k8s.io
  kind: ClusterRole
  name: ai-fixer
subjects:
  - kind: ServiceAccount
    name: ai-fixer
    namespace: ai-fixer

监控配置 ​

ServiceMonitor ​

yaml
apiVersion: monitoring.coreos.com/v1
kind: ServiceMonitor
metadata:
  name: ai-fixer
  namespace: ai-fixer
spec:
  selector:
    matchLabels:
      app.kubernetes.io/name: ai-fixer
  endpoints:
    - port: http
      path: /metrics
      interval: 30s

Grafana Dashboard ​

导入预置的 Dashboard：

bash
# 获取 Dashboard JSON
cat deploy/grafana/k8s-fixer-overview.json

# 在 Grafana 中导入
# 1. 打开 Grafana
# 2. Dashboards → Import
# 3. 上传 JSON 文件
# 4. 选择 Prometheus 数据源

升级 ​

更新配置 ​

bash
helm upgrade ai-fixer . -n ai-fixer -f my-values.yaml

更新镜像 ​

bash
helm upgrade ai-fixer . -n ai-fixer \
  --set image.tag=v1.1.0

运行数据库迁移 ​

升级后可能需要运行数据库迁移：

bash
kubectl create job --from=cronjob/ai-fixer-migrate ai-fixer-migrate-$(date +%s) -n ai-fixer

回滚 ​

bash
# 查看历史
helm history ai-fixer -n ai-fixer

# 回滚到指定版本
helm rollback ai-fixer 1 -n ai-fixer

卸载 ​

bash
helm uninstall ai-fixer -n ai-fixer
kubectl delete namespace ai-fixer

故障排查 ​

Pod 启动失败 ​

bash
# 查看 Pod 详情
kubectl describe pod -l app.kubernetes.io/name=ai-fixer -n ai-fixer

# 查看日志
kubectl logs -l app.kubernetes.io/name=ai-fixer -n ai-fixer

# 查看事件
kubectl get events -n ai-fixer --sort-by='.lastTimestamp'

数据库连接失败 ​

bash
# 检查 Secret
kubectl get secret ai-fixer-secrets -n ai-fixer -o yaml

# 测试连接
kubectl exec -it deployment/ai-fixer -n ai-fixer -- python -c "
from app.config.settings import settings
print(settings.DATABASE_URL)
"

内存不足 ​

bash
# 查看资源使用
kubectl top pods -n ai-fixer

# 调整资源限制
helm upgrade ai-fixer . -n ai-fixer \
  --set resources.limits.memory=4Gi

最佳实践 ​

1. 使用命名空间隔离 ​

bash
kubectl create namespace ai-fixer

2. 配置资源限制 ​

yaml
resources:
  requests:
    cpu: 500m
    memory: 512Mi
  limits:
    cpu: 2000m
    memory: 2Gi

3. 启用自动扩缩容 ​

yaml
autoscaling:
  enabled: true
  minReplicas: 1
  maxReplicas: 5

4. 配置健康检查 ​

yaml
livenessProbe:
  httpGet:
    path: /healthz
    port: 8080
readinessProbe:
  httpGet:
    path: /healthz
    port: 8080

5. 使用 Secret 管理敏感信息 ​

bash
kubectl create secret generic ai-fixer-secrets \
  --from-literal=database-url='postgresql+asyncpg://...' \
  --from-literal=redis-url='redis://...' \
  --from-literal=llm-api-key='sk-...' \
  -n ai-fixer

下一步 ​

基于 MIT 许可发布

Copyright © 2024-present ai-fixer