diff --git a/k8s/worker-deployment.yaml b/k8s/worker-deployment.yaml index b56eaf0..225a5ea 100644 --- a/k8s/worker-deployment.yaml +++ b/k8s/worker-deployment.yaml @@ -68,6 +68,74 @@ spec: --- apiVersion: apps/v1 kind: Deployment +metadata: + name: ai-service-analysis-worker + namespace: ai-service +spec: + replicas: 1 + selector: + matchLabels: + app: ai-service-analysis-worker + template: + metadata: + labels: + app: ai-service-analysis-worker + spec: + terminationGracePeriodSeconds: 20 + hostAliases: + - ip: "77.105.173.42" + hostnames: + - "s3-minio.estateliga.work" + containers: + - name: worker + image: localhost:30300/admin/ai-service:latest + command: ["/usr/local/bin/ai-service-worker"] + ports: + - containerPort: 8081 + env: + - name: WORKER_ID + valueFrom: + fieldRef: + fieldPath: metadata.name + - name: WORKER_TASK_TYPES + value: "call_analysis" + - name: WORKER_MODEL_PROFILES + value: "qwen2.5-14b" + - name: WORKER_CLAIM_LIMIT + value: "4" + - name: WORKER_LEASE_TIMEOUT + value: "5m" + envFrom: + - configMapRef: + name: ai-service-config + - secretRef: + name: ai-service-secrets + startupProbe: + httpGet: + path: /readyz + port: 8081 + periodSeconds: 5 + failureThreshold: 30 + readinessProbe: + httpGet: + path: /readyz + port: 8081 + periodSeconds: 10 + livenessProbe: + httpGet: + path: /healthz + port: 8081 + periodSeconds: 10 + resources: + requests: + cpu: 50m + memory: 96Mi + limits: + cpu: 500m + memory: 384Mi +--- +apiVersion: apps/v1 +kind: Deployment metadata: name: ai-service-transcription-worker namespace: ai-service