init

vw-document-ai-indexer/deploy/dev/deploy.sh

@@ -0,0 +1,46 @@
+
+# login AKS
+# az cloud set --name AzureCloud   # Switch CLI to Azure cloud
+# az login                              # Log in to Azure China account (browser or device code flow)
+# az account set -s 079d8bd8-b4cc-4892-9307-aa6dedf890e9 #! set subs
+# az aks get-credentials -g rg-aiflow-lab -n aks-aiflow-lab --overwrite-existing --file ~/.kube/config
+kubectl config use-context aks-aiflow-lab
+kubectl config current-context
+
+# kubectl create secret generic azure-files-cred \
+#   --from-literal=azurestorageaccountname=saaisearchlab \
+#   --from-literal=azurestorageaccountkey=xxxxxxxxxxxxxxxxxxxx \
+#   -n knowledge-agent            
+
+# kubectl delete configmap document-ai-indexer-config -n knowledge-agent
+
+docker build . -t document-ai-indexer:2.0.2
+docker tag document-ai-indexer:2.0.2 acraiflowlab.azurecr.io/document-ai-indexer:2.0.2
+docker push acraiflowlab.azurecr.io/document-ai-indexer:2.0.2
+
+
+# dev
+kubectl delete configmap document-ai-indexer-config -n knowledge-agent
+kubectl create configmap document-ai-indexer-config -n knowledge-agent --from-file=env.yaml --from-file=config.yaml
+
+# kubectl create namespace knowledge-agent
+
+# # kubectl delete pod document-ai-indexer -n knowledge-agent
+# kubectl apply -f document-ai-indexer_k8s.yml -n knowledge-agent
+
+# kubectl logs -f document-ai-indexer  -n knowledge-agent
+
+# Deploy CronJob
+kubectl apply -f deploy/dev/document-ai-indexer-cronjob.yml --namespace knowledge-agent
+
+# Check CronJob Status
+kubectl get cronjobs -n knowledge-agent --namespace knowledge-agent
+# Check Job Execution History
+kubectl get jobs -n knowledge-agent --namespace knowledge-agent
+
+###########
+# Manually trigger a job (for testing)
+kubectl delete job manual-test -n knowledge-agent
+kubectl create job --from=cronjob/document-ai-indexer-cronjob manual-test -n knowledge-agent
+# Check Job Logs
+kubectl logs -f job/manual-test -n knowledge-agent

vw-document-ai-indexer/deploy/dev/document-ai-indexer-cronjob.yml

@@ -0,0 +1,64 @@
+apiVersion: batch/v1
+kind: CronJob
+metadata:
+  name: document-ai-indexer-cronjob
+spec:
+  # Scheduling configuration - execute every 10 minutes
+  schedule: "*/10 * * * *"
+  
+  # Concurrency policy: Disable concurrent execution. If the previous job is still running, new execution will be skipped.
+  concurrencyPolicy: Forbid
+
+  # Successful jobs history limit: Keep the last 3 successful job records.
+  successfulJobsHistoryLimit: 10
+
+  # Failed jobs history limit: Keep the last failed job record.
+  failedJobsHistoryLimit: 10
+
+  # Job template
+  jobTemplate:
+    spec:
+      backoffLimit: 0
+      template:
+        metadata:
+          labels:
+            app: document-ai-indexer
+            job-type: cronjob
+        spec:
+          restartPolicy: Never
+
+          volumes:
+            # 1. ConfigMap volume
+            - name: config-volume
+              configMap:
+                name: document-ai-indexer-config
+                items:
+                  - key: env.yaml
+                    path: env.yaml
+                  - key: config.yaml
+                    path: config.yaml
+
+            # 2. Azure File Share volume
+            - name: data-volume
+              azureFile:
+                secretName: azure-files-cred        # Quote Secret
+                shareName: fs-document-ai-indexer   # Your file share name
+                readOnly: false                     # Write permission
+
+          containers:
+            - name: document-ai-indexer
+              image: acraiflowlab.azurecr.io/document-ai-indexer:2.0.1
+              imagePullPolicy: Always
+              # Mount the volume into the container
+              volumeMounts:
+                # ConfigMap Mount
+                - name: config-volume
+                  mountPath: /app/env.yaml
+                  subPath: env.yaml
+                - name: config-volume
+                  mountPath: /app/config.yaml
+                  subPath: config.yaml
+
+                # Azure File Shared mount
+                - name: data-volume
+                  mountPath: /app/run_tmp            # Program write/read directory

vw-document-ai-indexer/deploy/dev/document-ai-indexer_k8s.yml

@@ -0,0 +1,42 @@
+apiVersion: v1
+kind: Pod
+metadata:
+  name: document-ai-indexer
+spec:
+  restartPolicy: Never
+
+  volumes:
+    # 1. 原有的 ConfigMap 卷
+    - name: config-volume
+      configMap:
+        name: document-ai-indexer-config
+        items:
+          - key: env.yaml
+            path: env.yaml
+          - key: config.yaml
+            path: config.yaml
+
+    # 2. Azure File Share 卷
+    - name: data-volume
+      azureFile:
+        secretName: azure-files-cred        # 引用你创建的 Secret
+        shareName: fs-document-ai-indexer   # 你的文件共享名称
+        readOnly: false                     # 写权限
+
+  containers:
+    - name: document-ai-indexer
+      image: acraiflowlab.azurecr.io/document-ai-indexer:2.0.1
+      imagePullPolicy: Always
+      # 挂载卷到容器内
+      volumeMounts:
+        # ConfigMap 挂载
+        - name: config-volume
+          mountPath: /app/env.yaml
+          subPath: env.yaml
+        - name: config-volume
+          mountPath: /app/config.yaml
+          subPath: config.yaml
+
+        # Azure File 共享挂载
+        - name: data-volume
+          mountPath: /app/run_tmp            # 程序写入/读取目录