Merge branch 'develop' into main

Author: clemlesne

.github/dependabot.yml

@@ -0,0 +1,27 @@
+version: 2
+updates:
+  - package-ecosystem: github-actions
+    directory: ""
+    target-branch: develop
+    schedule:
+      interval: daily
+  - package-ecosystem: pip
+    directory: src/conversation-api
+    target-branch: develop
+    schedule:
+      interval: daily
+  - package-ecosystem: docker
+    directory: src/conversation-api
+    target-branch: develop
+    schedule:
+      interval: daily
+  - package-ecosystem: npm
+    directory: src/conversation-ui
+    target-branch: develop
+    schedule:
+      interval: daily
+  - package-ecosystem: docker
+    directory: src/conversation-ui
+    target-branch: develop
+    schedule:
+      interval: daily

README.md

@@ -1,5 +1,7 @@
 # 🔒 Private GPT
 
+> 👋🏻 Demo available at [private-gpt.shopping-cart-devops-demo.lesne.pro](https://private-gpt.shopping-cart-devops-demo.lesne.pro).
+
 Private GPT is a local version of Chat GPT, using Azure OpenAI. It is an enterprise grade platform to deploy a ChatGPT-like interface for your employees.
 
 Includes:
@@ -9,6 +11,7 @@ Includes:
 - Dead simple interface
 - Deployable on any Kubernetes cluster, with its Helm chart
 - Manage users effortlessly with OpenID Connect
+- Monitoring with Azure App Insights (logs, traces, user behaviors)
 - More than 150 tones and personalities (accountant, advisor, debater, excel sheet, instructor, logistician, etc.) to better help employees in their specific daily tasks
 - Plug and play storage system, including [Azure Cosmos DB](https://learn.microsoft.com/en-us/azure/cosmos-db/), [Redis](https://github.com/redis/redis) and [Qdrant](https://github.com/qdrant/qdrant).
 - Possibility to send temporary messages, for confidentiality
@@ -27,7 +30,7 @@ Create a local configuration file, a file named `config.toml` at the root of the
 
 ```toml
 # config.toml
-# Values are for example only, you should change them
+# /!\ All the file values are for example, you must change them
 [persistence]
 # Enum: "qdrant"
 search = "qdrant"
@@ -39,6 +42,9 @@ stream = "redis"
 [api]
 root_path = ""
 
+[appinsights]
+connection_str = "InstrumentationKey=00000000-0000-0000-0000-000000000000;IngestionEndpoint=https://westeurope-5.in.applicationinsights.azure.com/;LiveEndpoint=https://westeurope.livediagnostics.monitor.azure.com"
+
 [openai]
 ada_deploy_id = "ada"
 ada_max_tokens = 2049

cicd/helm/private-gpt/templates/conversation-api-config.yaml

@@ -15,6 +15,9 @@ data:
     [api]
     root_path = "/{{ include "private-gpt.fullname" . }}-conversation-api"
 
+    [appinsights]
+    connection_str = {{ .Values.appinsights.connection_str | required "A value for .Values.appinsights.connection_str" | quote }}
+
     [openai]
     ada_deploy_id = {{ .Values.api.openai.ada_deploy_id | quote }}
     ada_max_tokens = 2049

cicd/helm/private-gpt/templates/conversation-api-deployment.yaml

@@ -30,28 +30,42 @@ spec:
               containerPort: 8080
               protocol: TCP
           resources: {{- toYaml .Values.resources | nindent 12 | required "A value for .Values.resources is required" }}
+          startupProbe:
+            periodSeconds: 5
+            failureThreshold: 30
+            tcpSocket:
+              port: http
           livenessProbe:
             periodSeconds: 5
             timeoutSeconds: 5
             httpGet:
               path: /health/liveness
               port: http
-          startupProbe:
-            periodSeconds: 5
-            failureThreshold: 30
-            tcpSocket:
+          readinessProbe:
+            periodSeconds: 15
+            timeoutSeconds: 5
+            httpGet:
+              path: /health/readiness
               port: http
           volumeMounts:
             - name: config
               mountPath: /app/config.toml
               subPath: config.toml
+            - name: tmp
+              mountPath: /app/tmp
           env:
             - name: PG_ACS_API_TOKEN
               valueFrom:
                 secretKeyRef:
                   name: {{ include "private-gpt.fullname" . }}-conversation-api
                   key: PG_ACS_API_TOKEN
+            - name: TMPDIR
+              value: /app/tmp
       volumes:
+        # Store app configuration
         - name: config
           configMap:
             name: {{ include "private-gpt.fullname" . }}-conversation-api
+        # Store Azure App Insight telemetry buffer
+        - name: tmp
+          emptyDir: {}

cicd/helm/private-gpt/templates/conversation-api-scaler.yaml

@@ -17,4 +17,4 @@ spec:
     - type: memory
       metadata:
         type: Utilization
-        value: "50"
+        value: "75"

cicd/helm/private-gpt/templates/conversation-ui-config.yaml

@@ -37,6 +37,8 @@ data:
         sub_filter 'https://login.microsoftonline.com/common/v2.0' '{{ .Values.oidc.authority | required "A value for .Values.oidc.authority is required" }}';
         # Rewrite the OIDC API audience
         sub_filter 'e9d5f20f-7f14-4204-a9a2-0d91d6af5c82' '{{ .Values.oidc.api_audience | required "A value for .Values.oidc.api_audience is required" }}';
+        # Rewrite the Azure App Insights connection string
+        sub_filter 'InstrumentationKey=0b860d29-2a55-4d29-ab57-88cdd85a8da0;IngestionEndpoint=https://westeurope-5.in.applicationinsights.azure.com/;LiveEndpoint=https://westeurope.livediagnostics.monitor.azure.com' '{{ .Values.appinsights.connection_str | required "A value for .Values.appinsights.connection_str" }}';
         # Rewrite recursively
         sub_filter_once off;
         # Rewrite every content type

cicd/helm/private-gpt/templates/conversation-ui-scaler.yaml

@@ -17,4 +17,4 @@ spec:
     - type: memory
       metadata:
         type: Utilization
-        value: "50"
+        value: "75"

cicd/helm/private-gpt/values.yaml

@@ -23,6 +23,9 @@ resources:
     cpu: .5
     memory: 512Mi
 
+appinsights:
+  connection_str: null
+
 oidc:
   algorithms: []
   api_audience: null

src/conversation-api/ai/openai.py

@@ -1,6 +1,6 @@
 # Import utils
 from uuid import UUID
-from utils import (build_logger, get_config, hash_token)
+from utils import (build_logger, get_config, hash_token, AZ_CREDENTIAL)
 
 # Import misc
 from azure.identity import DefaultAzureCredential
@@ -33,8 +33,7 @@ async def refresh_oai_token_background():
     """
     while True:
         logger.info("Refreshing OpenAI token")
-        oai_cred = DefaultAzureCredential()
-        oai_token = oai_cred.get_token("https://cognitiveservices.azure.com/.default")
+        oai_token = AZ_CREDENTIAL.get_token("https://cognitiveservices.azure.com/.default")
         openai.api_key = oai_token.token
         # Execute every 20 minutes
         await asyncio.sleep(15 * 60)