fix(citations): icon not visible in ui (#6003)

Co-authored-by: Raunak Bhagat <r@rabh.io>

.github/CODEOWNERS

@@ -1 +1,3 @@
 * @onyx-dot-app/onyx-core-team
+# Helm charts Owners
+/helm/ @justin-tahara

.github/actions/custom-build-and-push/action.yml

@@ -25,12 +25,26 @@ inputs:
   tags:
     description: 'Image tags'
     required: true
+  no-cache:
+    description: 'Read from cache'
+    required: false
+    default: 'false'
   cache-from:
     description: 'Cache sources'
     required: false
   cache-to:
     description: 'Cache destinations'
     required: false
+  outputs:
+    description: 'Output destinations'
+    required: false
+  provenance:
+    description: 'Generate provenance attestation'
+    required: false
+    default: 'false'
+  build-args:
+    description: 'Build arguments'
+    required: false
   retry-wait-time:
     description: 'Time to wait before attempt 2 in seconds'
     required: false
@@ -55,8 +69,12 @@ runs:
         push: ${{ inputs.push }}
         load: ${{ inputs.load }}
         tags: ${{ inputs.tags }}
+        no-cache: ${{ inputs.no-cache }}
         cache-from: ${{ inputs.cache-from }}
         cache-to: ${{ inputs.cache-to }}
+        outputs: ${{ inputs.outputs }}
+        provenance: ${{ inputs.provenance }}
+        build-args: ${{ inputs.build-args }}
 
     - name: Wait before attempt 2
       if: steps.buildx1.outcome != 'success'
@@ -77,8 +95,12 @@ runs:
         push: ${{ inputs.push }}
         load: ${{ inputs.load }}
         tags: ${{ inputs.tags }}
+        no-cache: ${{ inputs.no-cache }}
         cache-from: ${{ inputs.cache-from }}
         cache-to: ${{ inputs.cache-to }}
+        outputs: ${{ inputs.outputs }}
+        provenance: ${{ inputs.provenance }}
+        build-args: ${{ inputs.build-args }}
 
     - name: Wait before attempt 3
       if: steps.buildx1.outcome != 'success' && steps.buildx2.outcome != 'success'
@@ -99,8 +121,12 @@ runs:
         push: ${{ inputs.push }}
         load: ${{ inputs.load }}
         tags: ${{ inputs.tags }}
+        no-cache: ${{ inputs.no-cache }}
         cache-from: ${{ inputs.cache-from }}
         cache-to: ${{ inputs.cache-to }}
+        outputs: ${{ inputs.outputs }}
+        provenance: ${{ inputs.provenance }}
+        build-args: ${{ inputs.build-args }}
 
     - name: Report failure
       if: steps.buildx1.outcome != 'success' && steps.buildx2.outcome != 'success' && steps.buildx3.outcome != 'success'

.github/actions/prepare-build/action.yml

@@ -0,0 +1,50 @@
+name: "Prepare Build (OpenAPI generation)"
+description: "Sets up Python with uv, installs deps, generates OpenAPI schema and Python client, uploads artifact"
+runs:
+  using: "composite"
+  steps:
+    - name: Checkout code
+      uses: actions/checkout@v4
+
+    - name: Setup uv
+      uses: astral-sh/setup-uv@v3
+
+    - name: Setup Python
+      uses: actions/setup-python@v5
+      with:
+        python-version: "3.11"
+
+    - name: Install Python dependencies with uv
+      shell: bash
+      run: |
+        uv pip install --system \
+          -r backend/requirements/default.txt \
+          -r backend/requirements/dev.txt
+
+    - name: Generate OpenAPI schema
+      shell: bash
+      working-directory: backend
+      env:
+        PYTHONPATH: "."
+      run: |
+        python scripts/onyx_openapi_schema.py --filename generated/openapi.json
+
+    - name: Generate OpenAPI Python client
+      shell: bash
+      run: |
+        docker run --rm \
+          -v "${{ github.workspace }}/backend/generated:/local" \
+          openapitools/openapi-generator-cli generate \
+          -i /local/openapi.json \
+          -g python \
+          -o /local/onyx_openapi_client \
+          --package-name onyx_openapi_client \
+          --skip-validate-spec \
+          --openapi-normalizer "SIMPLIFY_ONEOF_ANYOF=true,SET_OAS3_NULLABLE=true"
+
+    - name: Upload OpenAPI artifacts
+      uses: actions/upload-artifact@v4
+      with:
+        name: openapi-artifacts
+        path: backend/generated/
+

.github/pull_request_template.md

@@ -6,9 +6,6 @@
 
 [Describe the tests you ran to verify your changes]
 
-## Backporting (check the box to trigger backport action)
+## Additional Options
 
-Note: You have to check that the action passes, otherwise resolve the conflicts manually and tag the patches.
-
-- [ ] This PR should be backported (make sure to check that the backport attempt succeeds)
 - [ ] [Optional] Override Linear Check

.github/workflows/check-lazy-imports.yml

@@ -0,0 +1,24 @@
+name: Check Lazy Imports
+
+on:
+  merge_group:
+  pull_request:
+    branches:
+      - main
+      - 'release/**'
+
+jobs:
+  check-lazy-imports:
+    runs-on: ubuntu-latest
+
+    steps:
+    - name: Checkout code
+      uses: actions/checkout@v4
+
+    - name: Set up Python
+      uses: actions/setup-python@v4
+      with:
+        python-version: '3.11'
+
+    - name: Check lazy imports
+      run: python3 backend/scripts/check_lazy_imports.py

.github/workflows/docker-build-push-backend-container-on-tag.yml

@@ -7,18 +7,63 @@ on:
 
 env:
   REGISTRY_IMAGE: ${{ contains(github.ref_name, 'cloud') && 'onyxdotapp/onyx-backend-cloud' || 'onyxdotapp/onyx-backend' }}
-  LATEST_TAG: ${{ contains(github.ref_name, 'latest') }}
+  DEPLOYMENT: ${{ contains(github.ref_name, 'cloud') && 'cloud' || 'standalone' }}
+
+  # tag nightly builds with "edge"
+  EDGE_TAG: ${{ startsWith(github.ref_name, 'nightly-latest') }}
 
 jobs:
   build-and-push:
     # TODO: investigate a matrix build like the web container
     # See https://runs-on.com/runners/linux/
-    runs-on: [runs-on, runner=8cpu-linux-x64, "run-id=${{ github.run_id }}"]
-
+    runs-on:
+      - runs-on
+      - runner=${{ matrix.platform == 'linux/amd64' && '8cpu-linux-x64' || '8cpu-linux-arm64' }}
+      - run-id=${{ github.run_id }}
+      - tag=platform-${{ matrix.platform }}
+    strategy:
+      fail-fast: false
+      matrix:
+        platform:
+          - linux/amd64
+          - linux/arm64
+          
     steps:
+      - name: Prepare
+        run: |
+          platform=${{ matrix.platform }}
+          echo "PLATFORM_PAIR=${platform//\//-}" >> $GITHUB_ENV
+
+      - name: Check if stable release version
+        id: check_version
+        run: |
+          if [[ "${{ github.ref_name }}" =~ ^v[0-9]+\.[0-9]+\.[0-9]+$ ]] && [[ "${{ github.ref_name }}" != *"cloud"* ]]; then
+            echo "is_stable=true" >> $GITHUB_OUTPUT
+          else
+            echo "is_stable=false" >> $GITHUB_OUTPUT
+          fi
+          if [[ "${{ github.ref_name }}" =~ ^v[0-9]+\.[0-9]+\.[0-9]+-beta\.[0-9]+$ ]] && [[ "${{ github.ref_name }}" != *"cloud"* ]]; then
+            echo "is_beta=true" >> $GITHUB_OUTPUT
+          else
+            echo "is_beta=false" >> $GITHUB_OUTPUT
+          fi
+
       - name: Checkout code
         uses: actions/checkout@v4
 
+      - name: Docker meta
+        id: meta
+        uses: docker/metadata-action@v5
+        with:
+          images: ${{ env.REGISTRY_IMAGE }}
+          flavor: |
+            latest=false
+          tags: |
+            type=raw,value=${{ github.ref_name }}
+            type=raw,value=${{ steps.check_version.outputs.is_stable == 'true' && 'latest' || '' }}
+            type=raw,value=${{ env.EDGE_TAG == 'true' && 'edge' || '' }}
+            type=raw,value=${{ steps.check_version.outputs.is_beta == 'true' && 'beta' || '' }}
+            
       - name: Set up Docker Buildx
         uses: docker/setup-buildx-action@v3
 
@@ -34,30 +79,120 @@ jobs:
           sudo apt-get install -y build-essential
 
       - name: Backend Image Docker Build and Push
-        uses: docker/build-push-action@v5
+        id: build
+        uses: docker/build-push-action@v6
         with:
           context: ./backend
           file: ./backend/Dockerfile
-          platforms: linux/amd64,linux/arm64
+          platforms: ${{ matrix.platform }}
           push: true
-          tags: |
-            ${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}
-            ${{ env.LATEST_TAG == 'true' && format('{0}:latest', env.REGISTRY_IMAGE) || '' }}
           build-args: |
             ONYX_VERSION=${{ github.ref_name }}
+          labels: ${{ steps.meta.outputs.labels }}
+          outputs: type=image,name=${{ env.REGISTRY_IMAGE }},push-by-digest=true,name-canonical=true,push=true
+          cache-from: type=s3,prefix=cache/${{ github.repository }}/${{ env.DEPLOYMENT }}/backend-${{ env.PLATFORM_PAIR }}/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }}
+          cache-to: type=s3,prefix=cache/${{ github.repository }}/${{ env.DEPLOYMENT }}/backend-${{ env.PLATFORM_PAIR }}/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }},mode=max
 
+      - name: Export digest      
+        run: |
+          mkdir -p /tmp/digests
+          digest="${{ steps.build.outputs.digest }}"
+          touch "/tmp/digests/${digest#sha256:}"
+
+      - name: Upload digest
+        uses: actions/upload-artifact@v4
+        with:
+          name: backend-digests-${{ env.PLATFORM_PAIR }}-${{ github.run_id }}
+          path: /tmp/digests/*
+          if-no-files-found: error
+          retention-days: 1
+          
+  merge:
+    runs-on: ubuntu-latest
+    needs:
+      - build-and-push
+    steps:
+      # Needed for trivyignore
+      - name: Checkout
+        uses: actions/checkout@v4
+      
+      - name: Check if stable release version
+        id: check_version
+        run: |
+          if [[ "${{ github.ref_name }}" =~ ^v[0-9]+\.[0-9]+\.[0-9]+$ ]] && [[ "${{ github.ref_name }}" != *"cloud"* ]]; then
+            echo "is_stable=true" >> $GITHUB_OUTPUT
+          else
+            echo "is_stable=false" >> $GITHUB_OUTPUT
+          fi
+          if [[ "${{ github.ref_name }}" =~ ^v[0-9]+\.[0-9]+\.[0-9]+-beta\.[0-9]+$ ]] && [[ "${{ github.ref_name }}" != *"cloud"* ]]; then
+            echo "is_beta=true" >> $GITHUB_OUTPUT
+          else
+            echo "is_beta=false" >> $GITHUB_OUTPUT
+          fi
+        
+      - name: Download digests
+        uses: actions/download-artifact@v4
+        with:
+          path: /tmp/digests
+          pattern: backend-digests-*-${{ github.run_id }}
+          merge-multiple: true
+
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+
+      - name: Docker meta
+        id: meta
+        uses: docker/metadata-action@v5
+        with:
+          images: ${{ env.REGISTRY_IMAGE }}
+          flavor: |
+            latest=false
+          tags: |
+            type=raw,value=${{ github.ref_name }}
+            type=raw,value=${{ steps.check_version.outputs.is_stable == 'true' && 'latest' || '' }}
+            type=raw,value=${{ env.EDGE_TAG == 'true' && 'edge' || '' }}
+            type=raw,value=${{ steps.check_version.outputs.is_beta == 'true' && 'beta' || '' }}
+
+      - name: Login to Docker Hub
+        uses: docker/login-action@v3
+        with:
+          username: ${{ secrets.DOCKER_USERNAME }}
+          password: ${{ secrets.DOCKER_TOKEN }}
+
+      - name: Create manifest list and push
+        working-directory: /tmp/digests
+        run: |
+          docker buildx imagetools create $(jq -cr '.tags | map("-t " + .) | join(" ")' <<< "$DOCKER_METADATA_OUTPUT_JSON") \
+            $(printf '${{ env.REGISTRY_IMAGE }}@sha256:%s ' *)
+
+      - name: Inspect image
+        run: |
+          docker buildx imagetools inspect ${{ env.REGISTRY_IMAGE }}:${{ steps.meta.outputs.version }}
+          
       # trivy has their own rate limiting issues causing this action to flake
       # we worked around it by hardcoding to different db repos in env
       # can re-enable when they figure it out
       # https://github.com/aquasecurity/trivy/discussions/7538
       # https://github.com/aquasecurity/trivy-action/issues/389
+      # Security: Using pinned digest (0.65.0@sha256:a22415a38938a56c379387a8163fcb0ce38b10ace73e593475d3658d578b2436)
+      # Security: No Docker socket mount needed for remote registry scanning
       - name: Run Trivy vulnerability scanner
-        uses: aquasecurity/trivy-action@master
-        env:
-          TRIVY_DB_REPOSITORY: "public.ecr.aws/aquasecurity/trivy-db:2"
-          TRIVY_JAVA_DB_REPOSITORY: "public.ecr.aws/aquasecurity/trivy-java-db:1"
+        uses: nick-fields/retry@v3
         with:
-          # To run locally: trivy image --severity HIGH,CRITICAL onyxdotapp/onyx-backend
-          image-ref: docker.io/${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}
-          severity: "CRITICAL,HIGH"
-          trivyignores: ./backend/.trivyignore
+          timeout_minutes: 30
+          max_attempts: 3
+          retry_wait_seconds: 10
+          command: |
+            docker run --rm -v $HOME/.cache/trivy:/root/.cache/trivy \
+              -v ${{ github.workspace }}/backend/.trivyignore:/tmp/.trivyignore:ro \
+              -e TRIVY_DB_REPOSITORY="public.ecr.aws/aquasecurity/trivy-db:2" \
+              -e TRIVY_JAVA_DB_REPOSITORY="public.ecr.aws/aquasecurity/trivy-java-db:1" \
+              -e TRIVY_USERNAME="${{ secrets.DOCKER_USERNAME }}" \
+              -e TRIVY_PASSWORD="${{ secrets.DOCKER_TOKEN }}" \
+              aquasec/trivy@sha256:a22415a38938a56c379387a8163fcb0ce38b10ace73e593475d3658d578b2436 \
+              image \
+              --skip-version-check \
+              --timeout 20m \
+              --severity CRITICAL,HIGH \
+              --ignorefile /tmp/.trivyignore \
+              docker.io/${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}

.github/workflows/docker-build-push-cloud-web-container-on-tag.yml

@@ -4,12 +4,12 @@ name: Build and Push Cloud Web Image on Tag
 on:
   push:
     tags:
-      - "*"
+      - "*cloud*"
 
 env:
   REGISTRY_IMAGE: onyxdotapp/onyx-web-server-cloud
-  LATEST_TAG: ${{ contains(github.ref_name, 'latest') }}
-
+  DEPLOYMENT: cloud
+  
 jobs:
   build:
     runs-on:
@@ -38,9 +38,10 @@ jobs:
         uses: docker/metadata-action@v5
         with:
           images: ${{ env.REGISTRY_IMAGE }}
+          flavor: |
+            latest=false
           tags: |
-            type=raw,value=${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}
-            type=raw,value=${{ env.LATEST_TAG == 'true' && format('{0}:latest', env.REGISTRY_IMAGE) || '' }}
+            type=raw,value=${{ github.ref_name }}
 
       - name: Set up Docker Buildx
         uses: docker/setup-buildx-action@v3
@@ -53,7 +54,7 @@ jobs:
 
       - name: Build and push by digest
         id: build
-        uses: docker/build-push-action@v5
+        uses: docker/build-push-action@v6
         with:
           context: ./web
           file: ./web/Dockerfile
@@ -70,10 +71,12 @@ jobs:
             NEXT_PUBLIC_FORGOT_PASSWORD_ENABLED=true
             NEXT_PUBLIC_INCLUDE_ERROR_POPUP_SUPPORT_LINK=true
             NODE_OPTIONS=--max-old-space-size=8192
-          # needed due to weird interactions with the builds for different platforms
-          no-cache: true
           labels: ${{ steps.meta.outputs.labels }}
           outputs: type=image,name=${{ env.REGISTRY_IMAGE }},push-by-digest=true,name-canonical=true,push=true
+          cache-from: type=s3,prefix=cache/${{ github.repository }}/${{ env.DEPLOYMENT }}/cloudweb-${{ env.PLATFORM_PAIR }}/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }}
+          cache-to: type=s3,prefix=cache/${{ github.repository }}/${{ env.DEPLOYMENT }}/cloudweb-${{ env.PLATFORM_PAIR }}/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }},mode=max
+          # no-cache needed due to weird interactions with the builds for different platforms
+          # NOTE(rkuo): this may not be true any more with the proper cache prefixing by architecture - currently testing with it off
 
       - name: Export digest
         run: |
@@ -84,7 +87,7 @@ jobs:
       - name: Upload digest
         uses: actions/upload-artifact@v4
         with:
-          name: digests-${{ env.PLATFORM_PAIR }}
+          name: cloudweb-digests-${{ env.PLATFORM_PAIR }}-${{ github.run_id }}
           path: /tmp/digests/*
           if-no-files-found: error
           retention-days: 1
@@ -98,7 +101,7 @@ jobs:
         uses: actions/download-artifact@v4
         with:
           path: /tmp/digests
-          pattern: digests-*
+          pattern: cloudweb-digests-*-${{ github.run_id }}
           merge-multiple: true
 
       - name: Set up Docker Buildx
@@ -109,6 +112,10 @@ jobs:
         uses: docker/metadata-action@v5
         with:
           images: ${{ env.REGISTRY_IMAGE }}
+          flavor: |
+            latest=false
+          tags: |
+            type=raw,value=${{ github.ref_name }}
 
       - name: Login to Docker Hub
         uses: docker/login-action@v3
@@ -132,10 +139,20 @@ jobs:
       # https://github.com/aquasecurity/trivy/discussions/7538
       # https://github.com/aquasecurity/trivy-action/issues/389
       - name: Run Trivy vulnerability scanner
-        uses: aquasecurity/trivy-action@master
-        env:
-          TRIVY_DB_REPOSITORY: "public.ecr.aws/aquasecurity/trivy-db:2"
-          TRIVY_JAVA_DB_REPOSITORY: "public.ecr.aws/aquasecurity/trivy-java-db:1"
+        uses: nick-fields/retry@v3
         with:
-          image-ref: docker.io/${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}
-          severity: "CRITICAL,HIGH"
+          timeout_minutes: 30
+          max_attempts: 3
+          retry_wait_seconds: 10
+          command: |
+            docker run --rm -v $HOME/.cache/trivy:/root/.cache/trivy \
+              -e TRIVY_DB_REPOSITORY="public.ecr.aws/aquasecurity/trivy-db:2" \
+              -e TRIVY_JAVA_DB_REPOSITORY="public.ecr.aws/aquasecurity/trivy-java-db:1" \
+              -e TRIVY_USERNAME="${{ secrets.DOCKER_USERNAME }}" \
+              -e TRIVY_PASSWORD="${{ secrets.DOCKER_TOKEN }}" \
+              aquasec/trivy@sha256:a22415a38938a56c379387a8163fcb0ce38b10ace73e593475d3658d578b2436 \
+              image \
+              --skip-version-check \
+              --timeout 20m \
+              --severity CRITICAL,HIGH \
+              docker.io/${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}

.github/workflows/docker-build-push-model-server-container-on-tag.yml

@@ -7,10 +7,13 @@ on:
 
 env:
   REGISTRY_IMAGE: ${{ contains(github.ref_name, 'cloud') && 'onyxdotapp/onyx-model-server-cloud' || 'onyxdotapp/onyx-model-server' }}
-  LATEST_TAG: ${{ contains(github.ref_name, 'latest') }}
   DOCKER_BUILDKIT: 1
   BUILDKIT_PROGRESS: plain
+  DEPLOYMENT: ${{ contains(github.ref_name, 'cloud') && 'cloud' || 'standalone' }}
 
+  # tag nightly builds with "edge"
+  EDGE_TAG: ${{ startsWith(github.ref_name, 'nightly-latest') }}
+  
 jobs:
 
 #   Bypassing this for now as the idea of not building is glitching
@@ -51,6 +54,8 @@ jobs:
     if: needs.check_model_server_changes.outputs.changed == 'true'
     runs-on:
       [runs-on, runner=8cpu-linux-x64, "run-id=${{ github.run_id }}-amd64"]
+    env:
+      PLATFORM_PAIR: linux-amd64
     steps:
       - name: Checkout code
         uses: actions/checkout@v4
@@ -75,7 +80,7 @@ jobs:
           password: ${{ secrets.DOCKER_TOKEN }}
 
       - name: Build and Push AMD64
-        uses: docker/build-push-action@v5
+        uses: docker/build-push-action@v6
         with:
           context: ./backend
           file: ./backend/Dockerfile.model_server
@@ -83,15 +88,20 @@ jobs:
           push: true
           tags: ${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}-amd64
           build-args: |
-            DANSWER_VERSION=${{ github.ref_name }}
+            ONYX_VERSION=${{ github.ref_name }}
           outputs: type=registry
           provenance: false
+          cache-from: type=s3,prefix=cache/${{ github.repository }}/${{ env.DEPLOYMENT }}/model-server-${{ env.PLATFORM_PAIR }}/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }}
+          cache-to: type=s3,prefix=cache/${{ github.repository }}/${{ env.DEPLOYMENT }}/model-server-${{ env.PLATFORM_PAIR }}/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }},mode=max
+#           no-cache: true
 
   build-arm64:
     needs: [check_model_server_changes]
     if: needs.check_model_server_changes.outputs.changed == 'true'
     runs-on:
-      [runs-on, runner=8cpu-linux-x64, "run-id=${{ github.run_id }}-arm64"]
+      [runs-on, runner=8cpu-linux-arm64, "run-id=${{ github.run_id }}-arm64"]
+    env:
+      PLATFORM_PAIR: linux-arm64
     steps:
       - name: Checkout code
         uses: actions/checkout@v4
@@ -116,7 +126,7 @@ jobs:
           password: ${{ secrets.DOCKER_TOKEN }}
 
       - name: Build and Push ARM64
-        uses: docker/build-push-action@v5
+        uses: docker/build-push-action@v6
         with:
           context: ./backend
           file: ./backend/Dockerfile.model_server
@@ -124,15 +134,31 @@ jobs:
           push: true
           tags: ${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}-arm64
           build-args: |
-            DANSWER_VERSION=${{ github.ref_name }}
+            ONYX_VERSION=${{ github.ref_name }}
           outputs: type=registry
           provenance: false
+          cache-from: type=s3,prefix=cache/${{ github.repository }}/${{ env.DEPLOYMENT }}/model-server-${{ env.PLATFORM_PAIR }}/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }}
+          cache-to: type=s3,prefix=cache/${{ github.repository }}/${{ env.DEPLOYMENT }}/model-server-${{ env.PLATFORM_PAIR }}/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }},mode=max
 
   merge-and-scan:
     needs: [build-amd64, build-arm64, check_model_server_changes]
     if: needs.check_model_server_changes.outputs.changed == 'true'
     runs-on: ubuntu-latest
     steps:
+      - name: Check if stable release version
+        id: check_version
+        run: |
+          if [[ "${{ github.ref_name }}" =~ ^v[0-9]+\.[0-9]+\.[0-9]+$ ]] && [[ "${{ github.ref_name }}" != *"cloud"* ]]; then
+            echo "is_stable=true" >> $GITHUB_OUTPUT
+          else
+            echo "is_stable=false" >> $GITHUB_OUTPUT
+          fi
+          if [[ "${{ github.ref_name }}" =~ ^v[0-9]+\.[0-9]+\.[0-9]+-beta\.[0-9]+$ ]] && [[ "${{ github.ref_name }}" != *"cloud"* ]]; then
+            echo "is_beta=true" >> $GITHUB_OUTPUT
+          else
+            echo "is_beta=false" >> $GITHUB_OUTPUT
+          fi
+
       - name: Login to Docker Hub
         uses: docker/login-action@v3
         with:
@@ -145,18 +171,37 @@ jobs:
           docker buildx imagetools create -t ${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }} \
             ${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}-amd64 \
             ${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}-arm64
-          if [[ "${{ env.LATEST_TAG }}" == "true" ]]; then
+          if [[ "${{ steps.check_version.outputs.is_stable }}" == "true" ]]; then
             docker buildx imagetools create -t ${{ env.REGISTRY_IMAGE }}:latest \
               ${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}-amd64 \
               ${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}-arm64
           fi
+          if [[ "${{ env.EDGE_TAG }}" == "true" ]]; then
+            docker buildx imagetools create -t ${{ env.REGISTRY_IMAGE }}:edge \
+              ${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}-amd64 \
+              ${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}-arm64
+          fi
+          if [[ "${{ steps.check_version.outputs.is_beta }}" == "true" ]]; then
+            docker buildx imagetools create -t ${{ env.REGISTRY_IMAGE }}:beta \
+              ${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}-amd64 \
+              ${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}-arm64
+          fi
 
       - name: Run Trivy vulnerability scanner
-        uses: aquasecurity/trivy-action@master
-        env:
-          TRIVY_DB_REPOSITORY: "public.ecr.aws/aquasecurity/trivy-db:2"
-          TRIVY_JAVA_DB_REPOSITORY: "public.ecr.aws/aquasecurity/trivy-java-db:1"
+        uses: nick-fields/retry@v3
         with:
-          image-ref: docker.io/${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}
-          severity: "CRITICAL,HIGH"
-          timeout: "10m"
+          timeout_minutes: 30
+          max_attempts: 3
+          retry_wait_seconds: 10
+          command: |
+            docker run --rm -v $HOME/.cache/trivy:/root/.cache/trivy \
+              -e TRIVY_DB_REPOSITORY="public.ecr.aws/aquasecurity/trivy-db:2" \
+              -e TRIVY_JAVA_DB_REPOSITORY="public.ecr.aws/aquasecurity/trivy-java-db:1" \
+              -e TRIVY_USERNAME="${{ secrets.DOCKER_USERNAME }}" \
+              -e TRIVY_PASSWORD="${{ secrets.DOCKER_TOKEN }}" \
+              aquasec/trivy@sha256:a22415a38938a56c379387a8163fcb0ce38b10ace73e593475d3658d578b2436 \
+              image \
+              --skip-version-check \
+              --timeout 20m \
+              --severity CRITICAL,HIGH \
+              docker.io/${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}

.github/workflows/docker-build-push-web-container-on-tag.yml

@@ -7,10 +7,29 @@ on:
 
 env:
   REGISTRY_IMAGE: onyxdotapp/onyx-web-server
-  LATEST_TAG: ${{ contains(github.ref_name, 'latest') }}
+
+  # tag nightly builds with "edge"
+  EDGE_TAG: ${{ startsWith(github.ref_name, 'nightly-latest') }}
+
+  DEPLOYMENT: standalone
 
 jobs:
+  precheck:
+    runs-on: [runs-on, runner=2cpu-linux-x64, "run-id=${{ github.run_id }}"]
+    outputs:
+      should-run: ${{ steps.set-output.outputs.should-run }}
+    steps:
+      - name: Check if tag contains "cloud"
+        id: set-output
+        run: |
+          if [[ "${{ github.ref_name }}" == *cloud* ]]; then
+            echo "should-run=false" >> "$GITHUB_OUTPUT"
+          else
+            echo "should-run=true" >> "$GITHUB_OUTPUT"
+          fi
   build:
+    needs: precheck
+    if: needs.precheck.outputs.should-run == 'true'
     runs-on:
       - runs-on
       - runner=${{ matrix.platform == 'linux/amd64' && '8cpu-linux-x64' || '8cpu-linux-arm64' }}
@@ -29,6 +48,20 @@ jobs:
           platform=${{ matrix.platform }}
           echo "PLATFORM_PAIR=${platform//\//-}" >> $GITHUB_ENV
 
+      - name: Check if stable release version
+        id: check_version
+        run: |
+          if [[ "${{ github.ref_name }}" =~ ^v[0-9]+\.[0-9]+\.[0-9]+$ ]]; then
+            echo "is_stable=true" >> $GITHUB_OUTPUT
+          else
+            echo "is_stable=false" >> $GITHUB_OUTPUT
+          fi
+          if [[ "${{ github.ref_name }}" =~ ^v[0-9]+\.[0-9]+\.[0-9]+-beta\.[0-9]+$ ]]; then
+            echo "is_beta=true" >> $GITHUB_OUTPUT
+          else
+            echo "is_beta=false" >> $GITHUB_OUTPUT
+          fi
+
       - name: Checkout
         uses: actions/checkout@v4
 
@@ -37,9 +70,13 @@ jobs:
         uses: docker/metadata-action@v5
         with:
           images: ${{ env.REGISTRY_IMAGE }}
+          flavor: |
+            latest=false
           tags: |
-            type=raw,value=${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}
-            type=raw,value=${{ env.LATEST_TAG == 'true' && format('{0}:latest', env.REGISTRY_IMAGE) || '' }}
+            type=raw,value=${{ github.ref_name }}
+            type=raw,value=${{ steps.check_version.outputs.is_stable == 'true' && 'latest' || '' }}
+            type=raw,value=${{ env.EDGE_TAG == 'true' && 'edge' || '' }}
+            type=raw,value=${{ steps.check_version.outputs.is_beta == 'true' && 'beta' || '' }}
 
       - name: Set up Docker Buildx
         uses: docker/setup-buildx-action@v3
@@ -52,7 +89,7 @@ jobs:
 
       - name: Build and push by digest
         id: build
-        uses: docker/build-push-action@v5
+        uses: docker/build-push-action@v6
         with:
           context: ./web
           file: ./web/Dockerfile
@@ -62,11 +99,13 @@ jobs:
             ONYX_VERSION=${{ github.ref_name }}
             NODE_OPTIONS=--max-old-space-size=8192
 
-          # needed due to weird interactions with the builds for different platforms
-          no-cache: true
           labels: ${{ steps.meta.outputs.labels }}
           outputs: type=image,name=${{ env.REGISTRY_IMAGE }},push-by-digest=true,name-canonical=true,push=true
-
+          cache-from: type=s3,prefix=cache/${{ github.repository }}/${{ env.DEPLOYMENT }}/web-${{ env.PLATFORM_PAIR }}/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }}
+          cache-to: type=s3,prefix=cache/${{ github.repository }}/${{ env.DEPLOYMENT }}/web-${{ env.PLATFORM_PAIR }}/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }},mode=max
+          # no-cache needed due to weird interactions with the builds for different platforms
+          # NOTE(rkuo): this may not be true any more with the proper cache prefixing by architecture - currently testing with it off
+          
       - name: Export digest
         run: |
           mkdir -p /tmp/digests
@@ -76,21 +115,36 @@ jobs:
       - name: Upload digest
         uses: actions/upload-artifact@v4
         with:
-          name: digests-${{ env.PLATFORM_PAIR }}
+          name: web-digests-${{ env.PLATFORM_PAIR }}-${{ github.run_id }}
           path: /tmp/digests/*
           if-no-files-found: error
           retention-days: 1
 
   merge:
-    runs-on: ubuntu-latest
     needs:
       - build
+    if: needs.precheck.outputs.should-run == 'true'
+    runs-on: ubuntu-latest
     steps:
+      - name: Check if stable release version
+        id: check_version
+        run: |
+          if [[ "${{ github.ref_name }}" =~ ^v[0-9]+\.[0-9]+\.[0-9]+$ ]] && [[ "${{ github.ref_name }}" != *"cloud"* ]]; then
+            echo "is_stable=true" >> $GITHUB_OUTPUT
+          else
+            echo "is_stable=false" >> $GITHUB_OUTPUT
+          fi
+          if [[ "${{ github.ref_name }}" =~ ^v[0-9]+\.[0-9]+\.[0-9]+-beta\.[0-9]+$ ]]; then
+            echo "is_beta=true" >> $GITHUB_OUTPUT
+          else
+            echo "is_beta=false" >> $GITHUB_OUTPUT
+          fi
+        
       - name: Download digests
         uses: actions/download-artifact@v4
         with:
           path: /tmp/digests
-          pattern: digests-*
+          pattern: web-digests-*-${{ github.run_id }}
           merge-multiple: true
 
       - name: Set up Docker Buildx
@@ -101,6 +155,13 @@ jobs:
         uses: docker/metadata-action@v5
         with:
           images: ${{ env.REGISTRY_IMAGE }}
+          flavor: |
+            latest=false
+          tags: |
+            type=raw,value=${{ github.ref_name }}
+            type=raw,value=${{ steps.check_version.outputs.is_stable == 'true' && 'latest' || '' }}
+            type=raw,value=${{ env.EDGE_TAG == 'true' && 'edge' || '' }}
+            type=raw,value=${{ steps.check_version.outputs.is_beta == 'true' && 'beta' || '' }}
 
       - name: Login to Docker Hub
         uses: docker/login-action@v3
@@ -124,10 +185,20 @@ jobs:
       # https://github.com/aquasecurity/trivy/discussions/7538
       # https://github.com/aquasecurity/trivy-action/issues/389
       - name: Run Trivy vulnerability scanner
-        uses: aquasecurity/trivy-action@master
-        env:
-          TRIVY_DB_REPOSITORY: "public.ecr.aws/aquasecurity/trivy-db:2"
-          TRIVY_JAVA_DB_REPOSITORY: "public.ecr.aws/aquasecurity/trivy-java-db:1"
+        uses: nick-fields/retry@v3
         with:
-          image-ref: docker.io/${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}
-          severity: "CRITICAL,HIGH"
+          timeout_minutes: 30
+          max_attempts: 3
+          retry_wait_seconds: 10
+          command: |
+            docker run --rm -v $HOME/.cache/trivy:/root/.cache/trivy \
+              -e TRIVY_DB_REPOSITORY="public.ecr.aws/aquasecurity/trivy-db:2" \
+              -e TRIVY_JAVA_DB_REPOSITORY="public.ecr.aws/aquasecurity/trivy-java-db:1" \
+              -e TRIVY_USERNAME="${{ secrets.DOCKER_USERNAME }}" \
+              -e TRIVY_PASSWORD="${{ secrets.DOCKER_TOKEN }}" \
+              aquasec/trivy@sha256:a22415a38938a56c379387a8163fcb0ce38b10ace73e593475d3658d578b2436 \
+              image \
+              --skip-version-check \
+              --timeout 20m \
+              --severity CRITICAL,HIGH \
+              docker.io/${{ env.REGISTRY_IMAGE }}:${{ github.ref_name }}

.github/workflows/docker-tag-beta.yml

@@ -0,0 +1,41 @@
+# This workflow is set up to be manually triggered via the GitHub Action tab.
+# Given a version, it will tag those backend and webserver images as "beta".
+
+name: Tag Beta Version
+
+on:
+  workflow_dispatch:
+    inputs:
+      version:
+        description: "The version (ie v1.0.0-beta.0) to tag as beta"
+        required: true
+
+jobs:
+  tag:
+    # See https://runs-on.com/runners/linux/
+    # use a lower powered instance since this just does i/o to docker hub
+    runs-on: [runs-on, runner=2cpu-linux-x64, "run-id=${{ github.run_id }}"]
+    steps:
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v1
+
+      - name: Login to Docker Hub
+        uses: docker/login-action@v1
+        with:
+          username: ${{ secrets.DOCKER_USERNAME }}
+          password: ${{ secrets.DOCKER_TOKEN }}
+
+      - name: Enable Docker CLI experimental features
+        run: echo "DOCKER_CLI_EXPERIMENTAL=enabled" >> $GITHUB_ENV
+
+      - name: Pull, Tag and Push Web Server Image
+        run: |
+          docker buildx imagetools create -t onyxdotapp/onyx-web-server:beta onyxdotapp/onyx-web-server:${{ github.event.inputs.version }}
+
+      - name: Pull, Tag and Push API Server Image
+        run: |
+          docker buildx imagetools create -t onyxdotapp/onyx-backend:beta onyxdotapp/onyx-backend:${{ github.event.inputs.version }}
+
+      - name: Pull, Tag and Push Model Server Image
+        run: |
+          docker buildx imagetools create -t onyxdotapp/onyx-model-server:beta onyxdotapp/onyx-model-server:${{ github.event.inputs.version }}

.github/workflows/docker-tag-latest.yml

@@ -35,3 +35,7 @@ jobs:
       - name: Pull, Tag and Push API Server Image
         run: |
           docker buildx imagetools create -t onyxdotapp/onyx-backend:latest onyxdotapp/onyx-backend:${{ github.event.inputs.version }}
+
+      - name: Pull, Tag and Push Model Server Image
+        run: |
+          docker buildx imagetools create -t onyxdotapp/onyx-model-server:latest onyxdotapp/onyx-model-server:${{ github.event.inputs.version }}

.github/workflows/helm-chart-releases.yml

@@ -0,0 +1,52 @@
+name: Release Onyx Helm Charts
+
+on:
+  push:
+    branches:
+      - main
+
+permissions: write-all
+
+jobs:
+  release:
+    permissions:
+      contents: write
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - name: Install Helm CLI
+        uses: azure/setup-helm@v4
+        with:
+          version: v3.12.1
+
+      - name: Add required Helm repositories
+        run: |
+          helm repo add ingress-nginx https://kubernetes.github.io/ingress-nginx
+          helm repo add onyx-vespa https://onyx-dot-app.github.io/vespa-helm-charts
+          helm repo add cloudnative-pg https://cloudnative-pg.github.io/charts
+          helm repo add ot-container-kit https://ot-container-kit.github.io/helm-charts
+          helm repo add minio https://charts.min.io/
+          helm repo update
+
+      - name: Build chart dependencies
+        run: |
+          set -euo pipefail
+          for chart_dir in deployment/helm/charts/*; do
+            if [ -f "$chart_dir/Chart.yaml" ]; then
+              echo "Building dependencies for $chart_dir"
+              helm dependency build "$chart_dir"
+            fi
+          done
+
+      - name: Publish Helm charts to gh-pages
+        uses: stefanprodan/helm-gh-pages@v1.7.0
+        with:
+          token: ${{ secrets.GITHUB_TOKEN }}
+          charts_dir: deployment/helm/charts
+          branch: gh-pages
+          commit_username: ${{ github.actor }}
+          commit_email: ${{ github.actor }}@users.noreply.github.com

.github/workflows/hotfix-release-branches.yml

@@ -1,171 +0,0 @@
-# This workflow is intended to be manually triggered via the GitHub Action tab.
-# Given a hotfix branch, it will attempt to open a PR to all release branches and
-# by default auto merge them
-
-name: Hotfix release branches
-
-on:
-  workflow_dispatch:
-    inputs:
-      hotfix_commit:
-        description: "Hotfix commit hash"
-        required: true
-      hotfix_suffix:
-        description: "Hotfix branch suffix (e.g. hotfix/v0.8-{suffix})"
-        required: true
-      release_branch_pattern:
-        description: "Release branch pattern (regex)"
-        required: true
-        default: "release/.*"
-      auto_merge:
-        description: "Automatically merge the hotfix PRs"
-        required: true
-        type: choice
-        default: "true"
-        options:
-          - true
-          - false
-
-jobs:
-  hotfix_release_branches:
-    permissions: write-all
-    # See https://runs-on.com/runners/linux/
-    # use a lower powered instance since this just does i/o to docker hub
-    runs-on: [runs-on, runner=2cpu-linux-x64, "run-id=${{ github.run_id }}"]
-    steps:
-      # needs RKUO_DEPLOY_KEY for write access to merge PR's
-      - name: Checkout Repository
-        uses: actions/checkout@v4
-        with:
-          ssh-key: "${{ secrets.RKUO_DEPLOY_KEY }}"
-          fetch-depth: 0
-
-      - name: Set up Git user
-        run: |
-          git config user.name "Richard Kuo [bot]"
-          git config user.email "rkuo[bot]@onyx.app"
-
-      - name: Fetch All Branches
-        run: |
-          git fetch --all --prune
-
-      - name: Verify Hotfix Commit Exists
-        run: |
-          git rev-parse --verify "${{ github.event.inputs.hotfix_commit }}" || { echo "Commit not found: ${{ github.event.inputs.hotfix_commit }}"; exit 1; }
-
-      - name: Get Release Branches
-        id: get_release_branches
-        run: |
-          BRANCHES=$(git branch -r | grep -E "${{ github.event.inputs.release_branch_pattern }}" | sed 's|origin/||' | tr -d ' ')
-          if [ -z "$BRANCHES" ]; then
-            echo "No release branches found matching pattern '${{ github.event.inputs.release_branch_pattern }}'."
-            exit 1
-          fi
-
-          echo "Found release branches:"
-          echo "$BRANCHES"
-
-          # Join the branches into a single line separated by commas
-          BRANCHES_JOINED=$(echo "$BRANCHES" | tr '\n' ',' | sed 's/,$//')
-
-          # Set the branches as an output
-          echo "branches=$BRANCHES_JOINED" >> $GITHUB_OUTPUT
-
-      # notes on all the vagaries of wiring up automated PR's
-      # https://github.com/peter-evans/create-pull-request/blob/main/docs/concepts-guidelines.md#triggering-further-workflow-runs
-      # we must use a custom token for GH_TOKEN to trigger the subsequent PR checks
-      - name: Create and Merge Pull Requests to Matching Release Branches
-        env:
-          HOTFIX_COMMIT: ${{ github.event.inputs.hotfix_commit }}
-          HOTFIX_SUFFIX: ${{ github.event.inputs.hotfix_suffix }}
-          AUTO_MERGE: ${{ github.event.inputs.auto_merge }}
-          GH_TOKEN: ${{ secrets.RKUO_PERSONAL_ACCESS_TOKEN }}
-        run: |
-          # Get the branches from the previous step
-          BRANCHES="${{ steps.get_release_branches.outputs.branches }}"
-
-          # Convert BRANCHES to an array
-          IFS=$',' read -ra BRANCH_ARRAY <<< "$BRANCHES"
-
-          # Loop through each release branch and create and merge a PR
-          for RELEASE_BRANCH in "${BRANCH_ARRAY[@]}"; do
-            echo "Processing $RELEASE_BRANCH..."
-            
-            # Parse out the release version by removing "release/" from the branch name
-            RELEASE_VERSION=${RELEASE_BRANCH#release/}
-            echo "Release version parsed: $RELEASE_VERSION"
-            
-            HOTFIX_BRANCH="hotfix/${RELEASE_VERSION}-${HOTFIX_SUFFIX}"
-            echo "Creating PR from $HOTFIX_BRANCH to $RELEASE_BRANCH"
-
-            # Checkout the release branch
-            echo "Checking out $RELEASE_BRANCH"
-            git checkout "$RELEASE_BRANCH"
-
-            # Create the new hotfix branch
-            if git rev-parse --verify "$HOTFIX_BRANCH" >/dev/null 2>&1; then
-              echo "Hotfix branch $HOTFIX_BRANCH already exists. Skipping branch creation."
-            else
-              echo "Branching $RELEASE_BRANCH to $HOTFIX_BRANCH"
-              git checkout -b "$HOTFIX_BRANCH"
-            fi
-            
-            # Check if the hotfix commit is a merge commit
-            if git rev-list --merges -n 1 "$HOTFIX_COMMIT" >/dev/null 2>&1; then
-              # -m 1 uses the target branch as the base (which is what we want)
-              echo "Hotfix commit $HOTFIX_COMMIT is a merge commit, using -m 1 for cherry-pick"
-              CHERRY_PICK_CMD="git cherry-pick -m 1 $HOTFIX_COMMIT"
-            else
-              CHERRY_PICK_CMD="git cherry-pick $HOTFIX_COMMIT"
-            fi
-
-            # Perform the cherry-pick
-            echo "Executing: $CHERRY_PICK_CMD"
-            eval "$CHERRY_PICK_CMD"
-
-            if [ $? -ne 0 ]; then
-              echo "Cherry-pick failed for $HOTFIX_COMMIT on $HOTFIX_BRANCH. Aborting..."
-              git cherry-pick --abort
-              continue
-            fi
-
-            # Push the hotfix branch to the remote
-            echo "Pushing $HOTFIX_BRANCH..."
-            git push origin "$HOTFIX_BRANCH"
-            echo "Hotfix branch $HOTFIX_BRANCH created and pushed."
-            
-            # Check if PR already exists
-            EXISTING_PR=$(gh pr list --head "$HOTFIX_BRANCH" --base "$RELEASE_BRANCH" --state open --json number --jq '.[0].number')
-
-            if [ -n "$EXISTING_PR" ]; then
-              echo "An open PR already exists: #$EXISTING_PR. Skipping..."
-              continue
-            fi
-            
-            # Create a new PR and capture the output
-            PR_OUTPUT=$(gh pr create --title "Merge $HOTFIX_BRANCH into $RELEASE_BRANCH" \
-              --body "Automated PR to merge \`$HOTFIX_BRANCH\` into \`$RELEASE_BRANCH\`." \
-              --head "$HOTFIX_BRANCH" --base "$RELEASE_BRANCH")
-
-            # Extract the URL from the output
-            PR_URL=$(echo "$PR_OUTPUT" | grep -Eo 'https://github.com/[^ ]+')
-            echo "Pull request created: $PR_URL"
-
-            # Extract PR number from URL
-            PR_NUMBER=$(basename "$PR_URL")
-            echo "Pull request created: $PR_NUMBER"
-
-            if [ "$AUTO_MERGE" == "true" ]; then
-              echo "Attempting to merge pull request #$PR_NUMBER"
-
-              # Attempt to merge the PR
-              gh pr merge "$PR_NUMBER" --merge --auto --delete-branch
-
-              if [ $? -eq 0 ]; then
-                echo "Pull request #$PR_NUMBER merged successfully."
-              else
-                # Optionally, handle the error or continue
-                echo "Failed to merge pull request #$PR_NUMBER."
-              fi
-            fi
-          done

.github/workflows/pr-backport-autotrigger.yml

@@ -1,124 +0,0 @@
-name: Backport on Merge
-
-# Note this workflow does not trigger the builds, be sure to manually tag the branches to trigger the builds
-
-on:
-  pull_request:
-    types: [closed] # Later we check for merge so only PRs that go in can get backported
-
-permissions:
-  contents: write
-  actions: write
-
-jobs:
-  backport:
-    if: github.event.pull_request.merged == true
-    runs-on: ubuntu-latest
-    env:
-      GITHUB_TOKEN: ${{ secrets.YUHONG_GH_ACTIONS }}
-    steps:
-      - name: Checkout code
-        uses: actions/checkout@v4
-        with:
-          ssh-key: "${{ secrets.RKUO_DEPLOY_KEY }}"
-          fetch-depth: 0
-
-      - name: Set up Git user
-        run: |
-          git config user.name "Richard Kuo [bot]"
-          git config user.email "rkuo[bot]@onyx.app"
-          git fetch --prune
-
-      - name: Check for Backport Checkbox
-        id: checkbox-check
-        run: |
-          PR_BODY="${{ github.event.pull_request.body }}"
-          if [[ "$PR_BODY" == *"[x] This PR should be backported"* ]]; then
-            echo "backport=true" >> $GITHUB_OUTPUT
-          else
-            echo "backport=false" >> $GITHUB_OUTPUT
-          fi
-
-      - name: List and sort release branches
-        id: list-branches
-        run: |
-          git fetch --all --tags
-          BRANCHES=$(git for-each-ref --format='%(refname:short)' refs/remotes/origin/release/* | sed 's|origin/release/||' | sort -Vr)
-          BETA=$(echo "$BRANCHES" | head -n 1)
-          STABLE=$(echo "$BRANCHES" | head -n 2 | tail -n 1)
-          echo "beta=release/$BETA" >> $GITHUB_OUTPUT
-          echo "stable=release/$STABLE" >> $GITHUB_OUTPUT
-          # Fetch latest tags for beta and stable
-          LATEST_BETA_TAG=$(git tag -l "v[0-9]*.[0-9]*.[0-9]*-beta.[0-9]*" | grep -E "^v[0-9]+\.[0-9]+\.[0-9]+-beta\.[0-9]+$" | grep -v -- "-cloud" | sort -Vr | head -n 1)
-          LATEST_STABLE_TAG=$(git tag -l "v[0-9]*.[0-9]*.[0-9]*" | grep -E "^v[0-9]+\.[0-9]+\.[0-9]+$" | sort -Vr | head -n 1)
-
-          # Handle case where no beta tags exist
-          if [[ -z "$LATEST_BETA_TAG" ]]; then
-            NEW_BETA_TAG="v1.0.0-beta.1"
-          else
-            NEW_BETA_TAG=$(echo $LATEST_BETA_TAG | awk -F '[.-]' '{print $1 "." $2 "." $3 "-beta." ($NF+1)}')
-          fi
-
-          # Increment latest stable tag
-          NEW_STABLE_TAG=$(echo $LATEST_STABLE_TAG | awk -F '.' '{print $1 "." $2 "." ($3+1)}')
-          echo "latest_beta_tag=$LATEST_BETA_TAG" >> $GITHUB_OUTPUT
-          echo "latest_stable_tag=$LATEST_STABLE_TAG" >> $GITHUB_OUTPUT
-          echo "new_beta_tag=$NEW_BETA_TAG" >> $GITHUB_OUTPUT
-          echo "new_stable_tag=$NEW_STABLE_TAG" >> $GITHUB_OUTPUT
-
-      - name: Echo branch and tag information
-        run: |
-          echo "Beta branch: ${{ steps.list-branches.outputs.beta }}"
-          echo "Stable branch: ${{ steps.list-branches.outputs.stable }}"
-          echo "Latest beta tag: ${{ steps.list-branches.outputs.latest_beta_tag }}"
-          echo "Latest stable tag: ${{ steps.list-branches.outputs.latest_stable_tag }}"
-          echo "New beta tag: ${{ steps.list-branches.outputs.new_beta_tag }}"
-          echo "New stable tag: ${{ steps.list-branches.outputs.new_stable_tag }}"
-
-      - name: Trigger Backport
-        if: steps.checkbox-check.outputs.backport == 'true'
-        run: |
-          set -e
-          echo "Backporting to beta ${{ steps.list-branches.outputs.beta }} and stable ${{ steps.list-branches.outputs.stable }}"
-
-          # Echo the merge commit SHA
-          echo "Merge commit SHA: ${{ github.event.pull_request.merge_commit_sha }}"
-
-          # Fetch all history for all branches and tags
-          git fetch --prune
-
-          # Reset and prepare the beta branch
-          git checkout ${{ steps.list-branches.outputs.beta }}
-          echo "Last 5 commits on beta branch:"
-          git log -n 5 --pretty=format:"%H"
-          echo ""  # Newline for formatting
-
-          # Cherry-pick the merge commit from the merged PR
-          git cherry-pick -m 1 ${{ github.event.pull_request.merge_commit_sha }} || {
-            echo "Cherry-pick to beta failed due to conflicts."
-            exit 1
-          }
-
-          # Create new beta branch/tag
-          git tag ${{ steps.list-branches.outputs.new_beta_tag }}
-          # Push the changes and tag to the beta branch using PAT
-          git push origin ${{ steps.list-branches.outputs.beta }}
-          git push origin ${{ steps.list-branches.outputs.new_beta_tag }}
-
-          # Reset and prepare the stable branch
-          git checkout ${{ steps.list-branches.outputs.stable }}
-          echo "Last 5 commits on stable branch:"
-          git log -n 5 --pretty=format:"%H"
-          echo ""  # Newline for formatting
-
-          # Cherry-pick the merge commit from the merged PR
-          git cherry-pick -m 1 ${{ github.event.pull_request.merge_commit_sha }} || {
-            echo "Cherry-pick to stable failed due to conflicts."
-            exit 1
-          }
-
-          # Create new stable branch/tag
-          git tag ${{ steps.list-branches.outputs.new_stable_tag }}
-          # Push the changes and tag to the stable branch using PAT
-          git push origin ${{ steps.list-branches.outputs.stable }}
-          git push origin ${{ steps.list-branches.outputs.new_stable_tag }}

.github/workflows/pr-external-dependency-unit-tests.yml

@@ -0,0 +1,110 @@
+name: External Dependency Unit Tests
+
+on:
+  merge_group:
+  pull_request:
+    branches: [main]
+
+env:
+  # AWS
+  S3_AWS_ACCESS_KEY_ID: ${{ secrets.S3_AWS_ACCESS_KEY_ID }}
+  S3_AWS_SECRET_ACCESS_KEY: ${{ secrets.S3_AWS_SECRET_ACCESS_KEY }}
+
+  # MinIO
+  S3_ENDPOINT_URL: "http://localhost:9004"
+
+  # Confluence
+  CONFLUENCE_TEST_SPACE_URL: ${{ vars.CONFLUENCE_TEST_SPACE_URL }}
+  CONFLUENCE_TEST_SPACE: ${{ vars.CONFLUENCE_TEST_SPACE }}
+  CONFLUENCE_TEST_PAGE_ID: ${{ secrets.CONFLUENCE_TEST_PAGE_ID }}
+  CONFLUENCE_USER_NAME: ${{ vars.CONFLUENCE_USER_NAME }}
+  CONFLUENCE_ACCESS_TOKEN: ${{ secrets.CONFLUENCE_ACCESS_TOKEN }}
+  CONFLUENCE_ACCESS_TOKEN_SCOPED: ${{ secrets.CONFLUENCE_ACCESS_TOKEN_SCOPED }}
+
+  # LLMs
+  OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
+  ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
+
+jobs:
+  discover-test-dirs:
+    runs-on: ubuntu-latest
+    outputs:
+      test-dirs: ${{ steps.set-matrix.outputs.test-dirs }}
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+      
+      - name: Discover test directories
+        id: set-matrix
+        run: |
+          # Find all subdirectories in backend/tests/external_dependency_unit
+          dirs=$(find backend/tests/external_dependency_unit -mindepth 1 -maxdepth 1 -type d -exec basename {} \; | sort | jq -R -s -c 'split("\n")[:-1]')
+          echo "test-dirs=$dirs" >> $GITHUB_OUTPUT
+
+  external-dependency-unit-tests:
+    needs: discover-test-dirs
+    # Use larger runner with more resources for Vespa
+    runs-on: [runs-on, runner=16cpu-linux-x64, "run-id=${{ github.run_id }}"]
+    
+    strategy:
+      fail-fast: false
+      matrix:
+        test-dir: ${{ fromJson(needs.discover-test-dirs.outputs.test-dirs) }}
+
+    env:
+      PYTHONPATH: ./backend
+      MODEL_SERVER_HOST: "disabled"
+
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.11"
+          cache: "pip"
+          cache-dependency-path: |
+            backend/requirements/default.txt
+            backend/requirements/dev.txt
+
+      - name: Install Dependencies
+        run: |
+          python -m pip install --upgrade pip
+          pip install --retries 5 --timeout 30 -r backend/requirements/default.txt
+          pip install --retries 5 --timeout 30 -r backend/requirements/dev.txt
+          playwright install chromium
+          playwright install-deps chromium
+
+      - name: Set up Standard Dependencies
+        run: |
+          cd deployment/docker_compose
+          docker compose -f docker-compose.yml -f docker-compose.dev.yml up -d minio relational_db cache index
+
+      - name: Wait for services
+        run: |
+          echo "Waiting for services to be ready..."
+          sleep 30
+          
+          # Wait for Vespa specifically
+          echo "Waiting for Vespa to be ready..."
+          timeout 300 bash -c 'until curl -f -s http://localhost:8081/ApplicationStatus > /dev/null 2>&1; do echo "Vespa not ready, waiting..."; sleep 10; done' || echo "Vespa timeout - continuing anyway"
+          
+          echo "Services should be ready now"
+
+      - name: Run migrations
+        run: |
+          cd backend
+          # Run migrations to head
+          alembic upgrade head
+          alembic heads --verbose
+
+      - name: Run Tests for ${{ matrix.test-dir }}
+        shell: script -q -e -c "bash --noprofile --norc -eo pipefail {0}"
+        run: |
+          py.test \
+            --durations=8 \
+            -o junit_family=xunit2 \
+            -xv \
+            --ff \
+            backend/tests/external_dependency_unit/${{ matrix.test-dir }}

.github/workflows/pr-helm-chart-testing.yml

@@ -19,9 +19,9 @@ jobs:
         fetch-depth: 0
         
     - name: Set up Helm
-      uses: azure/setup-helm@v4.2.0
+      uses: azure/setup-helm@v4.3.1
       with:
-        version: v3.17.0
+        version: v3.19.0
       
     - name: Set up chart-testing
       uses: helm/chart-testing-action@v2.7.0
@@ -37,6 +37,11 @@ jobs:
           echo "changed=true" >> "$GITHUB_OUTPUT"
         fi
 
+    # uncomment to force run chart-testing
+#     - name: Force run chart-testing (list-changed)
+#       id: list-changed
+#       run: echo "changed=true" >> $GITHUB_OUTPUT
+        
     # lint all charts if any changes were detected
     - name: Run chart-testing (lint)
       if: steps.list-changed.outputs.changed == 'true'
@@ -48,9 +53,176 @@ jobs:
       if: steps.list-changed.outputs.changed == 'true'
       uses: helm/kind-action@v1.12.0
 
-    - name: Run chart-testing (install)
+    - name: Pre-install cluster status check
       if: steps.list-changed.outputs.changed == 'true'
-      run: ct install --all --helm-extra-set-args="--set=nginx.enabled=false" --debug --config ct.yaml
+      run: |
+        echo "=== Pre-install Cluster Status ==="
+        kubectl get nodes -o wide
+        kubectl get pods --all-namespaces
+        kubectl get storageclass
+
+    - name: Add Helm repositories and update
+      if: steps.list-changed.outputs.changed == 'true'
+      run: |
+        echo "=== Adding Helm repositories ==="
+        helm repo add ingress-nginx https://kubernetes.github.io/ingress-nginx
+        helm repo add vespa https://onyx-dot-app.github.io/vespa-helm-charts
+        helm repo add cloudnative-pg https://cloudnative-pg.github.io/charts
+        helm repo add ot-container-kit https://ot-container-kit.github.io/helm-charts
+        helm repo add minio https://charts.min.io/
+        helm repo update
+
+    - name: Install Redis operator
+      if: steps.list-changed.outputs.changed == 'true'
+      shell: bash
+      run: |
+        echo "=== Installing redis-operator CRDs ==="
+        helm upgrade --install redis-operator ot-container-kit/redis-operator \
+          --namespace redis-operator --create-namespace --wait --timeout 300s
+
+    - name: Pre-pull required images
+      if: steps.list-changed.outputs.changed == 'true'
+      run: |
+        echo "=== Pre-pulling required images to avoid timeout ==="
+        KIND_CLUSTER=$(kubectl config current-context | sed 's/kind-//')
+        echo "Kind cluster: $KIND_CLUSTER"
+
+        IMAGES=(
+          "ghcr.io/cloudnative-pg/cloudnative-pg:1.27.0"
+          "quay.io/opstree/redis:v7.0.15"
+          "docker.io/onyxdotapp/onyx-web-server:latest"
+        )
+
+        for image in "${IMAGES[@]}"; do
+          echo "Pre-pulling $image"
+          if docker pull "$image"; then
+            kind load docker-image "$image" --name "$KIND_CLUSTER" || echo "Failed to load $image into kind"
+          else
+            echo "Failed to pull $image"
+          fi
+        done
+
+        echo "=== Images loaded into Kind cluster ==="
+        docker exec "$KIND_CLUSTER"-control-plane crictl images | grep -E "(cloudnative-pg|redis|onyx)" || echo "Some images may still be loading..."
+
+    - name: Validate chart dependencies
+      if: steps.list-changed.outputs.changed == 'true'
+      run: |
+        echo "=== Validating chart dependencies ==="
+        cd deployment/helm/charts/onyx
+        helm dependency update
+        helm lint .
+
+    - name: Run chart-testing (install) with enhanced monitoring
+      timeout-minutes: 25
+      if: steps.list-changed.outputs.changed == 'true'
+      run: |
+        echo "=== Starting chart installation with monitoring ==="
+        
+        # Function to monitor cluster state
+        monitor_cluster() {
+          while true; do
+            echo "=== Cluster Status Check at $(date) ==="
+            # Only show non-running pods to reduce noise
+            NON_RUNNING_PODS=$(kubectl get pods --all-namespaces --field-selector=status.phase!=Running,status.phase!=Succeeded --no-headers 2>/dev/null | wc -l)
+            if [ "$NON_RUNNING_PODS" -gt 0 ]; then
+              echo "Non-running pods:"
+              kubectl get pods --all-namespaces --field-selector=status.phase!=Running,status.phase!=Succeeded
+            else
+              echo "All pods running successfully"
+            fi
+            # Only show recent events if there are issues
+            RECENT_EVENTS=$(kubectl get events --sort-by=.lastTimestamp --all-namespaces --field-selector=type!=Normal 2>/dev/null | tail -5)
+            if [ -n "$RECENT_EVENTS" ]; then
+              echo "Recent warnings/errors:"
+              echo "$RECENT_EVENTS"
+            fi
+            sleep 60
+          done
+        }
+        
+        # Start monitoring in background
+        monitor_cluster &
+        MONITOR_PID=$!
+        
+        # Set up cleanup
+        cleanup() {
+          echo "=== Cleaning up monitoring process ==="
+          kill $MONITOR_PID 2>/dev/null || true
+          echo "=== Final cluster state ==="
+          kubectl get pods --all-namespaces
+          kubectl get events --all-namespaces --sort-by=.lastTimestamp | tail -20
+        }
+        
+        # Trap cleanup on exit
+        trap cleanup EXIT
+        
+        # Run the actual installation with detailed logging
+        echo "=== Starting ct install ==="
+        set +e
+        ct install --all \
+          --helm-extra-set-args="\
+            --set=nginx.enabled=false \
+            --set=minio.enabled=false \
+            --set=vespa.enabled=false \
+            --set=slackbot.enabled=false \
+            --set=postgresql.enabled=true \
+            --set=postgresql.nameOverride=cloudnative-pg \
+            --set=postgresql.cluster.storage.storageClass=standard \
+            --set=redis.enabled=true \
+            --set=redis.storageSpec.volumeClaimTemplate.spec.storageClassName=standard \
+            --set=webserver.replicaCount=1 \
+            --set=api.replicaCount=0 \
+            --set=inferenceCapability.replicaCount=0 \
+            --set=indexCapability.replicaCount=0 \
+            --set=celery_beat.replicaCount=0 \
+            --set=celery_worker_heavy.replicaCount=0 \
+            --set=celery_worker_docfetching.replicaCount=0 \
+            --set=celery_worker_docprocessing.replicaCount=0 \
+            --set=celery_worker_light.replicaCount=0 \
+            --set=celery_worker_monitoring.replicaCount=0 \
+            --set=celery_worker_primary.replicaCount=0 \
+            --set=celery_worker_user_file_processing.replicaCount=0 \
+            --set=celery_worker_user_files_indexing.replicaCount=0" \
+          --helm-extra-args="--timeout 900s --debug" \
+          --debug --config ct.yaml
+        CT_EXIT=$?
+        set -e
+
+        if [[ $CT_EXIT -ne 0 ]]; then
+          echo "ct install failed with exit code $CT_EXIT"
+          exit $CT_EXIT
+        else
+          echo "=== Installation completed successfully ==="
+        fi
+
+        kubectl get pods --all-namespaces
+
+    - name: Post-install verification
+      if: steps.list-changed.outputs.changed == 'true'
+      run: |
+        echo "=== Post-install verification ==="
+        kubectl get pods --all-namespaces
+        kubectl get services --all-namespaces
+        # Only show issues if they exist
+        kubectl describe pods --all-namespaces | grep -A 5 -B 2 "Failed\|Error\|Warning" || echo "No pod issues found"
+
+    - name: Cleanup on failure
+      if: failure() && steps.list-changed.outputs.changed == 'true'
+      run: |
+        echo "=== Cleanup on failure ==="
+        echo "=== Final cluster state ==="
+        kubectl get pods --all-namespaces
+        kubectl get events --all-namespaces --sort-by=.lastTimestamp | tail -10
+        
+        echo "=== Pod descriptions for debugging ==="
+        kubectl describe pods --all-namespaces | grep -A 10 -B 3 "Failed\|Error\|Warning\|Pending" || echo "No problematic pods found"
+        
+        echo "=== Recent logs for debugging ==="
+        kubectl logs --all-namespaces --tail=50 | grep -i "error\|timeout\|failed\|pull" || echo "No error logs found"
+
+        echo "=== Helm releases ==="
+        helm list --all-namespaces
       # the following would install only changed charts, but we only have one chart so 
       # don't worry about that for now
       # run: ct install --target-branch ${{ github.event.repository.default_branch }}

.github/workflows/pr-integration-tests.yml

@@ -11,142 +11,207 @@ on:
       - "release/**"
 
 env:
+  # Private Registry Configuration
+  PRIVATE_REGISTRY: experimental-registry.blacksmith.sh:5000
+  PRIVATE_REGISTRY_USERNAME: ${{ secrets.PRIVATE_REGISTRY_USERNAME }}
+  PRIVATE_REGISTRY_PASSWORD: ${{ secrets.PRIVATE_REGISTRY_PASSWORD }}
+
+  # Test Environment Variables
   OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
   SLACK_BOT_TOKEN: ${{ secrets.SLACK_BOT_TOKEN }}
-  CONFLUENCE_TEST_SPACE_URL: ${{ secrets.CONFLUENCE_TEST_SPACE_URL }}
-  CONFLUENCE_USER_NAME: ${{ secrets.CONFLUENCE_USER_NAME }}
+  CONFLUENCE_TEST_SPACE_URL: ${{ vars.CONFLUENCE_TEST_SPACE_URL }}
+  CONFLUENCE_USER_NAME: ${{ vars.CONFLUENCE_USER_NAME }}
   CONFLUENCE_ACCESS_TOKEN: ${{ secrets.CONFLUENCE_ACCESS_TOKEN }}
+  CONFLUENCE_ACCESS_TOKEN_SCOPED: ${{ secrets.CONFLUENCE_ACCESS_TOKEN_SCOPED }}
+  JIRA_BASE_URL: ${{ secrets.JIRA_BASE_URL }}
+  JIRA_USER_EMAIL: ${{ secrets.JIRA_USER_EMAIL }}
+  JIRA_API_TOKEN: ${{ secrets.JIRA_API_TOKEN }}
+  JIRA_API_TOKEN_SCOPED: ${{ secrets.JIRA_API_TOKEN_SCOPED }}
+  PERM_SYNC_SHAREPOINT_CLIENT_ID: ${{ secrets.PERM_SYNC_SHAREPOINT_CLIENT_ID }}
+  PERM_SYNC_SHAREPOINT_PRIVATE_KEY: ${{ secrets.PERM_SYNC_SHAREPOINT_PRIVATE_KEY }}
+  PERM_SYNC_SHAREPOINT_CERTIFICATE_PASSWORD: ${{ secrets.PERM_SYNC_SHAREPOINT_CERTIFICATE_PASSWORD }}
+  PERM_SYNC_SHAREPOINT_DIRECTORY_ID: ${{ secrets.PERM_SYNC_SHAREPOINT_DIRECTORY_ID }}
+  EXA_API_KEY: ${{ secrets.EXA_API_KEY }}
 
 jobs:
-  integration-tests:
-    # See https://runs-on.com/runners/linux/
-    runs-on: [runs-on, runner=32cpu-linux-x64, "run-id=${{ github.run_id }}"]
+  discover-test-dirs:
+    runs-on: blacksmith-2vcpu-ubuntu-2404-arm
+    outputs:
+      test-dirs: ${{ steps.set-matrix.outputs.test-dirs }}
     steps:
       - name: Checkout code
         uses: actions/checkout@v4
 
-      - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
+      - name: Discover test directories
+        id: set-matrix
+        run: |
+          # Find all leaf-level directories in both test directories
+          tests_dirs=$(find backend/tests/integration/tests -mindepth 1 -maxdepth 1 -type d ! -name "__pycache__" -exec basename {} \; | sort)
+          connector_dirs=$(find backend/tests/integration/connector_job_tests -mindepth 1 -maxdepth 1 -type d ! -name "__pycache__" -exec basename {} \; | sort)
 
+          # Create JSON array with directory info
+          all_dirs=""
+          for dir in $tests_dirs; do
+            all_dirs="$all_dirs{\"path\":\"tests/$dir\",\"name\":\"tests-$dir\"},"
+          done
+          for dir in $connector_dirs; do
+            all_dirs="$all_dirs{\"path\":\"connector_job_tests/$dir\",\"name\":\"connector-$dir\"},"
+          done
+
+          # Remove trailing comma and wrap in array
+          all_dirs="[${all_dirs%,}]"
+          echo "test-dirs=$all_dirs" >> $GITHUB_OUTPUT
+
+  prepare-build:
+    runs-on: blacksmith-2vcpu-ubuntu-2404-arm
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Prepare build
+        uses: ./.github/actions/prepare-build
+
+  build-backend-image:
+    runs-on: blacksmith-16vcpu-ubuntu-2404-arm
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Login to Private Registry
+        uses: docker/login-action@v3
+        with:
+          registry: ${{ env.PRIVATE_REGISTRY }}
+          username: ${{ env.PRIVATE_REGISTRY_USERNAME }}
+          password: ${{ env.PRIVATE_REGISTRY_PASSWORD }}
+
+      - name: Set up Docker Buildx
+        uses: useblacksmith/setup-docker-builder@v1
+
+      - name: Build and push Backend Docker image
+        uses: useblacksmith/build-push-action@v2
+        with:
+          context: ./backend
+          file: ./backend/Dockerfile
+          platforms: linux/arm64
+          tags: ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-backend:test-${{ github.run_id }}
+          push: true
+          outputs: type=registry
+          no-cache: true
+
+
+  build-model-server-image:
+    runs-on: blacksmith-16vcpu-ubuntu-2404-arm
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Login to Private Registry
+        uses: docker/login-action@v3
+        with:
+          registry: ${{ env.PRIVATE_REGISTRY }}
+          username: ${{ env.PRIVATE_REGISTRY_USERNAME }}
+          password: ${{ env.PRIVATE_REGISTRY_PASSWORD }}
+
+      - name: Set up Docker Buildx
+        uses: useblacksmith/setup-docker-builder@v1
+
+      - name: Build and push Model Server Docker image
+        uses: useblacksmith/build-push-action@v2
+        with:
+          context: ./backend
+          file: ./backend/Dockerfile.model_server
+          platforms: linux/arm64
+          tags: ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-model-server:test-${{ github.run_id }}
+          push: true
+          outputs: type=registry
+          provenance: false
+          no-cache: ${{ vars.DOCKER_NO_CACHE == 'true' }}
+
+
+  build-integration-image:
+    needs: prepare-build
+    runs-on: blacksmith-16vcpu-ubuntu-2404-arm
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Login to Private Registry
+        uses: docker/login-action@v3
+        with:
+          registry: ${{ env.PRIVATE_REGISTRY }}
+          username: ${{ env.PRIVATE_REGISTRY_USERNAME }}
+          password: ${{ env.PRIVATE_REGISTRY_PASSWORD }}
+
+      - name: Download OpenAPI artifacts
+        uses: actions/download-artifact@v4
+        with:
+          name: openapi-artifacts
+          path: backend/generated/
+
+      - name: Set up Docker Buildx
+        uses: useblacksmith/setup-docker-builder@v1
+
+      - name: Build and push integration test image with Docker Bake
+        env:
+          REGISTRY: ${{ env.PRIVATE_REGISTRY }}
+          TAG: test-${{ github.run_id }}
+        run: cd backend && docker buildx bake --no-cache --push integration
+
+  integration-tests:
+    needs:
+      [
+        discover-test-dirs,
+        build-backend-image,
+        build-model-server-image,
+        build-integration-image,
+      ]
+    runs-on: blacksmith-8vcpu-ubuntu-2404-arm
+
+    strategy:
+      fail-fast: false
+      matrix:
+        test-dir: ${{ fromJson(needs.discover-test-dirs.outputs.test-dirs) }}
+
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Login to Private Registry
+        uses: docker/login-action@v3
+        with:
+          registry: ${{ env.PRIVATE_REGISTRY }}
+          username: ${{ env.PRIVATE_REGISTRY_USERNAME }}
+          password: ${{ env.PRIVATE_REGISTRY_PASSWORD }}
+
+      # needed for pulling Vespa, Redis, Postgres, and Minio images
+      # otherwise, we hit the "Unauthenticated users" limit
+      # https://docs.docker.com/docker-hub/usage/
       - name: Login to Docker Hub
         uses: docker/login-action@v3
         with:
           username: ${{ secrets.DOCKER_USERNAME }}
           password: ${{ secrets.DOCKER_TOKEN }}
 
-      # tag every docker image with "test" so that we can spin up the correct set
-      # of images during testing
-
-      # We don't need to build the Web Docker image since it's not yet used
-      # in the integration tests. We have a separate action to verify that it builds
-      # successfully.
-      - name: Pull Web Docker image
+      - name: Pull Docker images
         run: |
-          docker pull onyxdotapp/onyx-web-server:latest
-          docker tag onyxdotapp/onyx-web-server:latest onyxdotapp/onyx-web-server:test
+          # Pull all images from registry in parallel
+          echo "Pulling Docker images in parallel..."
+          # Pull images from private registry
+          (docker pull --platform linux/arm64 ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-backend:test-${{ github.run_id }}) &
+          (docker pull --platform linux/arm64 ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-model-server:test-${{ github.run_id }}) &
+          (docker pull --platform linux/arm64 ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-integration:test-${{ github.run_id }}) &
 
-      # we use the runs-on cache for docker builds
-      # in conjunction with runs-on runners, it has better speed and unlimited caching
-      # https://runs-on.com/caching/s3-cache-for-github-actions/
-      # https://runs-on.com/caching/docker/
-      # https://github.com/moby/buildkit#s3-cache-experimental
+          # Wait for all background jobs to complete
+          wait
+          echo "All Docker images pulled successfully"
 
-      # images are built and run locally for testing purposes. Not pushed.
-      - name: Build Backend Docker image
-        uses: ./.github/actions/custom-build-and-push
-        with:
-          context: ./backend
-          file: ./backend/Dockerfile
-          platforms: linux/amd64
-          tags: onyxdotapp/onyx-backend:test
-          push: false
-          load: true
-          cache-from: type=s3,prefix=cache/${{ github.repository }}/integration-tests/backend/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }}
-          cache-to: type=s3,prefix=cache/${{ github.repository }}/integration-tests/backend/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }},mode=max
-
-      - name: Build Model Server Docker image
-        uses: ./.github/actions/custom-build-and-push
-        with:
-          context: ./backend
-          file: ./backend/Dockerfile.model_server
-          platforms: linux/amd64
-          tags: onyxdotapp/onyx-model-server:test
-          push: false
-          load: true
-          cache-from: type=s3,prefix=cache/${{ github.repository }}/integration-tests/model-server/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }}
-          cache-to: type=s3,prefix=cache/${{ github.repository }}/integration-tests/model-server/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }},mode=max
-
-      - name: Build integration test Docker image
-        uses: ./.github/actions/custom-build-and-push
-        with:
-          context: ./backend
-          file: ./backend/tests/integration/Dockerfile
-          platforms: linux/amd64
-          tags: onyxdotapp/onyx-integration:test
-          push: false
-          load: true
-          cache-from: type=s3,prefix=cache/${{ github.repository }}/integration-tests/integration/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }}
-          cache-to: type=s3,prefix=cache/${{ github.repository }}/integration-tests/integration/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }},mode=max
-
-      # Start containers for multi-tenant tests
-      - name: Start Docker containers for multi-tenant tests
-        run: |
-          cd deployment/docker_compose
-          ENABLE_PAID_ENTERPRISE_EDITION_FEATURES=true \
-          MULTI_TENANT=true \
-          AUTH_TYPE=cloud \
-          REQUIRE_EMAIL_VERIFICATION=false \
-          DISABLE_TELEMETRY=true \
-          IMAGE_TAG=test \
-          DEV_MODE=true \
-          docker compose -f docker-compose.multitenant-dev.yml -p onyx-stack up -d
-        id: start_docker_multi_tenant
-
-      # In practice, `cloud` Auth type would require OAUTH credentials to be set.
-      - name: Run Multi-Tenant Integration Tests
-        run: |
-          echo "Waiting for 3 minutes to ensure API server is ready..."
-          sleep 180
-          echo "Running integration tests..."
-          docker run --rm --network onyx-stack_default \
-            --name test-runner \
-            -e POSTGRES_HOST=relational_db \
-            -e POSTGRES_USER=postgres \
-            -e POSTGRES_PASSWORD=password \
-            -e POSTGRES_DB=postgres \
-            -e POSTGRES_USE_NULL_POOL=true \
-            -e VESPA_HOST=index \
-            -e REDIS_HOST=cache \
-            -e API_SERVER_HOST=api_server \
-            -e OPENAI_API_KEY=${OPENAI_API_KEY} \
-            -e SLACK_BOT_TOKEN=${SLACK_BOT_TOKEN} \
-            -e TEST_WEB_HOSTNAME=test-runner \
-            -e AUTH_TYPE=cloud \
-            -e MULTI_TENANT=true \
-            -e REQUIRE_EMAIL_VERIFICATION=false \
-            -e DISABLE_TELEMETRY=true \
-            -e IMAGE_TAG=test \
-            -e DEV_MODE=true \
-            onyxdotapp/onyx-integration:test \
-            /app/tests/integration/multitenant_tests
-        continue-on-error: true
-        id: run_multitenant_tests
-
-      - name: Check multi-tenant test results
-        run: |
-          if [ ${{ steps.run_multitenant_tests.outcome }} == 'failure' ]; then
-            echo "Multi-tenant integration tests failed. Exiting with error."
-            exit 1
-          else
-            echo "All multi-tenant integration tests passed successfully."
-          fi
-
-      - name: Stop multi-tenant Docker containers
-        run: |
-          cd deployment/docker_compose
-          docker compose -f docker-compose.multitenant-dev.yml -p onyx-stack down -v
+          # Re-tag to remove registry prefix for docker-compose
+          docker tag ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-backend:test-${{ github.run_id }} onyxdotapp/onyx-backend:test
+          docker tag ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-model-server:test-${{ github.run_id }} onyxdotapp/onyx-model-server:test
+          docker tag ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-integration:test-${{ github.run_id }} onyxdotapp/onyx-integration:test
 
       # NOTE: Use pre-ping/null pool to reduce flakiness due to dropped connections
+      # NOTE: don't need web server for integration tests
       - name: Start Docker containers
         run: |
           cd deployment/docker_compose
@@ -158,14 +223,24 @@ jobs:
           DISABLE_TELEMETRY=true \
           IMAGE_TAG=test \
           INTEGRATION_TESTS_MODE=true \
-          docker compose -f docker-compose.dev.yml -p onyx-stack up -d
+          CHECK_TTL_MANAGEMENT_TASK_FREQUENCY_IN_HOURS=0.001 \
+          docker compose -f docker-compose.yml -f docker-compose.dev.yml up \
+            relational_db \
+            index \
+            cache \
+            minio \
+            api_server \
+            inference_model_server \
+            indexing_model_server \
+            background \
+            -d
         id: start_docker
 
       - name: Wait for service to be ready
         run: |
           echo "Starting wait-for-service script..."
 
-          docker logs -f onyx-stack-api_server-1 &
+          docker logs -f onyx-api_server-1 &
 
           start_time=$(date +%s)
           timeout=300  # 5 minutes in seconds
@@ -201,43 +276,46 @@ jobs:
           docker compose -f docker-compose.mock-it-services.yml \
             -p mock-it-services-stack up -d
 
-      # NOTE: Use pre-ping/null to reduce flakiness due to dropped connections
-      - name: Run Standard Integration Tests
-        run: |
-          echo "Running integration tests..."
-          docker run --rm --network onyx-stack_default \
-            --name test-runner \
-            -e POSTGRES_HOST=relational_db \
-            -e POSTGRES_USER=postgres \
-            -e POSTGRES_PASSWORD=password \
-            -e POSTGRES_DB=postgres \
-            -e POSTGRES_POOL_PRE_PING=true \
-            -e POSTGRES_USE_NULL_POOL=true \
-            -e VESPA_HOST=index \
-            -e REDIS_HOST=cache \
-            -e API_SERVER_HOST=api_server \
-            -e OPENAI_API_KEY=${OPENAI_API_KEY} \
-            -e SLACK_BOT_TOKEN=${SLACK_BOT_TOKEN} \
-            -e CONFLUENCE_TEST_SPACE_URL=${CONFLUENCE_TEST_SPACE_URL} \
-            -e CONFLUENCE_USER_NAME=${CONFLUENCE_USER_NAME} \
-            -e CONFLUENCE_ACCESS_TOKEN=${CONFLUENCE_ACCESS_TOKEN} \
-            -e TEST_WEB_HOSTNAME=test-runner \
-            -e MOCK_CONNECTOR_SERVER_HOST=mock_connector_server \
-            -e MOCK_CONNECTOR_SERVER_PORT=8001 \
-            onyxdotapp/onyx-integration:test \
-            /app/tests/integration/tests \
-            /app/tests/integration/connector_job_tests
-        continue-on-error: true
-        id: run_tests
-
-      - name: Check test results
-        run: |
-          if [ ${{ steps.run_tests.outcome }} == 'failure' ]; then
-            echo "Integration tests failed. Exiting with error."
-            exit 1
-          else
-            echo "All integration tests passed successfully."
-          fi
+      - name: Run Integration Tests for ${{ matrix.test-dir.name }}
+        uses: nick-fields/retry@v3
+        with:
+          timeout_minutes: 20
+          max_attempts: 3
+          retry_wait_seconds: 10
+          command: |
+            echo "Running integration tests for ${{ matrix.test-dir.path }}..."
+            docker run --rm --network onyx_default \
+              --name test-runner \
+              -e POSTGRES_HOST=relational_db \
+              -e POSTGRES_USER=postgres \
+              -e POSTGRES_PASSWORD=password \
+              -e POSTGRES_DB=postgres \
+              -e DB_READONLY_USER=db_readonly_user \
+              -e DB_READONLY_PASSWORD=password \
+              -e POSTGRES_POOL_PRE_PING=true \
+              -e POSTGRES_USE_NULL_POOL=true \
+              -e VESPA_HOST=index \
+              -e REDIS_HOST=cache \
+              -e API_SERVER_HOST=api_server \
+              -e OPENAI_API_KEY=${OPENAI_API_KEY} \
+              -e SLACK_BOT_TOKEN=${SLACK_BOT_TOKEN} \
+              -e CONFLUENCE_TEST_SPACE_URL=${CONFLUENCE_TEST_SPACE_URL} \
+              -e CONFLUENCE_USER_NAME=${CONFLUENCE_USER_NAME} \
+              -e CONFLUENCE_ACCESS_TOKEN=${CONFLUENCE_ACCESS_TOKEN} \
+              -e CONFLUENCE_ACCESS_TOKEN_SCOPED=${CONFLUENCE_ACCESS_TOKEN_SCOPED} \
+              -e JIRA_BASE_URL=${JIRA_BASE_URL} \
+              -e JIRA_USER_EMAIL=${JIRA_USER_EMAIL} \
+              -e JIRA_API_TOKEN=${JIRA_API_TOKEN} \
+              -e JIRA_API_TOKEN_SCOPED=${JIRA_API_TOKEN_SCOPED} \
+              -e PERM_SYNC_SHAREPOINT_CLIENT_ID=${PERM_SYNC_SHAREPOINT_CLIENT_ID} \
+              -e PERM_SYNC_SHAREPOINT_PRIVATE_KEY="${PERM_SYNC_SHAREPOINT_PRIVATE_KEY}" \
+              -e PERM_SYNC_SHAREPOINT_CERTIFICATE_PASSWORD=${PERM_SYNC_SHAREPOINT_CERTIFICATE_PASSWORD} \
+              -e PERM_SYNC_SHAREPOINT_DIRECTORY_ID=${PERM_SYNC_SHAREPOINT_DIRECTORY_ID} \
+              -e TEST_WEB_HOSTNAME=test-runner \
+              -e MOCK_CONNECTOR_SERVER_HOST=mock_connector_server \
+              -e MOCK_CONNECTOR_SERVER_PORT=8001 \
+              onyxdotapp/onyx-integration:test \
+              /app/tests/integration/${{ matrix.test-dir.path }}
 
       # ------------------------------------------------------------
       # Always gather logs BEFORE "down":
@@ -245,19 +323,19 @@ jobs:
         if: always()
         run: |
           cd deployment/docker_compose
-          docker compose -f docker-compose.dev.yml -p onyx-stack logs --no-color api_server > $GITHUB_WORKSPACE/api_server.log || true
+          docker compose logs --no-color api_server > $GITHUB_WORKSPACE/api_server.log || true
 
       - name: Dump all-container logs (optional)
         if: always()
         run: |
           cd deployment/docker_compose
-          docker compose -f docker-compose.dev.yml -p onyx-stack logs --no-color > $GITHUB_WORKSPACE/docker-compose.log || true
+          docker compose logs --no-color > $GITHUB_WORKSPACE/docker-compose.log || true
 
       - name: Upload logs
         if: always()
         uses: actions/upload-artifact@v4
         with:
-          name: docker-all-logs
+          name: docker-all-logs-${{ matrix.test-dir.name }}
           path: ${{ github.workspace }}/docker-compose.log
       # ------------------------------------------------------------
 
@@ -265,4 +343,158 @@ jobs:
         if: always()
         run: |
           cd deployment/docker_compose
-          docker compose -f docker-compose.dev.yml -p onyx-stack down -v
+          docker compose down -v
+
+
+  multitenant-tests:
+    needs:
+      [
+        build-backend-image,
+        build-model-server-image,
+        build-integration-image,
+      ]
+    runs-on: blacksmith-8vcpu-ubuntu-2404-arm
+
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Login to Private Registry
+        uses: docker/login-action@v3
+        with:
+          registry: ${{ env.PRIVATE_REGISTRY }}
+          username: ${{ env.PRIVATE_REGISTRY_USERNAME }}
+          password: ${{ env.PRIVATE_REGISTRY_PASSWORD }}
+
+      - name: Login to Docker Hub
+        uses: docker/login-action@v3
+        with:
+          username: ${{ secrets.DOCKER_USERNAME }}
+          password: ${{ secrets.DOCKER_TOKEN }}
+
+      - name: Pull Docker images
+        run: |
+          (docker pull --platform linux/arm64 ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-backend:test-${{ github.run_id }}) &
+          (docker pull --platform linux/arm64 ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-model-server:test-${{ github.run_id }}) &
+          (docker pull --platform linux/arm64 ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-integration:test-${{ github.run_id }}) &
+          wait
+          docker tag ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-backend:test-${{ github.run_id }} onyxdotapp/onyx-backend:test
+          docker tag ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-model-server:test-${{ github.run_id }} onyxdotapp/onyx-model-server:test
+          docker tag ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-integration:test-${{ github.run_id }} onyxdotapp/onyx-integration:test
+
+      - name: Start Docker containers for multi-tenant tests
+        run: |
+          cd deployment/docker_compose
+          ENABLE_PAID_ENTERPRISE_EDITION_FEATURES=true \
+          MULTI_TENANT=true \
+          AUTH_TYPE=cloud \
+          REQUIRE_EMAIL_VERIFICATION=false \
+          DISABLE_TELEMETRY=true \
+          IMAGE_TAG=test \
+          DEV_MODE=true \
+          docker compose -f docker-compose.multitenant-dev.yml up \
+            relational_db \
+            index \
+            cache \
+            minio \
+            api_server \
+            inference_model_server \
+            indexing_model_server \
+            background \
+            -d
+        id: start_docker_multi_tenant
+
+      - name: Wait for service to be ready (multi-tenant)
+        run: |
+          echo "Starting wait-for-service script for multi-tenant..."
+          docker logs -f onyx-api_server-1 &
+          start_time=$(date +%s)
+          timeout=300
+          while true; do
+            current_time=$(date +%s)
+            elapsed_time=$((current_time - start_time))
+            if [ $elapsed_time -ge $timeout ]; then
+              echo "Timeout reached. Service did not become ready in 5 minutes."
+              exit 1
+            fi
+            response=$(curl -s -o /dev/null -w "%{http_code}" http://localhost:8080/health || echo "curl_error")
+            if [ "$response" = "200" ]; then
+              echo "Service is ready!"
+              break
+            elif [ "$response" = "curl_error" ]; then
+              echo "Curl encountered an error; retrying..."
+            else
+              echo "Service not ready yet (HTTP $response). Retrying in 5 seconds..."
+            fi
+            sleep 5
+          done
+          echo "Finished waiting for service."
+
+      - name: Run Multi-Tenant Integration Tests
+        run: |
+          echo "Running multi-tenant integration tests..."
+          docker run --rm --network onyx_default \
+            --name test-runner \
+            -e POSTGRES_HOST=relational_db \
+            -e POSTGRES_USER=postgres \
+            -e POSTGRES_PASSWORD=password \
+            -e DB_READONLY_USER=db_readonly_user \
+            -e DB_READONLY_PASSWORD=password \
+            -e POSTGRES_DB=postgres \
+            -e POSTGRES_USE_NULL_POOL=true \
+            -e VESPA_HOST=index \
+            -e REDIS_HOST=cache \
+            -e API_SERVER_HOST=api_server \
+            -e OPENAI_API_KEY=${OPENAI_API_KEY} \
+            -e SLACK_BOT_TOKEN=${SLACK_BOT_TOKEN} \
+            -e TEST_WEB_HOSTNAME=test-runner \
+            -e AUTH_TYPE=cloud \
+            -e MULTI_TENANT=true \
+            -e SKIP_RESET=true \
+            -e REQUIRE_EMAIL_VERIFICATION=false \
+            -e DISABLE_TELEMETRY=true \
+            -e IMAGE_TAG=test \
+            -e DEV_MODE=true \
+            onyxdotapp/onyx-integration:test \
+            /app/tests/integration/multitenant_tests
+
+      - name: Dump API server logs (multi-tenant)
+        if: always()
+        run: |
+          cd deployment/docker_compose
+          docker compose -f docker-compose.multitenant-dev.yml logs --no-color api_server > $GITHUB_WORKSPACE/api_server_multitenant.log || true
+
+      - name: Dump all-container logs (multi-tenant)
+        if: always()
+        run: |
+          cd deployment/docker_compose
+          docker compose -f docker-compose.multitenant-dev.yml logs --no-color > $GITHUB_WORKSPACE/docker-compose-multitenant.log || true
+
+      - name: Upload logs (multi-tenant)
+        if: always()
+        uses: actions/upload-artifact@v4
+        with:
+          name: docker-all-logs-multitenant
+          path: ${{ github.workspace }}/docker-compose-multitenant.log
+
+      - name: Stop multi-tenant Docker containers
+        if: always()
+        run: |
+          cd deployment/docker_compose
+          docker compose -f docker-compose.multitenant-dev.yml down -v
+
+  required: 
+    runs-on: blacksmith-2vcpu-ubuntu-2404-arm
+    needs: [integration-tests, multitenant-tests]
+    if: ${{ always() }}
+    steps:
+      - uses: actions/github-script@v7
+        with:
+          script: |
+            const needs = ${{ toJSON(needs) }};
+            const failed = Object.values(needs).some(n => n.result !== 'success');
+            if (failed) {
+              core.setFailed('One or more upstream jobs failed or were cancelled.');
+            } else {
+              core.notice('All required jobs succeeded.');
+            }

.github/workflows/pr-jest-tests.yml

@@ -0,0 +1,35 @@
+name: Run Jest Tests
+concurrency:
+  group: Run-Jest-Tests-${{ github.workflow }}-${{ github.head_ref || github.event.workflow_run.head_branch || github.run_id }}
+  cancel-in-progress: true
+
+on: push
+
+jobs:
+  jest-tests:
+    name: Jest Tests
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Setup node
+        uses: actions/setup-node@v4
+        with:
+          node-version: 22
+
+      - name: Install node dependencies
+        working-directory: ./web
+        run: npm ci
+
+      - name: Run Jest tests
+        working-directory: ./web
+        run: npm test -- --ci --coverage --maxWorkers=50%
+
+      - name: Upload coverage reports
+        if: always()
+        uses: actions/upload-artifact@v4
+        with:
+          name: jest-coverage-${{ github.run_id }}
+          path: ./web/coverage
+          retention-days: 7

.github/workflows/pr-labeler.yml

@@ -0,0 +1,38 @@
+name: PR Labeler
+
+on:
+  pull_request_target:
+    branches:
+      - main
+    types:
+      - opened
+      - reopened
+      - synchronize
+      - edited
+
+permissions:
+  contents: read
+  pull-requests: write
+
+jobs:
+  validate_pr_title:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Check PR title for Conventional Commits
+        env:
+          PR_TITLE: ${{ github.event.pull_request.title }}
+        run: |
+          echo "PR Title: $PR_TITLE"
+          if [[ ! "$PR_TITLE" =~ ^(feat|fix|docs|test|ci|refactor|perf|chore|revert|build)(\(.+\))?:\ .+ ]]; then
+            echo "::error::❌ Your PR title does not follow the Conventional Commits format.
+              This check ensures that all pull requests use clear, consistent titles that help automate changelogs and improve project history.
+
+              Please update your PR title to follow the Conventional Commits style.  
+              Here is a link to a blog explaining the reason why we've included the Conventional Commits style into our PR titles: https://xfuture-blog.com/working-with-conventional-commits
+
+              **Here are some examples of valid PR titles:**
+              - feat: add user authentication
+              - fix(login): handle null password error
+              - docs(readme): update installation instructions"
+            exit 1
+          fi

.github/workflows/pr-mit-integration-tests.yml

@@ -5,90 +5,210 @@ concurrency:
 
 on:
   merge_group:
-  pull_request:
-    branches:
-      - main
-      - "release/**"
+    types: [checks_requested]
 
 env:
+  # Private Registry Configuration
+  PRIVATE_REGISTRY: experimental-registry.blacksmith.sh:5000
+  PRIVATE_REGISTRY_USERNAME: ${{ secrets.PRIVATE_REGISTRY_USERNAME }}
+  PRIVATE_REGISTRY_PASSWORD: ${{ secrets.PRIVATE_REGISTRY_PASSWORD }}
+
+  # Test Environment Variables
   OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
   SLACK_BOT_TOKEN: ${{ secrets.SLACK_BOT_TOKEN }}
-  CONFLUENCE_TEST_SPACE_URL: ${{ secrets.CONFLUENCE_TEST_SPACE_URL }}
-  CONFLUENCE_USER_NAME: ${{ secrets.CONFLUENCE_USER_NAME }}
+  CONFLUENCE_TEST_SPACE_URL: ${{ vars.CONFLUENCE_TEST_SPACE_URL }}
+  CONFLUENCE_USER_NAME: ${{ vars.CONFLUENCE_USER_NAME }}
   CONFLUENCE_ACCESS_TOKEN: ${{ secrets.CONFLUENCE_ACCESS_TOKEN }}
+  CONFLUENCE_ACCESS_TOKEN_SCOPED: ${{ secrets.CONFLUENCE_ACCESS_TOKEN_SCOPED }}
+  JIRA_BASE_URL: ${{ secrets.JIRA_BASE_URL }}
+  JIRA_USER_EMAIL: ${{ secrets.JIRA_USER_EMAIL }}
+  JIRA_API_TOKEN: ${{ secrets.JIRA_API_TOKEN }}
+  JIRA_API_TOKEN_SCOPED: ${{ secrets.JIRA_API_TOKEN_SCOPED }}
+  PERM_SYNC_SHAREPOINT_CLIENT_ID: ${{ secrets.PERM_SYNC_SHAREPOINT_CLIENT_ID }}
+  PERM_SYNC_SHAREPOINT_PRIVATE_KEY: ${{ secrets.PERM_SYNC_SHAREPOINT_PRIVATE_KEY }}
+  PERM_SYNC_SHAREPOINT_CERTIFICATE_PASSWORD: ${{ secrets.PERM_SYNC_SHAREPOINT_CERTIFICATE_PASSWORD }}
+  PERM_SYNC_SHAREPOINT_DIRECTORY_ID: ${{ secrets.PERM_SYNC_SHAREPOINT_DIRECTORY_ID }}
 
 jobs:
-  integration-tests-mit:
-    # See https://runs-on.com/runners/linux/
-    runs-on: [runs-on, runner=32cpu-linux-x64, "run-id=${{ github.run_id }}"]
+  discover-test-dirs:
+    runs-on: blacksmith-2vcpu-ubuntu-2404-arm
+    outputs:
+      test-dirs: ${{ steps.set-matrix.outputs.test-dirs }}
     steps:
       - name: Checkout code
         uses: actions/checkout@v4
 
-      - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
+      - name: Discover test directories
+        id: set-matrix
+        run: |
+          # Find all leaf-level directories in both test directories
+          tests_dirs=$(find backend/tests/integration/tests -mindepth 1 -maxdepth 1 -type d ! -name "__pycache__" -exec basename {} \; | sort)
+          connector_dirs=$(find backend/tests/integration/connector_job_tests -mindepth 1 -maxdepth 1 -type d ! -name "__pycache__" -exec basename {} \; | sort)
 
+          # Create JSON array with directory info
+          all_dirs=""
+          for dir in $tests_dirs; do
+            all_dirs="$all_dirs{\"path\":\"tests/$dir\",\"name\":\"tests-$dir\"},"
+          done
+          for dir in $connector_dirs; do
+            all_dirs="$all_dirs{\"path\":\"connector_job_tests/$dir\",\"name\":\"connector-$dir\"},"
+          done
+
+          # Remove trailing comma and wrap in array
+          all_dirs="[${all_dirs%,}]"
+          echo "test-dirs=$all_dirs" >> $GITHUB_OUTPUT
+
+  prepare-build:
+    runs-on: blacksmith-2vcpu-ubuntu-2404-arm
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Prepare build
+        uses: ./.github/actions/prepare-build
+
+  build-backend-image:
+    runs-on: blacksmith-16vcpu-ubuntu-2404-arm
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Login to Private Registry
+        uses: docker/login-action@v3
+        with:
+          registry: ${{ env.PRIVATE_REGISTRY }}
+          username: ${{ env.PRIVATE_REGISTRY_USERNAME }}
+          password: ${{ env.PRIVATE_REGISTRY_PASSWORD }}
+
+      - name: Set up Docker Buildx
+        uses: useblacksmith/setup-docker-builder@v1
+
+      - name: Build and push Backend Docker image
+        uses: useblacksmith/build-push-action@v2
+        with:
+          context: ./backend
+          file: ./backend/Dockerfile
+          platforms: linux/arm64
+          tags: ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-backend:test-${{ github.run_id }}
+          push: true
+          outputs: type=registry
+          no-cache: true
+
+
+  build-model-server-image:
+    runs-on: blacksmith-16vcpu-ubuntu-2404-arm
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Login to Private Registry
+        uses: docker/login-action@v3
+        with:
+          registry: ${{ env.PRIVATE_REGISTRY }}
+          username: ${{ env.PRIVATE_REGISTRY_USERNAME }}
+          password: ${{ env.PRIVATE_REGISTRY_PASSWORD }}
+
+      - name: Set up Docker Buildx
+        uses: useblacksmith/setup-docker-builder@v1
+
+      - name: Build and push Model Server Docker image
+        uses: useblacksmith/build-push-action@v2
+        with:
+          context: ./backend
+          file: ./backend/Dockerfile.model_server
+          platforms: linux/arm64
+          tags: ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-model-server:test-${{ github.run_id }}
+          push: true
+          outputs: type=registry
+          provenance: false
+          no-cache: ${{ vars.DOCKER_NO_CACHE == 'true' }}
+
+
+  build-integration-image:
+    needs: prepare-build
+    runs-on: blacksmith-16vcpu-ubuntu-2404-arm
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Login to Private Registry
+        uses: docker/login-action@v3
+        with:
+          registry: ${{ env.PRIVATE_REGISTRY }}
+          username: ${{ env.PRIVATE_REGISTRY_USERNAME }}
+          password: ${{ env.PRIVATE_REGISTRY_PASSWORD }}
+
+      - name: Download OpenAPI artifacts
+        uses: actions/download-artifact@v4
+        with:
+          name: openapi-artifacts
+          path: backend/generated/
+
+      - name: Set up Docker Buildx
+        uses: useblacksmith/setup-docker-builder@v1
+
+      - name: Build and push integration test image with Docker Bake
+        env:
+          REGISTRY: ${{ env.PRIVATE_REGISTRY }}
+          TAG: test-${{ github.run_id }}
+        run: cd backend && docker buildx bake --no-cache --push integration
+
+  integration-tests-mit:
+    needs:
+      [
+        discover-test-dirs,
+        build-backend-image,
+        build-model-server-image,
+        build-integration-image,
+      ]
+    # See https://docs.blacksmith.sh/blacksmith-runners/overview
+    runs-on: blacksmith-8vcpu-ubuntu-2404-arm
+
+    strategy:
+      fail-fast: false
+      matrix:
+        test-dir: ${{ fromJson(needs.discover-test-dirs.outputs.test-dirs) }}
+
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Login to Private Registry
+        uses: docker/login-action@v3
+        with:
+          registry: ${{ env.PRIVATE_REGISTRY }}
+          username: ${{ env.PRIVATE_REGISTRY_USERNAME }}
+          password: ${{ env.PRIVATE_REGISTRY_PASSWORD }}
+
+      # needed for pulling Vespa, Redis, Postgres, and Minio images
+      # otherwise, we hit the "Unauthenticated users" limit
+      # https://docs.docker.com/docker-hub/usage/
       - name: Login to Docker Hub
         uses: docker/login-action@v3
         with:
           username: ${{ secrets.DOCKER_USERNAME }}
           password: ${{ secrets.DOCKER_TOKEN }}
 
-      # tag every docker image with "test" so that we can spin up the correct set
-      # of images during testing
-
-      # We don't need to build the Web Docker image since it's not yet used
-      # in the integration tests. We have a separate action to verify that it builds
-      # successfully.
-      - name: Pull Web Docker image
+      - name: Pull Docker images
         run: |
-          docker pull onyxdotapp/onyx-web-server:latest
-          docker tag onyxdotapp/onyx-web-server:latest onyxdotapp/onyx-web-server:test
+          # Pull all images from registry in parallel
+          echo "Pulling Docker images in parallel..."
+          # Pull images from private registry
+          (docker pull --platform linux/arm64 ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-backend:test-${{ github.run_id }}) &
+          (docker pull --platform linux/arm64 ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-model-server:test-${{ github.run_id }}) &
+          (docker pull --platform linux/arm64 ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-integration:test-${{ github.run_id }}) &
 
-      # we use the runs-on cache for docker builds
-      # in conjunction with runs-on runners, it has better speed and unlimited caching
-      # https://runs-on.com/caching/s3-cache-for-github-actions/
-      # https://runs-on.com/caching/docker/
-      # https://github.com/moby/buildkit#s3-cache-experimental
+          # Wait for all background jobs to complete
+          wait
+          echo "All Docker images pulled successfully"
 
-      # images are built and run locally for testing purposes. Not pushed.
-      - name: Build Backend Docker image
-        uses: ./.github/actions/custom-build-and-push
-        with:
-          context: ./backend
-          file: ./backend/Dockerfile
-          platforms: linux/amd64
-          tags: onyxdotapp/onyx-backend:test
-          push: false
-          load: true
-          cache-from: type=s3,prefix=cache/${{ github.repository }}/integration-tests/backend/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }}
-          cache-to: type=s3,prefix=cache/${{ github.repository }}/integration-tests/backend/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }},mode=max
-
-      - name: Build Model Server Docker image
-        uses: ./.github/actions/custom-build-and-push
-        with:
-          context: ./backend
-          file: ./backend/Dockerfile.model_server
-          platforms: linux/amd64
-          tags: onyxdotapp/onyx-model-server:test
-          push: false
-          load: true
-          cache-from: type=s3,prefix=cache/${{ github.repository }}/integration-tests/model-server/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }}
-          cache-to: type=s3,prefix=cache/${{ github.repository }}/integration-tests/model-server/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }},mode=max
-
-      - name: Build integration test Docker image
-        uses: ./.github/actions/custom-build-and-push
-        with:
-          context: ./backend
-          file: ./backend/tests/integration/Dockerfile
-          platforms: linux/amd64
-          tags: onyxdotapp/onyx-integration:test
-          push: false
-          load: true
-          cache-from: type=s3,prefix=cache/${{ github.repository }}/integration-tests/integration/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }}
-          cache-to: type=s3,prefix=cache/${{ github.repository }}/integration-tests/integration/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }},mode=max
+          # Re-tag to remove registry prefix for docker-compose
+          docker tag ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-backend:test-${{ github.run_id }} onyxdotapp/onyx-backend:test
+          docker tag ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-model-server:test-${{ github.run_id }} onyxdotapp/onyx-model-server:test
+          docker tag ${{ env.PRIVATE_REGISTRY }}/integration-test-onyx-integration:test-${{ github.run_id }} onyxdotapp/onyx-integration:test
 
       # NOTE: Use pre-ping/null pool to reduce flakiness due to dropped connections
+      # NOTE: don't need web server for integration tests
       - name: Start Docker containers
         run: |
           cd deployment/docker_compose
@@ -99,14 +219,23 @@ jobs:
           DISABLE_TELEMETRY=true \
           IMAGE_TAG=test \
           INTEGRATION_TESTS_MODE=true \
-          docker compose -f docker-compose.dev.yml -p onyx-stack up -d
+          docker compose -f docker-compose.yml -f docker-compose.dev.yml up \
+            relational_db \
+            index \
+            cache \
+            minio \
+            api_server \
+            inference_model_server \
+            indexing_model_server \
+            background \
+            -d
         id: start_docker
 
       - name: Wait for service to be ready
         run: |
           echo "Starting wait-for-service script..."
 
-          docker logs -f onyx-stack-api_server-1 &
+          docker logs -f onyx-api_server-1 &
 
           start_time=$(date +%s)
           timeout=300  # 5 minutes in seconds
@@ -143,42 +272,46 @@ jobs:
             -p mock-it-services-stack up -d
 
       # NOTE: Use pre-ping/null to reduce flakiness due to dropped connections
-      - name: Run Standard Integration Tests
-        run: |
-          echo "Running integration tests..."
-          docker run --rm --network onyx-stack_default \
-            --name test-runner \
-            -e POSTGRES_HOST=relational_db \
-            -e POSTGRES_USER=postgres \
-            -e POSTGRES_PASSWORD=password \
-            -e POSTGRES_DB=postgres \
-            -e POSTGRES_POOL_PRE_PING=true \
-            -e POSTGRES_USE_NULL_POOL=true \
-            -e VESPA_HOST=index \
-            -e REDIS_HOST=cache \
-            -e API_SERVER_HOST=api_server \
-            -e OPENAI_API_KEY=${OPENAI_API_KEY} \
-            -e SLACK_BOT_TOKEN=${SLACK_BOT_TOKEN} \
-            -e CONFLUENCE_TEST_SPACE_URL=${CONFLUENCE_TEST_SPACE_URL} \
-            -e CONFLUENCE_USER_NAME=${CONFLUENCE_USER_NAME} \
-            -e CONFLUENCE_ACCESS_TOKEN=${CONFLUENCE_ACCESS_TOKEN} \
-            -e TEST_WEB_HOSTNAME=test-runner \
-            -e MOCK_CONNECTOR_SERVER_HOST=mock_connector_server \
-            -e MOCK_CONNECTOR_SERVER_PORT=8001 \
-            onyxdotapp/onyx-integration:test \
-            /app/tests/integration/tests \
-            /app/tests/integration/connector_job_tests
-        continue-on-error: true
-        id: run_tests
-
-      - name: Check test results
-        run: |
-          if [ ${{ steps.run_tests.outcome }} == 'failure' ]; then
-            echo "Integration tests failed. Exiting with error."
-            exit 1
-          else
-            echo "All integration tests passed successfully."
-          fi
+      - name: Run Integration Tests for ${{ matrix.test-dir.name }}
+        uses: nick-fields/retry@v3
+        with:
+          timeout_minutes: 20
+          max_attempts: 3
+          retry_wait_seconds: 10
+          command: |
+            echo "Running integration tests for ${{ matrix.test-dir.path }}..."
+            docker run --rm --network onyx_default \
+              --name test-runner \
+              -e POSTGRES_HOST=relational_db \
+              -e POSTGRES_USER=postgres \
+              -e POSTGRES_PASSWORD=password \
+              -e POSTGRES_DB=postgres \
+              -e DB_READONLY_USER=db_readonly_user \
+              -e DB_READONLY_PASSWORD=password \
+              -e POSTGRES_POOL_PRE_PING=true \
+              -e POSTGRES_USE_NULL_POOL=true \
+              -e VESPA_HOST=index \
+              -e REDIS_HOST=cache \
+              -e API_SERVER_HOST=api_server \
+              -e OPENAI_API_KEY=${OPENAI_API_KEY} \
+              -e SLACK_BOT_TOKEN=${SLACK_BOT_TOKEN} \
+              -e CONFLUENCE_TEST_SPACE_URL=${CONFLUENCE_TEST_SPACE_URL} \
+              -e CONFLUENCE_USER_NAME=${CONFLUENCE_USER_NAME} \
+              -e CONFLUENCE_ACCESS_TOKEN=${CONFLUENCE_ACCESS_TOKEN} \
+              -e CONFLUENCE_ACCESS_TOKEN_SCOPED=${CONFLUENCE_ACCESS_TOKEN_SCOPED} \
+              -e JIRA_BASE_URL=${JIRA_BASE_URL} \
+              -e JIRA_USER_EMAIL=${JIRA_USER_EMAIL} \
+              -e JIRA_API_TOKEN=${JIRA_API_TOKEN} \
+              -e JIRA_API_TOKEN_SCOPED=${JIRA_API_TOKEN_SCOPED} \
+              -e PERM_SYNC_SHAREPOINT_CLIENT_ID=${PERM_SYNC_SHAREPOINT_CLIENT_ID} \
+              -e PERM_SYNC_SHAREPOINT_PRIVATE_KEY="${PERM_SYNC_SHAREPOINT_PRIVATE_KEY}" \
+              -e PERM_SYNC_SHAREPOINT_CERTIFICATE_PASSWORD=${PERM_SYNC_SHAREPOINT_CERTIFICATE_PASSWORD} \
+              -e PERM_SYNC_SHAREPOINT_DIRECTORY_ID=${PERM_SYNC_SHAREPOINT_DIRECTORY_ID} \
+              -e TEST_WEB_HOSTNAME=test-runner \
+              -e MOCK_CONNECTOR_SERVER_HOST=mock_connector_server \
+              -e MOCK_CONNECTOR_SERVER_PORT=8001 \
+              onyxdotapp/onyx-integration:test \
+              /app/tests/integration/${{ matrix.test-dir.path }}
 
       # ------------------------------------------------------------
       # Always gather logs BEFORE "down":
@@ -186,19 +319,19 @@ jobs:
         if: always()
         run: |
           cd deployment/docker_compose
-          docker compose -f docker-compose.dev.yml -p onyx-stack logs --no-color api_server > $GITHUB_WORKSPACE/api_server.log || true
+          docker compose logs --no-color api_server > $GITHUB_WORKSPACE/api_server.log || true
 
       - name: Dump all-container logs (optional)
         if: always()
         run: |
           cd deployment/docker_compose
-          docker compose -f docker-compose.dev.yml -p onyx-stack logs --no-color > $GITHUB_WORKSPACE/docker-compose.log || true
+          docker compose logs --no-color > $GITHUB_WORKSPACE/docker-compose.log || true
 
       - name: Upload logs
         if: always()
         uses: actions/upload-artifact@v4
         with:
-          name: docker-all-logs
+          name: docker-all-logs-${{ matrix.test-dir.name }}
           path: ${{ github.workspace }}/docker-compose.log
       # ------------------------------------------------------------
 
@@ -206,4 +339,21 @@ jobs:
         if: always()
         run: |
           cd deployment/docker_compose
-          docker compose -f docker-compose.dev.yml -p onyx-stack down -v
+          docker compose down -v
+
+  
+  required: 
+    runs-on: blacksmith-2vcpu-ubuntu-2404-arm
+    needs: [integration-tests-mit]
+    if: ${{ always() }}
+    steps:
+      - uses: actions/github-script@v7
+        with:
+          script: |
+            const needs = ${{ toJSON(needs) }};
+            const failed = Object.values(needs).some(n => n.result !== 'success');
+            if (failed) {
+              core.setFailed('One or more upstream jobs failed or were cancelled.');
+            } else {
+              core.notice('All required jobs succeeded.');
+            }

.github/workflows/pr-playwright-tests.yml

@@ -6,43 +6,171 @@ concurrency:
 on: push
 
 env:
+  # AWS ECR Configuration
+  AWS_REGION: ${{ secrets.AWS_REGION || 'us-west-2' }}
+  ECR_REGISTRY: ${{ secrets.ECR_REGISTRY }}
+  AWS_ACCESS_KEY_ID: ${{ secrets.AWS_ACCESS_KEY_ID_ECR }}
+  AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_ACCESS_KEY_ECR }}
+  BUILDX_NO_DEFAULT_ATTESTATIONS: 1
+
+  # Test Environment Variables
   OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
   SLACK_BOT_TOKEN: ${{ secrets.SLACK_BOT_TOKEN }}
   GEN_AI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
+  EXA_API_KEY: ${{ secrets.EXA_API_KEY }}
+
+  # for federated slack tests
+  SLACK_CLIENT_ID: ${{ secrets.SLACK_CLIENT_ID }}
+  SLACK_CLIENT_SECRET: ${{ secrets.SLACK_CLIENT_SECRET }}
+
   MOCK_LLM_RESPONSE: true
 
 jobs:
-  playwright-tests:
-    name: Playwright Tests
+  build-web-image:
+    runs-on: blacksmith-8vcpu-ubuntu-2404-arm
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
 
-    # See https://runs-on.com/runners/linux/
-    runs-on:
-      [
-        runs-on,
-        runner=32cpu-linux-x64,
-        disk=large,
-        "run-id=${{ github.run_id }}",
-      ]
+      - name: Configure AWS credentials
+        uses: aws-actions/configure-aws-credentials@v4
+        with:
+          aws-access-key-id: ${{ env.AWS_ACCESS_KEY_ID }}
+          aws-secret-access-key: ${{ env.AWS_SECRET_ACCESS_KEY }}
+          aws-region: ${{ env.AWS_REGION }}
+
+      - name: Login to Amazon ECR
+        id: login-ecr
+        uses: aws-actions/amazon-ecr-login@v2
+
+      - name: Set up Docker Buildx
+        uses: useblacksmith/setup-docker-builder@v1
+
+      - name: Build and push Web Docker image
+        uses: useblacksmith/build-push-action@v2
+        with:
+          context: ./web
+          file: ./web/Dockerfile
+          platforms: linux/arm64
+          tags: ${{ env.ECR_REGISTRY }}/integration-test-onyx-web-server:playwright-test-${{ github.run_id }}
+          provenance: false
+          sbom: false
+          push: true
+          outputs: type=registry
+          no-cache: ${{ vars.DOCKER_NO_CACHE == 'true' }}
+
+  build-backend-image:
+    runs-on: blacksmith-8vcpu-ubuntu-2404-arm
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Configure AWS credentials
+        uses: aws-actions/configure-aws-credentials@v4
+        with:
+          aws-access-key-id: ${{ env.AWS_ACCESS_KEY_ID }}
+          aws-secret-access-key: ${{ env.AWS_SECRET_ACCESS_KEY }}
+          aws-region: ${{ env.AWS_REGION }}
+
+      - name: Login to Amazon ECR
+        id: login-ecr
+        uses: aws-actions/amazon-ecr-login@v2
+
+      - name: Set up Docker Buildx
+        uses: useblacksmith/setup-docker-builder@v1
+
+      - name: Build and push Backend Docker image
+        uses: useblacksmith/build-push-action@v2
+        with:
+          context: ./backend
+          file: ./backend/Dockerfile
+          platforms: linux/arm64
+          tags: ${{ env.ECR_REGISTRY }}/integration-test-onyx-backend:playwright-test-${{ github.run_id }}
+          provenance: false
+          sbom: false
+          push: true
+          outputs: type=registry
+          no-cache: ${{ vars.DOCKER_NO_CACHE == 'true' }}
+
+  build-model-server-image:
+    runs-on: blacksmith-8vcpu-ubuntu-2404-arm
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Configure AWS credentials
+        uses: aws-actions/configure-aws-credentials@v4
+        with:
+          aws-access-key-id: ${{ env.AWS_ACCESS_KEY_ID }}
+          aws-secret-access-key: ${{ env.AWS_SECRET_ACCESS_KEY }}
+          aws-region: ${{ env.AWS_REGION }}
+
+      - name: Login to Amazon ECR
+        id: login-ecr
+        uses: aws-actions/amazon-ecr-login@v2
+
+      - name: Set up Docker Buildx
+        uses: useblacksmith/setup-docker-builder@v1
+
+      - name: Build and push Model Server Docker image
+        uses: useblacksmith/build-push-action@v2
+        with:
+          context: ./backend
+          file: ./backend/Dockerfile.model_server
+          platforms: linux/arm64
+          tags: ${{ env.ECR_REGISTRY }}/integration-test-onyx-model-server:playwright-test-${{ github.run_id }}
+          provenance: false
+          sbom: false
+          push: true
+          outputs: type=registry
+          no-cache: ${{ vars.DOCKER_NO_CACHE == 'true' }}
+
+  playwright-tests:
+    needs: [build-web-image, build-backend-image, build-model-server-image]
+    name: Playwright Tests
+    runs-on: blacksmith-8vcpu-ubuntu-2404-arm
     steps:
       - name: Checkout code
         uses: actions/checkout@v4
         with:
           fetch-depth: 0
 
-      - name: Set up Python
-        uses: actions/setup-python@v5
+      - name: Configure AWS credentials
+        uses: aws-actions/configure-aws-credentials@v4
         with:
-          python-version: "3.11"
-          cache: "pip"
-          cache-dependency-path: |
-            backend/requirements/default.txt
-            backend/requirements/dev.txt
-            backend/requirements/model_server.txt
-      - run: |
-          python -m pip install --upgrade pip
-          pip install --retries 5 --timeout 30 -r backend/requirements/default.txt
-          pip install --retries 5 --timeout 30 -r backend/requirements/dev.txt
-          pip install --retries 5 --timeout 30 -r backend/requirements/model_server.txt
+          aws-access-key-id: ${{ env.AWS_ACCESS_KEY_ID }}
+          aws-secret-access-key: ${{ env.AWS_SECRET_ACCESS_KEY }}
+          aws-region: ${{ env.AWS_REGION }}
+
+      - name: Login to Amazon ECR
+        id: login-ecr
+        uses: aws-actions/amazon-ecr-login@v2
+
+      # needed for pulling Vespa, Redis, Postgres, and Minio images
+      # otherwise, we hit the "Unauthenticated users" limit
+      # https://docs.docker.com/docker-hub/usage/
+      - name: Login to Docker Hub
+        uses: docker/login-action@v3
+        with:
+          username: ${{ secrets.DOCKER_USERNAME }}
+          password: ${{ secrets.DOCKER_TOKEN }}
+
+      - name: Pull Docker images
+        run: |
+          # Pull all images from ECR in parallel
+          echo "Pulling Docker images in parallel..."
+          (docker pull ${{ env.ECR_REGISTRY }}/integration-test-onyx-web-server:playwright-test-${{ github.run_id }}) &
+          (docker pull ${{ env.ECR_REGISTRY }}/integration-test-onyx-backend:playwright-test-${{ github.run_id }}) &
+          (docker pull ${{ env.ECR_REGISTRY }}/integration-test-onyx-model-server:playwright-test-${{ github.run_id }}) &
+
+          # Wait for all background jobs to complete
+          wait
+          echo "All Docker images pulled successfully"
+
+          # Re-tag with expected names for docker-compose
+          docker tag ${{ env.ECR_REGISTRY }}/integration-test-onyx-web-server:playwright-test-${{ github.run_id }} onyxdotapp/onyx-web-server:test
+          docker tag ${{ env.ECR_REGISTRY }}/integration-test-onyx-backend:playwright-test-${{ github.run_id }} onyxdotapp/onyx-backend:test
+          docker tag ${{ env.ECR_REGISTRY }}/integration-test-onyx-model-server:playwright-test-${{ github.run_id }} onyxdotapp/onyx-model-server:test
 
       - name: Setup node
         uses: actions/setup-node@v4
@@ -57,79 +185,29 @@ jobs:
         working-directory: ./web
         run: npx playwright install --with-deps
 
-      - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
-
-      - name: Login to Docker Hub
-        uses: docker/login-action@v3
-        with:
-          username: ${{ secrets.DOCKER_USERNAME }}
-          password: ${{ secrets.DOCKER_TOKEN }}
-
-      # tag every docker image with "test" so that we can spin up the correct set
-      # of images during testing
-
-      # we use the runs-on cache for docker builds
-      # in conjunction with runs-on runners, it has better speed and unlimited caching
-      # https://runs-on.com/caching/s3-cache-for-github-actions/
-      # https://runs-on.com/caching/docker/
-      # https://github.com/moby/buildkit#s3-cache-experimental
-
-      # images are built and run locally for testing purposes. Not pushed.
-
-      - name: Build Web Docker image
-        uses: ./.github/actions/custom-build-and-push
-        with:
-          context: ./web
-          file: ./web/Dockerfile
-          platforms: linux/amd64
-          tags: onyxdotapp/onyx-web-server:test
-          push: false
-          load: true
-          cache-from: type=s3,prefix=cache/${{ github.repository }}/integration-tests/web-server/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }}
-          cache-to: type=s3,prefix=cache/${{ github.repository }}/integration-tests/web-server/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }},mode=max
-
-      - name: Build Backend Docker image
-        uses: ./.github/actions/custom-build-and-push
-        with:
-          context: ./backend
-          file: ./backend/Dockerfile
-          platforms: linux/amd64
-          tags: onyxdotapp/onyx-backend:test
-          push: false
-          load: true
-          cache-from: type=s3,prefix=cache/${{ github.repository }}/integration-tests/backend/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }}
-          cache-to: type=s3,prefix=cache/${{ github.repository }}/integration-tests/backend/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }},mode=max
-
-      - name: Build Model Server Docker image
-        uses: ./.github/actions/custom-build-and-push
-        with:
-          context: ./backend
-          file: ./backend/Dockerfile.model_server
-          platforms: linux/amd64
-          tags: onyxdotapp/onyx-model-server:test
-          push: false
-          load: true
-          cache-from: type=s3,prefix=cache/${{ github.repository }}/integration-tests/model-server/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }}
-          cache-to: type=s3,prefix=cache/${{ github.repository }}/integration-tests/model-server/,region=${{ env.RUNS_ON_AWS_REGION }},bucket=${{ env.RUNS_ON_S3_BUCKET_CACHE }},mode=max
+      - name: Create .env file for Docker Compose
+        run: |
+          cat <<EOF > deployment/docker_compose/.env
+          ENABLE_PAID_ENTERPRISE_EDITION_FEATURES=true
+          AUTH_TYPE=basic
+          GEN_AI_API_KEY=${{ env.OPENAI_API_KEY }}
+          EXA_API_KEY=${{ env.EXA_API_KEY }}
+          REQUIRE_EMAIL_VERIFICATION=false
+          DISABLE_TELEMETRY=true
+          IMAGE_TAG=test
+          EOF
 
       - name: Start Docker containers
         run: |
           cd deployment/docker_compose
-          ENABLE_PAID_ENTERPRISE_EDITION_FEATURES=true \
-          AUTH_TYPE=basic \
-          GEN_AI_API_KEY=${{ secrets.OPENAI_API_KEY }} \
-          REQUIRE_EMAIL_VERIFICATION=false \
-          DISABLE_TELEMETRY=true \
-          IMAGE_TAG=test \
-          docker compose -f docker-compose.dev.yml -p danswer-stack up -d
+          docker compose -f docker-compose.yml -f docker-compose.dev.yml up -d
         id: start_docker
 
       - name: Wait for service to be ready
         run: |
           echo "Starting wait-for-service script..."
 
-          docker logs -f danswer-stack-api_server-1 &
+          docker logs -f onyx-api_server-1 &
 
           start_time=$(date +%s)
           timeout=300  # 5 minutes in seconds
@@ -137,15 +215,15 @@ jobs:
           while true; do
             current_time=$(date +%s)
             elapsed_time=$((current_time - start_time))
-            
+
             if [ $elapsed_time -ge $timeout ]; then
               echo "Timeout reached. Service did not become ready in 5 minutes."
               exit 1
             fi
-            
+
             # Use curl with error handling to ignore specific exit code 56
             response=$(curl -s -o /dev/null -w "%{http_code}" http://localhost:8080/health || echo "curl_error")
-            
+
             if [ "$response" = "200" ]; then
               echo "Service is ready!"
               break
@@ -154,28 +232,24 @@ jobs:
             else
               echo "Service not ready yet (HTTP status $response). Retrying in 5 seconds..."
             fi
-            
+
             sleep 5
           done
           echo "Finished waiting for service."
 
-      - name: Run pytest playwright test init
-        working-directory: ./backend
-        env:
-          PYTEST_IGNORE_SKIP: true
-        run: pytest -s tests/integration/tests/playwright/test_playwright.py
-
       - name: Run Playwright tests
         working-directory: ./web
-        run: npx playwright test
+        run: |
+          # Create test-results directory to ensure it exists for artifact upload
+          mkdir -p test-results
+          npx playwright test
 
       - uses: actions/upload-artifact@v4
         if: always()
         with:
-          # Chromatic automatically defaults to the test-results directory.
-          # Replace with the path to your custom directory and adjust the CHROMATIC_ARCHIVE_LOCATION environment variable accordingly.
-          name: test-results
-          path: ./web/test-results
+          # Includes test results and trace.zip files
+          name: playwright-test-results-${{ github.run_id }}
+          path: ./web/test-results/
           retention-days: 30
 
       # save before stopping the containers so the logs can be captured
@@ -183,7 +257,7 @@ jobs:
         if: success() || failure()
         run: |
           cd deployment/docker_compose
-          docker compose -f docker-compose.dev.yml -p danswer-stack logs > docker-compose.log
+          docker compose logs > docker-compose.log
           mv docker-compose.log ${{ github.workspace }}/docker-compose.log
 
       - name: Upload logs
@@ -196,7 +270,7 @@ jobs:
       - name: Stop Docker containers
         run: |
           cd deployment/docker_compose
-          docker compose -f docker-compose.dev.yml -p danswer-stack down -v
+          docker compose down -v
 
 # NOTE: Chromatic UI diff testing is currently disabled.
 # We are using Playwright for local and CI testing without visual regression checks.

.github/workflows/pr-python-checks.yml

@@ -31,20 +31,35 @@ jobs:
         pip install --retries 5 --timeout 30 -r backend/requirements/dev.txt
         pip install --retries 5 --timeout 30 -r backend/requirements/model_server.txt
 
+    - name: Generate OpenAPI schema
+      working-directory: ./backend
+      env:
+        PYTHONPATH: "."
+      run: |
+        python scripts/onyx_openapi_schema.py --filename generated/openapi.json
+
+    - name: Generate OpenAPI Python client
+      working-directory: ./backend
+      run: |
+        docker run --rm \
+          -v "${{ github.workspace }}/backend/generated:/local" \
+          openapitools/openapi-generator-cli generate \
+          -i /local/openapi.json \
+          -g python \
+          -o /local/onyx_openapi_client \
+          --package-name onyx_openapi_client \
+          --skip-validate-spec \
+          --openapi-normalizer "SIMPLIFY_ONEOF_ANYOF=true,SET_OAS3_NULLABLE=true"
+            
     - name: Run MyPy
       run: |
         cd backend
         mypy .
 
-    - name: Run ruff
-      run: |
-        cd backend
-        ruff .
-
     - name: Check import order with reorder-python-imports
       run: |
         cd backend
-        find ./danswer -name "*.py" | xargs reorder-python-imports --py311-plus
+        find ./onyx -name "*.py" | xargs reorder-python-imports --py311-plus
 
     - name: Check code formatting with Black
       run: |

.github/workflows/pr-python-connector-tests.yml

@@ -12,54 +12,110 @@ env:
   # AWS
   AWS_ACCESS_KEY_ID_DAILY_CONNECTOR_TESTS: ${{ secrets.AWS_ACCESS_KEY_ID_DAILY_CONNECTOR_TESTS }}
   AWS_SECRET_ACCESS_KEY_DAILY_CONNECTOR_TESTS: ${{ secrets.AWS_SECRET_ACCESS_KEY_DAILY_CONNECTOR_TESTS }}
-  
+
+  # Cloudflare R2
+  R2_ACCOUNT_ID_DAILY_CONNECTOR_TESTS: ${{ vars.R2_ACCOUNT_ID_DAILY_CONNECTOR_TESTS }}
+  R2_ACCESS_KEY_ID_DAILY_CONNECTOR_TESTS: ${{ secrets.R2_ACCESS_KEY_ID_DAILY_CONNECTOR_TESTS }}
+  R2_SECRET_ACCESS_KEY_DAILY_CONNECTOR_TESTS: ${{ secrets.R2_SECRET_ACCESS_KEY_DAILY_CONNECTOR_TESTS }}
+
+  # Google Cloud Storage
+  GCS_ACCESS_KEY_ID_DAILY_CONNECTOR_TESTS: ${{ secrets.GCS_ACCESS_KEY_ID_DAILY_CONNECTOR_TESTS }}
+  GCS_SECRET_ACCESS_KEY_DAILY_CONNECTOR_TESTS: ${{ secrets.GCS_SECRET_ACCESS_KEY_DAILY_CONNECTOR_TESTS }}
+
   # Confluence
-  CONFLUENCE_TEST_SPACE_URL: ${{ secrets.CONFLUENCE_TEST_SPACE_URL }}
-  CONFLUENCE_TEST_SPACE: ${{ secrets.CONFLUENCE_TEST_SPACE }}
-  CONFLUENCE_IS_CLOUD: ${{ secrets.CONFLUENCE_IS_CLOUD }}
+  CONFLUENCE_TEST_SPACE_URL: ${{ vars.CONFLUENCE_TEST_SPACE_URL }}
+  CONFLUENCE_TEST_SPACE: ${{ vars.CONFLUENCE_TEST_SPACE }}
   CONFLUENCE_TEST_PAGE_ID: ${{ secrets.CONFLUENCE_TEST_PAGE_ID }}
-  CONFLUENCE_USER_NAME: ${{ secrets.CONFLUENCE_USER_NAME }}
+  CONFLUENCE_USER_NAME: ${{ vars.CONFLUENCE_USER_NAME }}
   CONFLUENCE_ACCESS_TOKEN: ${{ secrets.CONFLUENCE_ACCESS_TOKEN }}
+  CONFLUENCE_ACCESS_TOKEN_SCOPED: ${{ secrets.CONFLUENCE_ACCESS_TOKEN_SCOPED }}
+
   # Jira
+  JIRA_BASE_URL: ${{ secrets.JIRA_BASE_URL }}
   JIRA_USER_EMAIL: ${{ secrets.JIRA_USER_EMAIL }}
   JIRA_API_TOKEN: ${{ secrets.JIRA_API_TOKEN }}
+  JIRA_API_TOKEN_SCOPED: ${{ secrets.JIRA_API_TOKEN_SCOPED }}
+
+  # Gong
+  GONG_ACCESS_KEY: ${{ secrets.GONG_ACCESS_KEY }}
+  GONG_ACCESS_KEY_SECRET: ${{ secrets.GONG_ACCESS_KEY_SECRET }}
+
   # Google
   GOOGLE_DRIVE_SERVICE_ACCOUNT_JSON_STR: ${{ secrets.GOOGLE_DRIVE_SERVICE_ACCOUNT_JSON_STR }}
   GOOGLE_DRIVE_OAUTH_CREDENTIALS_JSON_STR_TEST_USER_1: ${{ secrets.GOOGLE_DRIVE_OAUTH_CREDENTIALS_JSON_STR_TEST_USER_1 }}
   GOOGLE_DRIVE_OAUTH_CREDENTIALS_JSON_STR: ${{ secrets.GOOGLE_DRIVE_OAUTH_CREDENTIALS_JSON_STR }}
   GOOGLE_GMAIL_SERVICE_ACCOUNT_JSON_STR: ${{ secrets.GOOGLE_GMAIL_SERVICE_ACCOUNT_JSON_STR }}
   GOOGLE_GMAIL_OAUTH_CREDENTIALS_JSON_STR: ${{ secrets.GOOGLE_GMAIL_OAUTH_CREDENTIALS_JSON_STR }}
+
   # Slab
   SLAB_BOT_TOKEN: ${{ secrets.SLAB_BOT_TOKEN }}
+
   # Zendesk
   ZENDESK_SUBDOMAIN: ${{ secrets.ZENDESK_SUBDOMAIN }}
   ZENDESK_EMAIL: ${{ secrets.ZENDESK_EMAIL }}
   ZENDESK_TOKEN: ${{ secrets.ZENDESK_TOKEN }}
+
   # Salesforce
   SF_USERNAME: ${{ secrets.SF_USERNAME }}
   SF_PASSWORD: ${{ secrets.SF_PASSWORD }}
   SF_SECURITY_TOKEN: ${{ secrets.SF_SECURITY_TOKEN }}
+
+  # Hubspot
+  HUBSPOT_ACCESS_TOKEN: ${{ secrets.HUBSPOT_ACCESS_TOKEN }}
+
+  # IMAP
+  IMAP_HOST: ${{ vars.IMAP_HOST }}
+  IMAP_USERNAME: ${{ vars.IMAP_USERNAME }}
+  IMAP_PASSWORD: ${{ secrets.IMAP_PASSWORD }}
+  IMAP_MAILBOXES: ${{ vars.IMAP_MAILBOXES }}
+
   # Airtable
-  AIRTABLE_TEST_BASE_ID: ${{ secrets.AIRTABLE_TEST_BASE_ID }}
-  AIRTABLE_TEST_TABLE_ID: ${{ secrets.AIRTABLE_TEST_TABLE_ID }}
-  AIRTABLE_TEST_TABLE_NAME: ${{ secrets.AIRTABLE_TEST_TABLE_NAME }}
+  AIRTABLE_TEST_BASE_ID: ${{ vars.AIRTABLE_TEST_BASE_ID }}
+  AIRTABLE_TEST_TABLE_ID: ${{ vars.AIRTABLE_TEST_TABLE_ID }}
+  AIRTABLE_TEST_TABLE_NAME: ${{ vars.AIRTABLE_TEST_TABLE_NAME }}
   AIRTABLE_ACCESS_TOKEN: ${{ secrets.AIRTABLE_ACCESS_TOKEN }}
+
   # Sharepoint
-  SHAREPOINT_CLIENT_ID: ${{ secrets.SHAREPOINT_CLIENT_ID }}
+  SHAREPOINT_CLIENT_ID: ${{ vars.SHAREPOINT_CLIENT_ID }}
   SHAREPOINT_CLIENT_SECRET: ${{ secrets.SHAREPOINT_CLIENT_SECRET }}
-  SHAREPOINT_CLIENT_DIRECTORY_ID: ${{ secrets.SHAREPOINT_CLIENT_DIRECTORY_ID }}
-  SHAREPOINT_SITE: ${{ secrets.SHAREPOINT_SITE }}
+  SHAREPOINT_CLIENT_DIRECTORY_ID: ${{ vars.SHAREPOINT_CLIENT_DIRECTORY_ID }}
+  SHAREPOINT_SITE: ${{ vars.SHAREPOINT_SITE }}
+
   # Github
   ACCESS_TOKEN_GITHUB: ${{ secrets.ACCESS_TOKEN_GITHUB }}
+
+  # Gitlab
+  GITLAB_ACCESS_TOKEN: ${{ secrets.GITLAB_ACCESS_TOKEN }}
+
   # Gitbook
   GITBOOK_SPACE_ID: ${{ secrets.GITBOOK_SPACE_ID }}
   GITBOOK_API_KEY: ${{ secrets.GITBOOK_API_KEY }}
+
   # Notion
   NOTION_INTEGRATION_TOKEN: ${{ secrets.NOTION_INTEGRATION_TOKEN }}
+
   # Highspot
   HIGHSPOT_KEY: ${{ secrets.HIGHSPOT_KEY }}
   HIGHSPOT_SECRET: ${{ secrets.HIGHSPOT_SECRET }}
 
+  # Slack
+  SLACK_BOT_TOKEN: ${{ secrets.SLACK_BOT_TOKEN }}
+
+  # Teams
+  TEAMS_APPLICATION_ID: ${{ secrets.TEAMS_APPLICATION_ID }}
+  TEAMS_DIRECTORY_ID: ${{ secrets.TEAMS_DIRECTORY_ID }}
+  TEAMS_SECRET: ${{ secrets.TEAMS_SECRET }}
+
+  # Bitbucket
+  BITBUCKET_WORKSPACE: ${{ secrets.BITBUCKET_WORKSPACE }}
+  BITBUCKET_REPOSITORIES: ${{ secrets.BITBUCKET_REPOSITORIES }}
+  BITBUCKET_PROJECTS: ${{ secrets.BITBUCKET_PROJECTS }}
+  BITBUCKET_EMAIL: ${{ vars.BITBUCKET_EMAIL }}
+  BITBUCKET_API_TOKEN: ${{ secrets.BITBUCKET_API_TOKEN }}
+
+  # Fireflies
+  FIREFLIES_API_KEY: ${{ secrets.FIREFLIES_API_KEY }}
+
 jobs:
   connectors-check:
     # See https://runs-on.com/runners/linux/
@@ -89,9 +145,76 @@ jobs:
           playwright install chromium
           playwright install-deps chromium
 
-      - name: Run Tests
+      - name: Detect Connector changes
+        id: changes
+        uses: dorny/paths-filter@v3
+        with:
+          filters: |
+            hubspot:
+              - 'backend/onyx/connectors/hubspot/**'
+              - 'backend/tests/daily/connectors/hubspot/**'
+            salesforce:
+              - 'backend/onyx/connectors/salesforce/**'
+              - 'backend/tests/daily/connectors/salesforce/**'
+            github:
+              - 'backend/onyx/connectors/github/**'
+              - 'backend/tests/daily/connectors/github/**'
+            file_processing:
+              - 'backend/onyx/file_processing/**'
+
+      - name: Run Tests (excluding HubSpot, Salesforce, and GitHub)
         shell: script -q -e -c "bash --noprofile --norc -eo pipefail {0}"
-        run: py.test -o junit_family=xunit2 -xv --ff backend/tests/daily/connectors
+        run: |
+          py.test \
+            -n 8 \
+            --dist loadfile \
+            --durations=8 \
+            -o junit_family=xunit2 \
+            -xv \
+            --ff \
+            backend/tests/daily/connectors \
+            --ignore backend/tests/daily/connectors/hubspot \
+            --ignore backend/tests/daily/connectors/salesforce \
+            --ignore backend/tests/daily/connectors/github
+
+      - name: Run HubSpot Connector Tests
+        if: ${{ github.event_name == 'schedule' || steps.changes.outputs.hubspot == 'true' || steps.changes.outputs.file_processing == 'true' }}
+        shell: script -q -e -c "bash --noprofile --norc -eo pipefail {0}"
+        run: |
+          py.test \
+            -n 8 \
+            --dist loadfile \
+            --durations=8 \
+            -o junit_family=xunit2 \
+            -xv \
+            --ff \
+            backend/tests/daily/connectors/hubspot
+
+      - name: Run Salesforce Connector Tests
+        if: ${{ github.event_name == 'schedule' || steps.changes.outputs.salesforce == 'true' || steps.changes.outputs.file_processing == 'true' }}
+        shell: script -q -e -c "bash --noprofile --norc -eo pipefail {0}"
+        run: |
+          py.test \
+            -n 8 \
+            --dist loadfile \
+            --durations=8 \
+            -o junit_family=xunit2 \
+            -xv \
+            --ff \
+            backend/tests/daily/connectors/salesforce
+
+      - name: Run GitHub Connector Tests
+        if: ${{ github.event_name == 'schedule' || steps.changes.outputs.github == 'true' || steps.changes.outputs.file_processing == 'true' }}
+        shell: script -q -e -c "bash --noprofile --norc -eo pipefail {0}"
+        run: |
+          py.test \
+            -n 8 \
+            --dist loadfile \
+            --durations=8 \
+            -o junit_family=xunit2 \
+            -xv \
+            --ff \
+            backend/tests/daily/connectors/github
 
       - name: Alert on Failure
         if: failure() && github.event_name == 'schedule'

.github/workflows/pr-python-model-tests.yml

@@ -15,7 +15,7 @@ env:
   # Bedrock
   AWS_ACCESS_KEY_ID: ${{ secrets.AWS_ACCESS_KEY_ID }}
   AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
-  AWS_REGION_NAME: ${{ secrets.AWS_REGION_NAME }}
+  AWS_REGION_NAME: ${{ vars.AWS_REGION_NAME }}
 
   # API keys for testing
   COHERE_API_KEY: ${{ secrets.COHERE_API_KEY }}
@@ -23,7 +23,7 @@ env:
   LITELLM_API_URL: ${{ secrets.LITELLM_API_URL }}
   OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
   AZURE_API_KEY: ${{ secrets.AZURE_API_KEY }}
-  AZURE_API_URL: ${{ secrets.AZURE_API_URL }}
+  AZURE_API_URL: ${{ vars.AZURE_API_URL }}
 
 jobs:
   model-check:
@@ -77,7 +77,7 @@ jobs:
           REQUIRE_EMAIL_VERIFICATION=false \
           DISABLE_TELEMETRY=true \
           IMAGE_TAG=test \
-          docker compose -f docker-compose.model-server-test.yml -p onyx-stack up -d indexing_model_server
+          docker compose -f docker-compose.model-server-test.yml up -d indexing_model_server
         id: start_docker
 
       - name: Wait for service to be ready
@@ -132,7 +132,7 @@ jobs:
         if: always()
         run: |
           cd deployment/docker_compose
-          docker compose -f docker-compose.model-server-test.yml -p onyx-stack logs --no-color > $GITHUB_WORKSPACE/docker-compose.log || true
+          docker compose -f docker-compose.model-server-test.yml logs --no-color > $GITHUB_WORKSPACE/docker-compose.log || true
 
       - name: Upload logs
         if: always()
@@ -145,5 +145,5 @@ jobs:
         if: always()
         run: |
           cd deployment/docker_compose
-          docker compose -f docker-compose.model-server-test.yml -p onyx-stack down -v
+          docker compose -f docker-compose.model-server-test.yml down -v
           

.github/workflows/pr-python-tests.yml

@@ -15,6 +15,9 @@ jobs:
     env:
       PYTHONPATH: ./backend
       REDIS_CLOUD_PYTEST_PASSWORD: ${{ secrets.REDIS_CLOUD_PYTEST_PASSWORD }}
+      SF_USERNAME: ${{ secrets.SF_USERNAME }}
+      SF_PASSWORD: ${{ secrets.SF_PASSWORD }}
+      SF_SECURITY_TOKEN: ${{ secrets.SF_SECURITY_TOKEN }}
       
     steps:
     - name: Checkout code
@@ -28,12 +31,14 @@ jobs:
         cache-dependency-path: |
           backend/requirements/default.txt
           backend/requirements/dev.txt
+          backend/requirements/model_server.txt
 
     - name: Install Dependencies
       run: |
         python -m pip install --upgrade pip
         pip install --retries 5 --timeout 30 -r backend/requirements/default.txt
         pip install --retries 5 --timeout 30 -r backend/requirements/dev.txt
+        pip install --retries 5 --timeout 30 -r backend/requirements/model_server.txt
 
     - name: Run Tests
       shell: script -q -e -c "bash --noprofile --norc -eo pipefail {0}"

.github/workflows/sync_foss.yml

@@ -0,0 +1,47 @@
+name: Sync FOSS Repo
+
+on:
+  schedule:
+    # Run daily at 3am PT (11am UTC during PST)
+    - cron: '0 11 * * *'
+  workflow_dispatch:
+
+jobs:
+  sync-foss:
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+    steps:
+      - name: Checkout main Onyx repo
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - name: Install git-filter-repo
+        run: |
+          sudo apt-get update && sudo apt-get install -y git-filter-repo
+
+      - name: Configure SSH for deploy key
+        env:
+          FOSS_REPO_DEPLOY_KEY: ${{ secrets.FOSS_REPO_DEPLOY_KEY }}
+        run: |
+          mkdir -p ~/.ssh
+          echo "$FOSS_REPO_DEPLOY_KEY" > ~/.ssh/id_ed25519
+          chmod 600 ~/.ssh/id_ed25519
+          ssh-keyscan github.com >> ~/.ssh/known_hosts
+
+      - name: Set Git config
+        run: |
+          git config --global user.name "onyx-bot"
+          git config --global user.email "bot@onyx.app"
+
+      - name: Build FOSS version
+        run: bash backend/scripts/make_foss_repo.sh
+
+      - name: Push to FOSS repo
+        env:
+          FOSS_REPO_URL: git@github.com:onyx-dot-app/onyx-foss.git
+        run: |
+          cd /tmp/foss_repo
+          git remote add public "$FOSS_REPO_URL"
+          git push --force public main

.github/workflows/tag-nightly.yml

@@ -15,6 +15,9 @@ jobs:
       # actions using GITHUB_TOKEN cannot trigger another workflow, but we do want this to trigger docker pushes
       # see https://github.com/orgs/community/discussions/27028#discussioncomment-3254367 for the workaround we
       # implement here which needs an actual user's deploy key
+
+      # Additional NOTE: even though this is named "rkuo", the actual key is tied to the onyx repo
+      # and not rkuo's personal account. It is fine to leave this key as is!
       - name: Checkout code
         uses: actions/checkout@v4
         with:

.gitignore

@@ -1,12 +1,47 @@
-.env
+# editors
+.vscode
+.zed
+
+# macos
 .DS_store
+
+# python
 .venv
 .mypy_cache
 .idea
-/deployment/data/nginx/app.conf
-.vscode/
-*.sw?
-/backend/tests/regression/answer_quality/search_test_config.yaml
+
+# testing
 /web/test-results/
 backend/onyx/agent_search/main/test_data.json
 backend/tests/regression/answer_quality/test_data.json
+backend/tests/regression/search_quality/eval-*
+backend/tests/regression/search_quality/search_eval_config.yaml
+backend/tests/regression/search_quality/*.json
+backend/onyx/evals/data/
+*.log
+
+# secret files
+.env
+jira_test_env
+settings.json
+
+# others
+/deployment/data/nginx/app.conf
+*.sw?
+/backend/tests/regression/answer_quality/search_test_config.yaml
+*.egg-info
+
+# Claude
+AGENTS.md
+CLAUDE.md
+
+
+# Local .terraform directories
+**/.terraform/*
+
+# Local .tfstate files
+*.tfstate
+*.tfstate.*
+
+# Local .terraform.lock.hcl file
+.terraform.lock.hcl

.mcp.json.template

@@ -0,0 +1,8 @@
+{
+  "mcpServers": {
+    "onyx-mcp": {
+      "type": "http",
+      "url": "http://localhost:8000/mcp"
+    }
+  }
+}

.pre-commit-config.yaml

@@ -1,12 +1,13 @@
 repos:
   - repo: https://github.com/psf/black
-    rev: 23.3.0
+    rev: 25.1.0
     hooks:
     - id: black
       language_version: python3.11
 
-  - repo: https://github.com/asottile/reorder_python_imports
-    rev: v3.9.0
+  # this is a fork which keeps compatibility with black
+  - repo: https://github.com/wimglenn/reorder-python-imports-black
+    rev: v3.14.0
     hooks:
     - id: reorder-python-imports
       args: ['--py311-plus', '--application-directories=backend/']
@@ -18,23 +19,50 @@ repos:
   # These settings will remove unused imports with side effects
   # Note: The repo currently does not and should not have imports with side effects
   - repo: https://github.com/PyCQA/autoflake
-    rev: v2.2.0
+    rev: v2.3.1
     hooks:
       - id: autoflake
         args: [ '--remove-all-unused-imports', '--remove-unused-variables', '--in-place' , '--recursive']
 
   - repo: https://github.com/astral-sh/ruff-pre-commit
     # Ruff version.
-    rev: v0.0.286
+    rev: v0.11.4
     hooks:
       - id: ruff
+
   - repo: https://github.com/pre-commit/mirrors-prettier
     rev: v3.1.0
     hooks:
     - id: prettier
       types_or: [html, css, javascript, ts, tsx]
-      additional_dependencies:
-      - prettier
+      language_version: system
+
+  - repo: https://github.com/sirwart/ripsecrets
+    rev: v0.1.11
+    hooks:
+      - id: ripsecrets
+        args:
+        - --additional-pattern
+        - ^sk-[A-Za-z0-9_\-]{20,}$
+
+
+  - repo: local
+    hooks:
+      - id: terraform-fmt
+        name: terraform fmt
+        entry: terraform fmt -recursive
+        language: system
+        pass_filenames: false
+        files: \.tf$
+      - id: check-lazy-imports
+        name: Check lazy imports are not directly imported
+        entry: python3 backend/scripts/check_lazy_imports.py
+        language: system
+        files: ^backend/(?!\.venv/).*\.py$
+        pass_filenames: false
+        # Note: pass_filenames is false because tsc must check the entire
+        # project, but the files filter ensures this only runs when relevant
+        # files change. Using --incremental for faster subsequent checks.
 
   # We would like to have a mypy pre-commit hook, but due to the fact that
   # pre-commit runs in it's own isolated environment, we would need to install

.vscode/env_template.txt

@@ -1,6 +1,6 @@
 # Copy this file to .env in the .vscode folder
 # Fill in the <REPLACE THIS> values as needed, it is recommended to set the GEN_AI_API_KEY value to avoid having to set up an LLM in the UI
-# Also check out danswer/backend/scripts/restart_containers.sh for a script to restart the containers which Danswer relies on outside of VSCode/Cursor processes
+# Also check out onyx/backend/scripts/restart_containers.sh for a script to restart the containers which Onyx relies on outside of VSCode/Cursor processes
 
 # For local dev, often user Authentication is not needed
 AUTH_TYPE=disabled
@@ -10,7 +10,7 @@ SKIP_WARM_UP=True
 
 # Always keep these on for Dev
 # Logs all model prompts to stdout
-LOG_DANSWER_MODEL_INTERACTIONS=True
+LOG_ONYX_MODEL_INTERACTIONS=True
 # More verbose logging
 LOG_LEVEL=debug
 
@@ -23,6 +23,9 @@ DISABLE_LLM_DOC_RELEVANCE=False
 # Useful if you want to toggle auth on/off (google_oauth/OIDC specifically)
 OAUTH_CLIENT_ID=<REPLACE THIS>
 OAUTH_CLIENT_SECRET=<REPLACE THIS>
+OPENID_CONFIG_URL=<REPLACE THIS>
+SAML_CONF_DIR=/<ABSOLUTE PATH TO ONYX>/onyx/backend/ee/onyx/configs/saml_config
+
 # Generally not useful for dev, we don't generally want to set up an SMTP server for dev
 REQUIRE_EMAIL_VERIFICATION=False
 
@@ -34,10 +37,10 @@ OPENAI_API_KEY=<REPLACE THIS>
 GEN_AI_MODEL_VERSION=gpt-4o
 FAST_GEN_AI_MODEL_VERSION=gpt-4o
 
-# For Danswer Slack Bot, overrides the UI values so no need to set this up via UI every time
-# Only needed if using DanswerBot
-#DANSWER_BOT_SLACK_APP_TOKEN=<REPLACE THIS>
-#DANSWER_BOT_SLACK_BOT_TOKEN=<REPLACE THIS>
+# For Onyx Slack Bot, overrides the UI values so no need to set this up via UI every time
+# Only needed if using OnyxBot
+#ONYX_BOT_SLACK_APP_TOKEN=<REPLACE THIS>
+#ONYX_BOT_SLACK_BOT_TOKEN=<REPLACE THIS>
 
 
 # Python stuff
@@ -45,8 +48,8 @@ PYTHONPATH=../backend
 PYTHONUNBUFFERED=1
 
 
-# Internet Search 
-BING_API_KEY=<REPLACE THIS>
+# Internet Search
+EXA_API_KEY=<REPLACE THIS>
 
 
 # Enable the full set of Danswer Enterprise Edition features
@@ -58,3 +61,23 @@ AGENT_RETRIEVAL_STATS=False   # Note: This setting will incur substantial re-ran
 AGENT_RERANKING_STATS=True
 AGENT_MAX_QUERY_RETRIEVAL_RESULTS=20
 AGENT_RERANKING_MAX_QUERY_RETRIEVAL_RESULTS=20
+
+# S3 File Store Configuration (MinIO for local development)
+S3_ENDPOINT_URL=http://localhost:9004
+S3_FILE_STORE_BUCKET_NAME=onyx-file-store-bucket
+S3_AWS_ACCESS_KEY_ID=minioadmin
+S3_AWS_SECRET_ACCESS_KEY=minioadmin
+
+# Show extra/uncommon connectors
+SHOW_EXTRA_CONNECTORS=True
+
+# Local langsmith tracing
+LANGSMITH_TRACING="true"
+LANGSMITH_ENDPOINT="https://api.smith.langchain.com"
+LANGSMITH_API_KEY=<REPLACE_THIS>
+LANGSMITH_PROJECT=<REPLACE_THIS>
+
+# Local Confluence OAuth testing
+# OAUTH_CONFLUENCE_CLOUD_CLIENT_ID=<REPLACE_THIS>
+# OAUTH_CONFLUENCE_CLOUD_CLIENT_SECRET=<REPLACE_THIS>
+# NEXT_PUBLIC_TEST_ENV=True

.vscode/tasks.template.jsonc

@@ -0,0 +1,101 @@
+{
+    "version": "2.0.0",
+    "tasks": [
+        {
+            "type": "austin",
+            "label": "Profile celery beat",
+            "envFile": "${workspaceFolder}/.env",
+            "options": {
+              "cwd": "${workspaceFolder}/backend"
+            },
+            "command": [
+                "sudo",
+                "-E"
+            ],
+            "args": [
+              "celery",
+              "-A",
+              "onyx.background.celery.versioned_apps.beat",
+              "beat",
+              "--loglevel=INFO"
+            ]
+        },
+        {
+            "type": "shell",
+            "label": "Generate Onyx OpenAPI Python client",
+            "cwd": "${workspaceFolder}/backend",
+            "envFile": "${workspaceFolder}/.env",
+            "options": {
+              "cwd": "${workspaceFolder}/backend"
+            },
+            "command": [
+                "openapi-generator"
+            ],
+            "args": [
+                "generate",
+                "-i",
+                "generated/openapi.json",
+                "-g",
+                "python",
+                "-o",
+                "generated/onyx_openapi_client",
+                "--package-name",
+                "onyx_openapi_client",
+            ]
+        },
+        {
+            "type": "shell",
+            "label": "Generate Typescript Fetch client (openapi-generator)",
+            "envFile": "${workspaceFolder}/.env",
+            "options": {
+              "cwd": "${workspaceFolder}"
+            },
+            "command": [
+                "openapi-generator"
+            ],
+            "args": [
+                "generate",
+                "-i",
+                "backend/generated/openapi.json",
+                "-g",
+                "typescript-fetch",
+                "-o",
+                "${workspaceFolder}/web/src/lib/generated/onyx_api",
+                "--additional-properties=disallowAdditionalPropertiesIfNotPresent=false,legacyDiscriminatorBehavior=false,supportsES6=true",
+            ]
+        },
+        {
+            "type": "shell",
+            "label": "Generate TypeScript Client (openapi-ts)",
+            "envFile": "${workspaceFolder}/.env",
+            "options": {
+              "cwd": "${workspaceFolder}/web"
+            },
+            "command": [
+                "npx"
+            ],
+            "args": [
+                "openapi-typescript",
+                "../backend/generated/openapi.json",
+                "--output",
+                "./src/lib/generated/onyx-schema.ts",
+            ]
+        },
+        {
+            "type": "shell",
+            "label": "Generate TypeScript Client (orval)",
+            "envFile": "${workspaceFolder}/.env",
+            "options": {
+              "cwd": "${workspaceFolder}/web"
+            },
+            "command": [
+                "npx"
+            ],
+            "args": [
+            	"orval",
+                "--config",
+                "orval.config.js",
+            ]
+        }
+    ]
+}

AGENTS.md.template

@@ -0,0 +1,327 @@
+# AGENTS.md
+
+This file provides guidance to Codex when working with code in this repository.
+
+## KEY NOTES
+
+- If you run into any missing python dependency errors, try running your command with `source backend/.venv/bin/activate` \
+to assume the python venv.
+- To make tests work, check the `.env` file at the root of the project to find an OpenAI key.
+- If using `playwright` to explore the frontend, you can usually log in with username `a@test.com` and password
+`a`. The app can be accessed at `http://localhost:3000`.
+- You should assume that all Onyx services are running. To verify, you can check the `backend/log` directory to
+make sure we see logs coming out from the relevant service.
+- To connect to the Postgres database, use: `docker exec -it onyx-relational_db-1 psql -U postgres -c "<SQL>"`
+- When making calls to the backend, always go through the frontend. E.g. make a call to `http://localhost:3000/api/persona` not `http://localhost:8080/api/persona`
+- Put ALL db operations under the `backend/onyx/db` / `backend/ee/onyx/db` directories. Don't run queries
+outside of those directories.
+
+## Project Overview
+
+**Onyx** (formerly Danswer) is an open-source Gen-AI and Enterprise Search platform that connects to company documents, apps, and people. It features a modular architecture with both Community Edition (MIT licensed) and Enterprise Edition offerings.
+
+
+### Background Workers (Celery)
+
+Onyx uses Celery for asynchronous task processing with multiple specialized workers:
+
+#### Worker Types
+
+1. **Primary Worker** (`celery_app.py`)
+   - Coordinates core background tasks and system-wide operations
+   - Handles connector management, document sync, pruning, and periodic checks
+   - Runs with 4 threads concurrency
+   - Tasks: connector deletion, vespa sync, pruning, LLM model updates, user file sync
+
+2. **Docfetching Worker** (`docfetching`)
+   - Fetches documents from external data sources (connectors)
+   - Spawns docprocessing tasks for each document batch
+   - Implements watchdog monitoring for stuck connectors
+   - Configurable concurrency (default from env)
+
+3. **Docprocessing Worker** (`docprocessing`)
+   - Processes fetched documents through the indexing pipeline:
+     - Upserts documents to PostgreSQL
+     - Chunks documents and adds contextual information
+     - Embeds chunks via model server
+     - Writes chunks to Vespa vector database
+     - Updates document metadata
+   - Configurable concurrency (default from env)
+
+4. **Light Worker** (`light`)
+   - Handles lightweight, fast operations
+   - Tasks: vespa operations, document permissions sync, external group sync
+   - Higher concurrency for quick tasks
+
+5. **Heavy Worker** (`heavy`)
+   - Handles resource-intensive operations
+   - Primary task: document pruning operations
+   - Runs with 4 threads concurrency
+
+6. **KG Processing Worker** (`kg_processing`)
+   - Handles Knowledge Graph processing and clustering
+   - Builds relationships between documents
+   - Runs clustering algorithms
+   - Configurable concurrency
+
+7. **Monitoring Worker** (`monitoring`)
+   - System health monitoring and metrics collection
+   - Monitors Celery queues, process memory, and system status
+   - Single thread (monitoring doesn't need parallelism)
+   - Cloud-specific monitoring tasks
+
+8. **User File Processing Worker** (`user_file_processing`)
+   - Processes user-uploaded files
+   - Handles user file indexing and project synchronization
+   - Configurable concurrency
+
+9. **Beat Worker** (`beat`)
+   - Celery's scheduler for periodic tasks
+   - Uses DynamicTenantScheduler for multi-tenant support
+   - Schedules tasks like:
+     - Indexing checks (every 15 seconds)
+     - Connector deletion checks (every 20 seconds)
+     - Vespa sync checks (every 20 seconds)
+     - Pruning checks (every 20 seconds)
+     - KG processing (every 60 seconds)
+     - Monitoring tasks (every 5 minutes)
+     - Cleanup tasks (hourly)
+
+#### Worker Deployment Modes
+
+Onyx supports two deployment modes for background workers, controlled by the `USE_LIGHTWEIGHT_BACKGROUND_WORKER` environment variable:
+
+**Lightweight Mode** (default, `USE_LIGHTWEIGHT_BACKGROUND_WORKER=true`):
+- Runs a single consolidated `background` worker that handles all background tasks:
+  - Pruning operations (from `heavy` worker)
+  - Knowledge graph processing (from `kg_processing` worker)
+  - Monitoring tasks (from `monitoring` worker)
+  - User file processing (from `user_file_processing` worker)
+- Lower resource footprint (single worker process)
+- Suitable for smaller deployments or development environments
+- Default concurrency: 6 threads
+
+**Standard Mode** (`USE_LIGHTWEIGHT_BACKGROUND_WORKER=false`):
+- Runs separate specialized workers as documented above (heavy, kg_processing, monitoring, user_file_processing)
+- Better isolation and scalability
+- Can scale individual workers independently based on workload
+- Suitable for production deployments with higher load
+
+The deployment mode affects:
+- **Backend**: Worker processes spawned by supervisord or dev scripts
+- **Helm**: Which Kubernetes deployments are created
+- **Dev Environment**: Which workers `dev_run_background_jobs.py` spawns
+
+#### Key Features
+
+- **Thread-based Workers**: All workers use thread pools (not processes) for stability
+- **Tenant Awareness**: Multi-tenant support with per-tenant task isolation. There is a 
+middleware layer that automatically finds the appropriate tenant ID when sending tasks 
+via Celery Beat.
+- **Task Prioritization**: High, Medium, Low priority queues
+- **Monitoring**: Built-in heartbeat and liveness checking
+- **Failure Handling**: Automatic retry and failure recovery mechanisms
+- **Redis Coordination**: Inter-process communication via Redis
+- **PostgreSQL State**: Task state and metadata stored in PostgreSQL
+
+
+#### Important Notes
+
+**Defining Tasks**: 
+- Always use `@shared_task` rather than `@celery_app`
+- Put tasks under `background/celery/tasks/` or `ee/background/celery/tasks`
+
+**Defining APIs**:
+When creating new FastAPI APIs, do NOT use the `response_model` field. Instead, just type the
+function.
+
+**Testing Updates**:
+If you make any updates to a celery worker and you want to test these changes, you will need
+to ask me to restart the celery worker. There is no auto-restart on code-change mechanism.
+
+### Code Quality
+```bash
+# Install and run pre-commit hooks
+pre-commit install
+pre-commit run --all-files
+```
+
+NOTE: Always make sure everything is strictly typed (both in Python and Typescript).
+
+## Architecture Overview
+
+### Technology Stack
+- **Backend**: Python 3.11, FastAPI, SQLAlchemy, Alembic, Celery
+- **Frontend**: Next.js 15+, React 18, TypeScript, Tailwind CSS
+- **Database**: PostgreSQL with Redis caching
+- **Search**: Vespa vector database
+- **Auth**: OAuth2, SAML, multi-provider support
+- **AI/ML**: LangChain, LiteLLM, multiple embedding models
+
+### Directory Structure
+
+```
+backend/
+├── onyx/
+│   ├── auth/                    # Authentication & authorization
+│   ├── chat/                    # Chat functionality & LLM interactions
+│   ├── connectors/              # Data source connectors
+│   ├── db/                      # Database models & operations
+│   ├── document_index/          # Vespa integration
+│   ├── federated_connectors/    # External search connectors
+│   ├── llm/                     # LLM provider integrations
+│   └── server/                  # API endpoints & routers
+├── ee/                          # Enterprise Edition features
+├── alembic/                     # Database migrations
+└── tests/                       # Test suites
+
+web/
+├── src/app/                     # Next.js app router pages
+├── src/components/              # Reusable React components
+└── src/lib/                     # Utilities & business logic
+```
+
+## Database & Migrations
+
+### Running Migrations
+```bash
+# Standard migrations
+alembic upgrade head
+
+# Multi-tenant (Enterprise)
+alembic -n schema_private upgrade head
+```
+
+### Creating Migrations
+```bash
+# Create migration
+alembic revision -m "description"
+
+# Multi-tenant migration
+alembic -n schema_private revision -m "description"
+```
+
+Write the migration manually and place it in the file that alembic creates when running the above command.
+
+## Testing Strategy
+
+There are 4 main types of tests within Onyx:
+
+### Unit Tests
+These should not assume any Onyx/external services are available to be called.
+Interactions with the outside world should be mocked using `unittest.mock`. Generally, only 
+write these for complex, isolated modules e.g. `citation_processing.py`.
+
+To run them:
+
+```bash
+python -m dotenv -f .vscode/.env run -- pytest -xv backend/tests/unit
+```
+
+### External Dependency Unit Tests
+These tests assume that all external dependencies of Onyx are available and callable (e.g. Postgres, Redis, 
+MinIO/S3, Vespa are running + OpenAI can be called + any request to the internet is fine + etc.).
+
+However, the actual Onyx containers are not running and with these tests we call the function to test directly.
+We can also mock components/calls at will. 
+
+The goal with these tests are to minimize mocking while giving some flexibility to mock things that are flakey, 
+need strictly controlled behavior, or need to have their internal behavior validated (e.g. verify a function is called
+with certain args, something that would be impossible with proper integration tests).
+
+A great example of this type of test is `backend/tests/external_dependency_unit/connectors/confluence/test_confluence_group_sync.py`.
+
+To run them:
+
+```bash
+python -m dotenv -f .vscode/.env run -- pytest backend/tests/external_dependency_unit
+```
+
+### Integration Tests
+Standard integration tests. Every test in `backend/tests/integration` runs against a real Onyx deployment. We cannot 
+mock anything in these tests. Prefer writing integration tests (or External Dependency Unit Tests if mocking/internal 
+verification is necessary) over any other type of test.
+
+Tests are parallelized at a directory level.
+
+When writing integration tests, make sure to check the root `conftest.py` for useful fixtures + the `backend/tests/integration/common_utils` directory for utilities. Prefer (if one exists), calling the appropriate Manager 
+class in the utils over directly calling the APIs with a library like `requests`. Prefer using fixtures rather than
+calling the utilities directly (e.g. do NOT create admin users with 
+`admin_user = UserManager.create(name="admin_user")`, instead use the `admin_user` fixture).
+
+A great example of this type of test is `backend/tests/integration/dev_apis/test_simple_chat_api.py`.
+
+To run them:
+
+```bash
+python -m dotenv -f .vscode/.env run -- pytest backend/tests/integration
+```
+
+### Playwright (E2E) Tests
+These tests are an even more complete version of the Integration Tests mentioned above. Has all services of Onyx 
+running, *including* the Web Server.
+
+Use these tests for anything that requires significant frontend <-> backend coordination.
+
+Tests are located at `web/tests/e2e`. Tests are written in TypeScript.
+
+To run them:
+
+```bash
+npx playwright test <TEST_NAME>
+```
+
+
+## Logs
+
+When (1) writing integration tests or (2) doing live tests (e.g. curl / playwright) you can get access
+to logs via the `backend/log/<service_name>_debug.log` file. All Onyx services (api_server, web_server, celery_X)
+will be tailing their logs to this file. 
+
+
+## Security Considerations
+
+- Never commit API keys or secrets to repository
+- Use encrypted credential storage for connector credentials
+- Follow RBAC patterns for new features
+- Implement proper input validation with Pydantic models
+- Use parameterized queries to prevent SQL injection
+
+## AI/LLM Integration
+
+- Multiple LLM providers supported via LiteLLM
+- Configurable models per feature (chat, search, embeddings)
+- Streaming support for real-time responses
+- Token management and rate limiting
+- Custom prompts and agent actions
+
+## UI/UX Patterns
+
+- Tailwind CSS with design system in `web/src/components/ui/`
+- Radix UI and Headless UI for accessible components
+- SWR for data fetching and caching
+- Form validation with react-hook-form
+- Error handling with popup notifications
+
+## Creating a Plan
+When creating a plan in the `plans` directory, make sure to include at least these elements:
+
+**Issues to Address**
+What the change is meant to do.
+
+**Important Notes**
+Things you come across in your research that are important to the implementation.
+
+**Implementation strategy**
+How you are going to make the changes happen. High level approach.
+
+**Tests**
+What unit (use rarely), external dependency unit, integration, and playwright tests you plan to write to 
+verify the correct behavior. Don't overtest. Usually, a given change only needs one type of test.
+
+Do NOT include these: *Timeline*, *Rollback plan*
+
+This is a minimal list - feel free to include more. Do NOT write code as part of your plan.
+Keep it high level. You can reference certain files or functions though.
+
+Before writing your plan, make sure to do research. Explore the relevant sections in the codebase.

CLAUDE.md.template

@@ -0,0 +1,332 @@
+# CLAUDE.md
+
+This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
+
+## KEY NOTES
+
+- If you run into any missing python dependency errors, try running your command with `source backend/.venv/bin/activate` \
+to assume the python venv.
+- To make tests work, check the `.env` file at the root of the project to find an OpenAI key.
+- If using `playwright` to explore the frontend, you can usually log in with username `a@test.com` and password
+`a`. The app can be accessed at `http://localhost:3000`.
+- You should assume that all Onyx services are running. To verify, you can check the `backend/log` directory to
+make sure we see logs coming out from the relevant service.
+- To connect to the Postgres database, use: `docker exec -it onyx-relational_db-1 psql -U postgres -c "<SQL>"`
+- When making calls to the backend, always go through the frontend. E.g. make a call to `http://localhost:3000/api/persona` not `http://localhost:8080/api/persona`
+- Put ALL db operations under the `backend/onyx/db` / `backend/ee/onyx/db` directories. Don't run queries
+outside of those directories.
+
+## Project Overview
+
+**Onyx** (formerly Danswer) is an open-source Gen-AI and Enterprise Search platform that connects to company documents, apps, and people. It features a modular architecture with both Community Edition (MIT licensed) and Enterprise Edition offerings.
+
+
+### Background Workers (Celery)
+
+Onyx uses Celery for asynchronous task processing with multiple specialized workers:
+
+#### Worker Types
+
+1. **Primary Worker** (`celery_app.py`)
+   - Coordinates core background tasks and system-wide operations
+   - Handles connector management, document sync, pruning, and periodic checks
+   - Runs with 4 threads concurrency
+   - Tasks: connector deletion, vespa sync, pruning, LLM model updates, user file sync
+
+2. **Docfetching Worker** (`docfetching`)
+   - Fetches documents from external data sources (connectors)
+   - Spawns docprocessing tasks for each document batch
+   - Implements watchdog monitoring for stuck connectors
+   - Configurable concurrency (default from env)
+
+3. **Docprocessing Worker** (`docprocessing`)
+   - Processes fetched documents through the indexing pipeline:
+     - Upserts documents to PostgreSQL
+     - Chunks documents and adds contextual information
+     - Embeds chunks via model server
+     - Writes chunks to Vespa vector database
+     - Updates document metadata
+   - Configurable concurrency (default from env)
+
+4. **Light Worker** (`light`)
+   - Handles lightweight, fast operations
+   - Tasks: vespa operations, document permissions sync, external group sync
+   - Higher concurrency for quick tasks
+
+5. **Heavy Worker** (`heavy`)
+   - Handles resource-intensive operations
+   - Primary task: document pruning operations
+   - Runs with 4 threads concurrency
+
+6. **KG Processing Worker** (`kg_processing`)
+   - Handles Knowledge Graph processing and clustering
+   - Builds relationships between documents
+   - Runs clustering algorithms
+   - Configurable concurrency
+
+7. **Monitoring Worker** (`monitoring`)
+   - System health monitoring and metrics collection
+   - Monitors Celery queues, process memory, and system status
+   - Single thread (monitoring doesn't need parallelism)
+   - Cloud-specific monitoring tasks
+
+8. **User File Processing Worker** (`user_file_processing`)
+   - Processes user-uploaded files
+   - Handles user file indexing and project synchronization
+   - Configurable concurrency
+
+9. **Beat Worker** (`beat`)
+   - Celery's scheduler for periodic tasks
+   - Uses DynamicTenantScheduler for multi-tenant support
+   - Schedules tasks like:
+     - Indexing checks (every 15 seconds)
+     - Connector deletion checks (every 20 seconds)
+     - Vespa sync checks (every 20 seconds)
+     - Pruning checks (every 20 seconds)
+     - KG processing (every 60 seconds)
+     - Monitoring tasks (every 5 minutes)
+     - Cleanup tasks (hourly)
+
+#### Worker Deployment Modes
+
+Onyx supports two deployment modes for background workers, controlled by the `USE_LIGHTWEIGHT_BACKGROUND_WORKER` environment variable:
+
+**Lightweight Mode** (default, `USE_LIGHTWEIGHT_BACKGROUND_WORKER=true`):
+- Runs a single consolidated `background` worker that handles all background tasks:
+  - Light worker tasks (Vespa operations, permissions sync, deletion)
+  - Document processing (indexing pipeline)
+  - Document fetching (connector data retrieval)
+  - Pruning operations (from `heavy` worker)
+  - Knowledge graph processing (from `kg_processing` worker)
+  - Monitoring tasks (from `monitoring` worker)
+  - User file processing (from `user_file_processing` worker)
+- Lower resource footprint (fewer worker processes)
+- Suitable for smaller deployments or development environments
+- Default concurrency: 20 threads (increased to handle combined workload)
+
+**Standard Mode** (`USE_LIGHTWEIGHT_BACKGROUND_WORKER=false`):
+- Runs separate specialized workers as documented above (light, docprocessing, docfetching, heavy, kg_processing, monitoring, user_file_processing)
+- Better isolation and scalability
+- Can scale individual workers independently based on workload
+- Suitable for production deployments with higher load
+
+The deployment mode affects:
+- **Backend**: Worker processes spawned by supervisord or dev scripts
+- **Helm**: Which Kubernetes deployments are created
+- **Dev Environment**: Which workers `dev_run_background_jobs.py` spawns
+
+#### Key Features
+
+- **Thread-based Workers**: All workers use thread pools (not processes) for stability
+- **Tenant Awareness**: Multi-tenant support with per-tenant task isolation. There is a
+middleware layer that automatically finds the appropriate tenant ID when sending tasks
+via Celery Beat.
+- **Task Prioritization**: High, Medium, Low priority queues
+- **Monitoring**: Built-in heartbeat and liveness checking
+- **Failure Handling**: Automatic retry and failure recovery mechanisms
+- **Redis Coordination**: Inter-process communication via Redis
+- **PostgreSQL State**: Task state and metadata stored in PostgreSQL
+
+
+#### Important Notes
+
+**Defining Tasks**: 
+- Always use `@shared_task` rather than `@celery_app`
+- Put tasks under `background/celery/tasks/` or `ee/background/celery/tasks`
+
+**Defining APIs**:
+When creating new FastAPI APIs, do NOT use the `response_model` field. Instead, just type the
+function.
+
+**Testing Updates**:
+If you make any updates to a celery worker and you want to test these changes, you will need
+to ask me to restart the celery worker. There is no auto-restart on code-change mechanism.
+
+### Code Quality
+```bash
+# Install and run pre-commit hooks
+pre-commit install
+pre-commit run --all-files
+```
+
+NOTE: Always make sure everything is strictly typed (both in Python and Typescript).
+
+## Architecture Overview
+
+### Technology Stack
+- **Backend**: Python 3.11, FastAPI, SQLAlchemy, Alembic, Celery
+- **Frontend**: Next.js 15+, React 18, TypeScript, Tailwind CSS
+- **Database**: PostgreSQL with Redis caching
+- **Search**: Vespa vector database
+- **Auth**: OAuth2, SAML, multi-provider support
+- **AI/ML**: LangChain, LiteLLM, multiple embedding models
+
+### Directory Structure
+
+```
+backend/
+├── onyx/
+│   ├── auth/                    # Authentication & authorization
+│   ├── chat/                    # Chat functionality & LLM interactions
+│   ├── connectors/              # Data source connectors
+│   ├── db/                      # Database models & operations
+│   ├── document_index/          # Vespa integration
+│   ├── federated_connectors/    # External search connectors
+│   ├── llm/                     # LLM provider integrations
+│   └── server/                  # API endpoints & routers
+├── ee/                          # Enterprise Edition features
+├── alembic/                     # Database migrations
+└── tests/                       # Test suites
+
+web/
+├── src/app/                     # Next.js app router pages
+├── src/components/              # Reusable React components
+└── src/lib/                     # Utilities & business logic
+```
+
+## Database & Migrations
+
+### Running Migrations
+```bash
+# Standard migrations
+alembic upgrade head
+
+# Multi-tenant (Enterprise)
+alembic -n schema_private upgrade head
+```
+
+### Creating Migrations
+```bash
+# Create migration
+alembic revision -m "description"
+
+# Multi-tenant migration
+alembic -n schema_private revision -m "description"
+```
+
+Write the migration manually and place it in the file that alembic creates when running the above command.
+
+## Testing Strategy
+
+First, you must activate the virtual environment with `source .venv/bin/activate`. 
+
+There are 4 main types of tests within Onyx:
+
+### Unit Tests
+These should not assume any Onyx/external services are available to be called.
+Interactions with the outside world should be mocked using `unittest.mock`. Generally, only 
+write these for complex, isolated modules e.g. `citation_processing.py`.
+
+To run them:
+
+```bash
+pytest -xv backend/tests/unit
+```
+
+### External Dependency Unit Tests
+These tests assume that all external dependencies of Onyx are available and callable (e.g. Postgres, Redis, 
+MinIO/S3, Vespa are running + OpenAI can be called + any request to the internet is fine + etc.).
+
+However, the actual Onyx containers are not running and with these tests we call the function to test directly.
+We can also mock components/calls at will. 
+
+The goal with these tests are to minimize mocking while giving some flexibility to mock things that are flakey, 
+need strictly controlled behavior, or need to have their internal behavior validated (e.g. verify a function is called
+with certain args, something that would be impossible with proper integration tests).
+
+A great example of this type of test is `backend/tests/external_dependency_unit/connectors/confluence/test_confluence_group_sync.py`.
+
+To run them:
+
+```bash
+python -m dotenv -f .vscode/.env run -- pytest backend/tests/external_dependency_unit
+```
+
+### Integration Tests
+Standard integration tests. Every test in `backend/tests/integration` runs against a real Onyx deployment. We cannot 
+mock anything in these tests. Prefer writing integration tests (or External Dependency Unit Tests if mocking/internal 
+verification is necessary) over any other type of test.
+
+Tests are parallelized at a directory level.
+
+When writing integration tests, make sure to check the root `conftest.py` for useful fixtures + the `backend/tests/integration/common_utils` directory for utilities. Prefer (if one exists), calling the appropriate Manager 
+class in the utils over directly calling the APIs with a library like `requests`. Prefer using fixtures rather than
+calling the utilities directly (e.g. do NOT create admin users with 
+`admin_user = UserManager.create(name="admin_user")`, instead use the `admin_user` fixture).
+
+A great example of this type of test is `backend/tests/integration/dev_apis/test_simple_chat_api.py`.
+
+To run them:
+
+```bash
+python -m dotenv -f .vscode/.env run -- pytest backend/tests/integration
+```
+
+### Playwright (E2E) Tests
+These tests are an even more complete version of the Integration Tests mentioned above. Has all services of Onyx 
+running, *including* the Web Server.
+
+Use these tests for anything that requires significant frontend <-> backend coordination.
+
+Tests are located at `web/tests/e2e`. Tests are written in TypeScript.
+
+To run them:
+
+```bash
+npx playwright test <TEST_NAME>
+```
+
+
+## Logs
+
+When (1) writing integration tests or (2) doing live tests (e.g. curl / playwright) you can get access
+to logs via the `backend/log/<service_name>_debug.log` file. All Onyx services (api_server, web_server, celery_X)
+will be tailing their logs to this file. 
+
+
+## Security Considerations
+
+- Never commit API keys or secrets to repository
+- Use encrypted credential storage for connector credentials
+- Follow RBAC patterns for new features
+- Implement proper input validation with Pydantic models
+- Use parameterized queries to prevent SQL injection
+
+## AI/LLM Integration
+
+- Multiple LLM providers supported via LiteLLM
+- Configurable models per feature (chat, search, embeddings)
+- Streaming support for real-time responses
+- Token management and rate limiting
+- Custom prompts and agent actions
+
+## UI/UX Patterns
+
+- Tailwind CSS with design system in `web/src/components/ui/`
+- Radix UI and Headless UI for accessible components
+- SWR for data fetching and caching
+- Form validation with react-hook-form
+- Error handling with popup notifications
+
+## Creating a Plan
+When creating a plan in the `plans` directory, make sure to include at least these elements:
+
+**Issues to Address**
+What the change is meant to do.
+
+**Important Notes**
+Things you come across in your research that are important to the implementation.
+
+**Implementation strategy**
+How you are going to make the changes happen. High level approach.
+
+**Tests**
+What unit (use rarely), external dependency unit, integration, and playwright tests you plan to write to 
+verify the correct behavior. Don't overtest. Usually, a given change only needs one type of test.
+
+Do NOT include these: *Timeline*, *Rollback plan*
+
+This is a minimal list - feel free to include more. Do NOT write code as part of your plan.
+Keep it high level. You can reference certain files or functions though.
+
+Before writing your plan, make sure to do research. Explore the relevant sections in the codebase.

CONTRIBUTING.md

@@ -1,4 +1,4 @@
-<!-- DANSWER_METADATA={"link": "https://github.com/onyx-dot-app/onyx/blob/main/CONTRIBUTING.md"} -->
+<!-- ONYX_METADATA={"link": "https://github.com/onyx-dot-app/onyx/blob/main/CONTRIBUTING.md"} -->
 
 # Contributing to Onyx
 
@@ -12,9 +12,8 @@ As an open source project in a rapidly changing space, we welcome all contributi
 
 The [GitHub Issues](https://github.com/onyx-dot-app/onyx/issues) page is a great place to start for contribution ideas.
 
-To ensure that your contribution is aligned with the project's direction, please reach out to Hagen (or any other maintainer) on the Onyx team
-via [Slack](https://join.slack.com/t/onyx-dot-app/shared_invite/zt-2twesxdr6-5iQitKZQpgq~hYIZ~dv3KA) /
-[Discord](https://discord.gg/TDJ59cGV2X) or [email](mailto:founders@onyx.app).
+To ensure that your contribution is aligned with the project's direction, please reach out to any maintainer on the Onyx team
+via [Discord](https://discord.gg/4NA5SbzrWb) or [email](mailto:hello@onyx.app).
 
 Issues that have been explicitly approved by the maintainers (aligned with the direction of the project)
 will be marked with the `approved by maintainers` label.
@@ -28,8 +27,7 @@ Your input is vital to making sure that Onyx moves in the right direction.
 Before starting on implementation, please raise a GitHub issue.
 
 Also, always feel free to message the founders (Chris Weaver / Yuhong Sun) on
-[Slack](https://join.slack.com/t/onyx-dot-app/shared_invite/zt-2twesxdr6-5iQitKZQpgq~hYIZ~dv3KA) /
-[Discord](https://discord.gg/TDJ59cGV2X) directly about anything at all.
+[Discord](https://discord.gg/4NA5SbzrWb) directly about anything at all.
 
 ### Contributing Code
 
@@ -46,9 +44,7 @@ Our goal is to make contributing as easy as possible. If you run into any issues
 That way we can help future contributors and users can avoid the same issue.
 
 We also have support channels and generally interesting discussions on our
-[Slack](https://join.slack.com/t/onyx-dot-app/shared_invite/zt-2twesxdr6-5iQitKZQpgq~hYIZ~dv3KA)
-and
-[Discord](https://discord.gg/TDJ59cGV2X).
+[Discord](https://discord.gg/4NA5SbzrWb).
 
 We would love to see you there!
 
@@ -59,6 +55,7 @@ Onyx being a fully functional app, relies on some external software, specificall
 - [Postgres](https://www.postgresql.org/) (Relational DB)
 - [Vespa](https://vespa.ai/) (Vector DB/Search Engine)
 - [Redis](https://redis.io/) (Cache)
+- [MinIO](https://min.io/) (File Store)
 - [Nginx](https://nginx.org/) (Not needed for development flows generally)
 
 > **Note:**
@@ -83,10 +80,6 @@ python -m venv .venv
 source .venv/bin/activate
 ```
 
-> **Note:**
-> This virtual environment MUST NOT be set up WITHIN the onyx directory if you plan on using mypy within certain IDEs.
-> For simplicity, we recommend setting up the virtual environment outside of the onyx directory.
-
 _For Windows, activate the virtual environment using Command Prompt:_
 
 ```bash
@@ -102,10 +95,21 @@ If using PowerShell, the command slightly differs:
 Install the required python dependencies:
 
 ```bash
-pip install -r onyx/backend/requirements/default.txt
-pip install -r onyx/backend/requirements/dev.txt
-pip install -r onyx/backend/requirements/ee.txt
-pip install -r onyx/backend/requirements/model_server.txt
+pip install -r backend/requirements/combined.txt
+```
+
+or
+
+```bash
+pip install -r backend/requirements/default.txt
+pip install -r backend/requirements/dev.txt
+pip install -r backend/requirements/ee.txt
+pip install -r backend/requirements/model_server.txt
+```
+
+Fix vscode/cursor auto-imports:
+```bash
+pip install -e .
 ```
 
 Install Playwright for Python (headless browser required by the Web Connector)
@@ -120,8 +124,15 @@ You may have to deactivate and reactivate your virtualenv for `playwright` to ap
 
 #### Frontend: Node dependencies
 
-Install [Node.js and npm](https://docs.npmjs.com/downloading-and-installing-node-js-and-npm) for the frontend.
-Once the above is done, navigate to `onyx/web` run:
+Onyx uses Node v22.20.0. We highly recommend you use [Node Version Manager (nvm)](https://github.com/nvm-sh/nvm)
+to manage your Node installations. Once installed, you can run
+
+```bash
+nvm install 22 && nvm use 22
+node -v # verify your active version
+``` 
+
+Navigate to `onyx/web` and run:
 
 ```bash
 npm i
@@ -132,8 +143,6 @@ npm i
 ### Backend
 
 For the backend, you'll need to setup pre-commit hooks (black / reorder-python-imports).
-First, install pre-commit (if you don't have it already) following the instructions
-[here](https://pre-commit.com/#installation).
 
 With the virtual environment active, install the pre-commit library with:
 
@@ -153,15 +162,17 @@ To run the mypy checks manually, run `python -m mypy .` from the `onyx/backend`
 
 ### Web
 
-We use `prettier` for formatting. The desired version (2.8.8) will be installed via a `npm i` from the `onyx/web` directory.
+We use `prettier` for formatting. The desired version will be installed via a `npm i` from the `onyx/web` directory.
 To run the formatter, use `npx prettier --write .` from the `onyx/web` directory.
-Please double check that prettier passes before creating a pull request.
+
+Pre-commit will also run prettier automatically on files you've recently touched. If re-formatted, your commit will fail.
+Re-stage your changes and commit again.
 
 # Running the application for development
 
 ## Developing using VSCode Debugger (recommended)
 
-We highly recommend using VSCode debugger for development.
+**We highly recommend using VSCode debugger for development.**
 See [CONTRIBUTING_VSCODE.md](./CONTRIBUTING_VSCODE.md) for more details.
 
 Otherwise, you can follow the instructions below to run the application for development.
@@ -171,10 +182,10 @@ Otherwise, you can follow the instructions below to run the application for deve
 
 You will need Docker installed to run these containers.
 
-First navigate to `onyx/deployment/docker_compose`, then start up Postgres/Vespa/Redis with:
+First navigate to `onyx/deployment/docker_compose`, then start up Postgres/Vespa/Redis/MinIO with:
 
 ```bash
-docker compose -f docker-compose.dev.yml -p onyx-stack up -d index relational_db cache
+docker compose up -d index relational_db cache minio
 ```
 
 (index refers to Vespa, relational_db refers to Postgres, and cache refers to Redis)
@@ -256,7 +267,7 @@ You can run the full Onyx application stack from pre-built images including all
 Navigate to `onyx/deployment/docker_compose` and run:
 
 ```bash
-docker compose -f docker-compose.dev.yml -p onyx-stack up -d
+docker compose up -d
 ```
 
 After Docker pulls and starts these containers, navigate to `http://localhost:3000` to use Onyx.
@@ -264,7 +275,7 @@ After Docker pulls and starts these containers, navigate to `http://localhost:30
 If you want to make changes to Onyx and run those changes in Docker, you can also build a local version of the Onyx container images that incorporates your changes like so:
 
 ```bash
-docker compose -f docker-compose.dev.yml -p onyx-stack up -d --build
+docker compose up -d --build
 ```
 
 

CONTRIBUTING_VSCODE.md

@@ -5,7 +5,7 @@ This guide explains how to set up and use VSCode's debugging capabilities with t
 ## Initial Setup
 
 1. **Environment Setup**:
-   - Copy `.vscode/.env.template` to `.vscode/.env`
+   - Copy `.vscode/env_template.txt` to `.vscode/.env`
    - Fill in the necessary environment variables in `.vscode/.env`
 2. **launch.json**:
    - Copy `.vscode/launch.template.jsonc` to `.vscode/launch.json`
@@ -17,10 +17,12 @@ Before starting, make sure the Docker Daemon is running.
 1. Open the Debug view in VSCode (Cmd+Shift+D on macOS)
 2. From the dropdown at the top, select "Clear and Restart External Volumes and Containers" and press the green play button
 3. From the dropdown at the top, select "Run All Onyx Services" and press the green play button
-4. CD into web, run "npm i" followed by npm run dev.
-5. Now, you can navigate to onyx in your browser (default is http://localhost:3000) and start using the app
-6. You can set breakpoints by clicking to the left of line numbers to help debug while the app is running
-7. Use the debug toolbar to step through code, inspect variables, etc.
+4. Now, you can navigate to onyx in your browser (default is http://localhost:3000) and start using the app
+5. You can set breakpoints by clicking to the left of line numbers to help debug while the app is running
+6. Use the debug toolbar to step through code, inspect variables, etc.
+
+Note: Clear and Restart External Volumes and Containers will reset your postgres and Vespa (relational-db and index).
+Only run this if you are okay with wiping your data.
 
 ## Features
 

README.md

@@ -1,117 +1,103 @@
-<!-- DANSWER_METADATA={"link": "https://github.com/onyx-dot-app/onyx/blob/main/README.md"} -->
-
 <a name="readme-top"></a>
 
 <h2 align="center">
-<a href="https://www.onyx.app/"> <img width="50%" src="https://github.com/onyx-dot-app/onyx/blob/logo/OnyxLogoCropped.jpg?raw=true)" /></a>
+    <a href="https://www.onyx.app/"> <img width="50%" src="https://github.com/onyx-dot-app/onyx/blob/logo/OnyxLogoCropped.jpg?raw=true)" /></a>
 </h2>
 
-<p align="center">
-<p align="center">Open Source Gen-AI + Enterprise Search.</p>
+<p align="center">Open Source AI Platform</p>
 
 <p align="center">
-<a href="https://docs.onyx.app/" target="_blank">
-    <img src="https://img.shields.io/badge/docs-view-blue" alt="Documentation">
-</a>
-<a href="https://join.slack.com/t/onyx-dot-app/shared_invite/zt-2twesxdr6-5iQitKZQpgq~hYIZ~dv3KA" target="_blank">
-    <img src="https://img.shields.io/badge/slack-join-blue.svg?logo=slack" alt="Slack">
-</a>
-<a href="https://discord.gg/TDJ59cGV2X" target="_blank">
-    <img src="https://img.shields.io/badge/discord-join-blue.svg?logo=discord&logoColor=white" alt="Discord">
-</a>
-<a href="https://github.com/onyx-dot-app/onyx/blob/main/README.md" target="_blank">
-    <img src="https://img.shields.io/static/v1?label=license&message=MIT&color=blue" alt="License">
-</a>
+    <a href="https://discord.gg/TDJ59cGV2X" target="_blank">
+        <img src="https://img.shields.io/badge/discord-join-blue.svg?logo=discord&logoColor=white" alt="Discord">
+    </a>
+    <a href="https://docs.onyx.app/" target="_blank">
+        <img src="https://img.shields.io/badge/docs-view-blue" alt="Documentation">
+    </a>
+    <a href="https://docs.onyx.app/" target="_blank">
+        <img src="https://img.shields.io/website?url=https://www.onyx.app&up_message=visit&up_color=blue" alt="Documentation">
+    </a>
+    <a href="https://github.com/onyx-dot-app/onyx/blob/main/LICENSE" target="_blank">
+        <img src="https://img.shields.io/static/v1?label=license&message=MIT&color=blue" alt="License">
+    </a>
 </p>
 
-<strong>[Onyx](https://www.onyx.app/)</strong> (formerly Danswer) is the AI platform connected to your company's docs, apps, and people.
-Onyx provides a feature rich Chat interface and plugs into any LLM of your choice.
-Keep knowledge and access controls sync-ed across over 40 connectors like Google Drive, Slack, Confluence, Salesforce, etc.
-Create custom AI agents with unique prompts, knowledge, and actions that the agents can take.
-Onyx can be deployed securely anywhere and for any scale - on a laptop, on-premise, or to cloud.
 
 
-<h3>Feature Highlights</h3>
+**[Onyx](https://www.onyx.app/)** is a feature-rich, self-hostable Chat UI that works with any LLM. It is easy to deploy and can run in a completely airgapped environment.
 
-**Deep research over your team's knowledge:**
+Onyx comes loaded with advanced features like Agents, Web Search, RAG, MCP, Deep Research, Connectors to 40+ knowledge sources, and more.
 
-https://private-user-images.githubusercontent.com/32520769/414509312-48392e83-95d0-4fb5-8650-a396e05e0a32.mp4?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.a9D8A0sgKE9AoaoE-mfFbJ6_OKYeqaf7TZ4Han2JfW8
+> [!TIP]
+> Run Onyx with one command (or see deployment section below):
+> ```
+> curl -fsSL https://raw.githubusercontent.com/onyx-dot-app/onyx/main/deployment/docker_compose/install.sh > install.sh && chmod +x install.sh && ./install.sh
+> ```
 
-
-**Use Onyx as a secure AI Chat with any LLM:**
+****
 
 ![Onyx Chat Silent Demo](https://github.com/onyx-dot-app/onyx/releases/download/v0.21.1/OnyxChatSilentDemo.gif)
 
 
-**Easily set up connectors to your apps:**
 
-![Onyx Connector Silent Demo](https://github.com/onyx-dot-app/onyx/releases/download/v0.21.1/OnyxConnectorSilentDemo.gif)
+## ⭐ Features
+- **🤖 Custom Agents:** Build AI Agents with unique instructions, knowledge and actions.
+- **🌍 Web Search:** Browse the web with Google PSE, Exa, and Serper as well as an in-house scraper or Firecrawl.
+- **🔍 RAG:** Best in class hybrid-search + knowledge graph for uploaded files and ingested documents from connectors. 
+- **🔄 Connectors:** Pull knowledge, metadata, and access information from over 40 applications.
+- **🔬 Deep Research:** Get in depth answers with an agentic multi-step search.
+- **▶️ Actions & MCP:** Give AI Agents the ability to interact with external systems.
+- **💻 Code Interpreter:** Execute code to analyze data, render graphs and create files.
+- **🎨 Image Generation:** Generate images based on user prompts.
+- **👥 Collaboration:** Chat sharing, feedback gathering, user management, usage analytics, and more.
+
+Onyx works with all LLMs (like OpenAI, Anthropic, Gemini, etc.) and self-hosted LLMs (like Ollama, vLLM, etc.)
+
+To learn more about the features, check out our [documentation](https://docs.onyx.app/welcome)!
 
 
-**Access Onyx where your team already works:**
 
-![Onyx Bot Demo](https://github.com/onyx-dot-app/onyx/releases/download/v0.21.1/OnyxBot.png)
+## 🚀 Deployment
+Onyx supports deployments in Docker, Kubernetes, Terraform, along with guides for major cloud providers.
+
+See guides below:
+- [Docker](https://docs.onyx.app/deployment/local/docker) or [Quickstart](https://docs.onyx.app/deployment/getting_started/quickstart) (best for most users)
+- [Kubernetes](https://docs.onyx.app/deployment/local/kubernetes) (best for large teams)
+- [Terraform](https://docs.onyx.app/deployment/local/terraform) (best for teams already using Terraform)
+- Cloud specific guides (best if specifically using [AWS EKS](https://docs.onyx.app/deployment/cloud/aws/eks), [Azure VMs](https://docs.onyx.app/deployment/cloud/azure), etc.)
+
+> [!TIP]  
+> **To try Onyx for free without deploying, check out [Onyx Cloud](https://cloud.onyx.app/signup)**.
 
 
-## Deployment
-**To try it out for free and get started in seconds, check out [Onyx Cloud](https://cloud.onyx.app/signup)**.
 
-Onyx can also be run locally (even on a laptop) or deployed on a virtual machine with a single
-`docker compose` command. Checkout our [docs](https://docs.onyx.app/quickstart) to learn more.
+## 🔍 Other Notable Benefits
+Onyx is built for teams of all sizes, from individual users to the largest global enterprises.
 
-We also have built-in support for high-availability/scalable deployment on Kubernetes.
-References [here](https://github.com/onyx-dot-app/onyx/tree/main/deployment).
+- **Enterprise Search**: far more than simple RAG, Onyx has custom indexing and retrieval that remains performant and accurate for scales of up to tens of millions of documents.
+- **Security**: SSO (OIDC/SAML/OAuth2), RBAC, encryption of credentials, etc.
+- **Management UI**: different user roles such as basic, curator, and admin.
+- **Document Permissioning**: mirrors user access from external apps for RAG use cases.
 
 
-## 🔍 Other Notable Benefits of Onyx
-- Custom deep learning models for indexing and inference time, only through Onyx + learning from user feedback.
-- Flexible security features like SSO (OIDC/SAML/OAuth2), RBAC, encryption of credentials, etc.
-- Knowledge curation features like document-sets, query history, usage analytics, etc.
-- Scalable deployment options tested up to many tens of thousands users and hundreds of millions of documents.
-
 
 ## 🚧 Roadmap
-- New methods in information retrieval (StructRAG, LightGraphRAG, etc.)
-- Personalized Search
-- Organizational understanding and ability to locate and suggest experts from your team.
-- Code Search
-- SQL and Structured Query Language
+To see ongoing and upcoming projects, check out our [roadmap](https://github.com/orgs/onyx-dot-app/projects/2)!
 
 
-## 🔌 Connectors
-Keep knowledge and access up to sync across 40+ connectors:
-
-- Google Drive
-- Confluence
-- Slack
-- Gmail
-- Salesforce
-- Microsoft Sharepoint
-- Github
-- Jira
-- Zendesk
-- Gong
-- Microsoft Teams
-- Dropbox
-- Local Files
-- Websites
-- And more ...
-
-See the full list [here](https://docs.onyx.app/connectors).
-
 
 ## 📚 Licensing
 There are two editions of Onyx:
 
-- Onyx Community Edition (CE) is available freely under the MIT Expat license. Simply follow the Deployment guide above.
+- Onyx Community Edition (CE) is available freely under the MIT license.
 - Onyx Enterprise Edition (EE) includes extra features that are primarily useful for larger organizations.
 For feature details, check out [our website](https://www.onyx.app/pricing).
 
-To try the Onyx Enterprise Edition:
-1. Checkout [Onyx Cloud](https://cloud.onyx.app/signup).
-2. For self-hosting the Enterprise Edition, contact us at [founders@onyx.app](mailto:founders@onyx.app) or book a call with us on our [Cal](https://cal.com/team/onyx/founders).
+
+
+## 👪 Community
+Join our open source community on **[Discord](https://discord.gg/TDJ59cGV2X)**!
+
 
 
 ## 💡 Contributing
 Looking to contribute? Please check out the [Contribution Guide](CONTRIBUTING.md) for more details.
-

backend/.gitignore

@@ -9,4 +9,6 @@ api_keys.py
 vespa-app.zip
 dynamic_config_storage/
 celerybeat-schedule*
-onyx/connectors/salesforce/data/
+onyx/connectors/salesforce/data/
+.test.env
+/generated

backend/Dockerfile

@@ -12,10 +12,11 @@ ARG ONYX_VERSION=0.0.0-dev
 # DO_NOT_TRACK is used to disable telemetry for Unstructured
 ENV ONYX_VERSION=${ONYX_VERSION} \
     DANSWER_RUNNING_IN_DOCKER="true" \
-    DO_NOT_TRACK="true"
+    DO_NOT_TRACK="true" \
+    PLAYWRIGHT_BROWSERS_PATH="/app/.cache/ms-playwright"
 
+COPY --from=ghcr.io/astral-sh/uv:latest /uv /uvx /bin/
 
-RUN echo "ONYX_VERSION: ${ONYX_VERSION}"
 # Install system dependencies
 # cmake needed for psycopg (postgres)
 # libpq-dev needed for psycopg (postgres)
@@ -47,22 +48,19 @@ RUN apt-get update && \
 # Remove py which is pulled in by retry, py is not needed and is a CVE
 COPY ./requirements/default.txt /tmp/requirements.txt
 COPY ./requirements/ee.txt /tmp/ee-requirements.txt
-RUN pip install --no-cache-dir --upgrade \
-        --retries 5 \
-        --timeout 30 \
+RUN uv pip install --system --no-cache-dir --upgrade \
         -r /tmp/requirements.txt \
         -r /tmp/ee-requirements.txt && \
     pip uninstall -y py && \
     playwright install chromium && \
     playwright install-deps chromium && \
-    ln -s /usr/local/bin/supervisord /usr/bin/supervisord
-
-# Cleanup for CVEs and size reduction
-# https://github.com/tornadoweb/tornado/issues/3107
-# xserver-common and xvfb included by playwright installation but not needed after
-# perl-base is part of the base Python Debian image but not needed for Onyx functionality
-# perl-base could only be removed with --allow-remove-essential
-RUN apt-get update && \
+    ln -s /usr/local/bin/supervisord /usr/bin/supervisord && \
+    # Cleanup for CVEs and size reduction
+    # https://github.com/tornadoweb/tornado/issues/3107
+    # xserver-common and xvfb included by playwright installation but not needed after
+    # perl-base is part of the base Python Debian image but not needed for Onyx functionality
+    # perl-base could only be removed with --allow-remove-essential
+    apt-get update && \
     apt-get remove -y --allow-remove-essential \
         perl-base \
         xserver-common \
@@ -72,12 +70,16 @@ RUN apt-get update && \
         libxmlsec1-dev \
         pkg-config \
         gcc && \
-    apt-get install -y libxmlsec1-openssl && \
+    # Install here to avoid some packages being cleaned up above
+    apt-get install -y \
+        libxmlsec1-openssl \
+        # Install postgresql-client for easy manual tests
+        postgresql-client && \
     apt-get autoremove -y && \
     rm -rf /var/lib/apt/lists/* && \
+    rm -rf ~/.cache/uv /tmp/*.txt && \
     rm -f /usr/local/lib/python3.11/site-packages/tornado/test/test.key
 
-
 # Pre-downloading models for setups with limited egress
 RUN python -c "from tokenizers import Tokenizer; \
 Tokenizer.from_pretrained('nomic-ai/nomic-embed-text-v1')"
@@ -85,31 +87,40 @@ Tokenizer.from_pretrained('nomic-ai/nomic-embed-text-v1')"
 # Pre-downloading NLTK for setups with limited egress
 RUN python -c "import nltk; \
 nltk.download('stopwords', quiet=True); \
-nltk.download('punkt', quiet=True);"
+nltk.download('punkt_tab', quiet=True);"
 # nltk.download('wordnet', quiet=True); introduce this back if lemmatization is needed
 
 # Set up application files
 WORKDIR /app
 
+# Create non-root user for security best practices
+RUN groupadd -g 1001 onyx && \
+    useradd -u 1001 -g onyx -m -s /bin/bash onyx && \
+    mkdir -p /var/log/onyx && \
+    chmod 755 /var/log/onyx && \
+    chown onyx:onyx /var/log/onyx
+
 # Enterprise Version Files
-COPY ./ee /app/ee
+COPY --chown=onyx:onyx ./ee /app/ee
 COPY supervisord.conf /etc/supervisor/conf.d/supervisord.conf
 
 # Set up application files
-COPY ./onyx /app/onyx
-COPY ./shared_configs /app/shared_configs
-COPY ./alembic /app/alembic
-COPY ./alembic_tenants /app/alembic_tenants
-COPY ./alembic.ini /app/alembic.ini
+COPY --chown=onyx:onyx ./onyx /app/onyx
+COPY --chown=onyx:onyx ./shared_configs /app/shared_configs
+COPY --chown=onyx:onyx ./alembic /app/alembic
+COPY --chown=onyx:onyx ./alembic_tenants /app/alembic_tenants
+COPY --chown=onyx:onyx ./alembic.ini /app/alembic.ini
 COPY supervisord.conf /usr/etc/supervisord.conf
-COPY ./static /app/static
+COPY --chown=onyx:onyx ./static /app/static
 
 # Escape hatch scripts
-COPY ./scripts/debugging /app/scripts/debugging
-COPY ./scripts/force_delete_connector_by_id.py /app/scripts/force_delete_connector_by_id.py
+COPY --chown=onyx:onyx ./scripts/debugging /app/scripts/debugging
+COPY --chown=onyx:onyx ./scripts/force_delete_connector_by_id.py /app/scripts/force_delete_connector_by_id.py
+COPY --chown=onyx:onyx ./scripts/supervisord_entrypoint.sh /app/scripts/supervisord_entrypoint.sh
+RUN chmod +x /app/scripts/supervisord_entrypoint.sh
 
 # Put logo in assets
-COPY ./assets /app/assets
+COPY --chown=onyx:onyx ./assets /app/assets
 
 ENV PYTHONPATH=/app
 

backend/Dockerfile.model_server

@@ -9,19 +9,42 @@ visit https://github.com/onyx-dot-app/onyx."
 # Default ONYX_VERSION, typically overriden during builds by GitHub Actions.
 ARG ONYX_VERSION=0.0.0-dev
 ENV ONYX_VERSION=${ONYX_VERSION} \
-    DANSWER_RUNNING_IN_DOCKER="true"
+    DANSWER_RUNNING_IN_DOCKER="true" \
+    HF_HOME=/app/.cache/huggingface
 
+COPY --from=ghcr.io/astral-sh/uv:latest /uv /uvx /bin/
 
-RUN echo "ONYX_VERSION: ${ONYX_VERSION}"
+# Create non-root user for security best practices
+RUN mkdir -p /app && \
+    groupadd -g 1001 onyx && \
+    useradd -u 1001 -g onyx -m -s /bin/bash onyx  && \
+    chown -R onyx:onyx /app && \
+    mkdir -p /var/log/onyx && \
+    chmod 755 /var/log/onyx && \
+    chown onyx:onyx /var/log/onyx
+
+# --- add toolchain needed for Rust/Python builds (fastuuid) ---
+ENV RUSTUP_HOME=/usr/local/rustup \
+    CARGO_HOME=/usr/local/cargo \
+    PATH=/usr/local/cargo/bin:$PATH
+
+RUN set -eux; \
+    apt-get update && apt-get install -y --no-install-recommends \
+        build-essential \
+        pkg-config \
+        curl \
+        ca-certificates \
+    # Install latest stable Rust (supports Cargo.lock v4)
+    && curl -sSf https://sh.rustup.rs | sh -s -- -y --profile minimal --default-toolchain stable \
+    && rustc --version && cargo --version \
+    && apt-get remove -y --allow-remove-essential perl-base \
+    && apt-get autoremove -y \
+    && rm -rf /var/lib/apt/lists/*
 
 COPY ./requirements/model_server.txt /tmp/requirements.txt
-RUN pip install --no-cache-dir --upgrade \
-        --retries 5 \
-        --timeout 30 \
-        -r /tmp/requirements.txt
-
-RUN apt-get remove -y --allow-remove-essential perl-base && \ 
-    apt-get autoremove -y
+RUN uv pip install --system --no-cache-dir --upgrade \
+        -r /tmp/requirements.txt && \
+    rm -rf ~/.cache/uv /tmp/*.txt
 
 # Pre-downloading models for setups with limited egress
 # Download tokenizers, distilbert for the Onyx model
@@ -36,16 +59,18 @@ snapshot_download(repo_id='onyx-dot-app/information-content-model'); \
 snapshot_download('nomic-ai/nomic-embed-text-v1'); \
 snapshot_download('mixedbread-ai/mxbai-rerank-xsmall-v1'); \
 from sentence_transformers import SentenceTransformer; \
-SentenceTransformer(model_name_or_path='nomic-ai/nomic-embed-text-v1', trust_remote_code=True);"
-
-# In case the user has volumes mounted to /root/.cache/huggingface that they've downloaded while
-# running Onyx, don't overwrite it with the built in cache folder
-RUN mv /root/.cache/huggingface /root/.cache/temp_huggingface
+SentenceTransformer(model_name_or_path='nomic-ai/nomic-embed-text-v1', trust_remote_code=True);" && \
+    # In case the user has volumes mounted to /app/.cache/huggingface that they've downloaded while
+    # running Onyx, move the current contents of the cache folder to a temporary location to ensure
+    # it's preserved in order to combine with the user's cache contents
+    mv /app/.cache/huggingface /app/.cache/temp_huggingface && \
+    chown -R onyx:onyx /app
 
 WORKDIR /app
 
 # Utils used by model server
 COPY ./onyx/utils/logger.py /app/onyx/utils/logger.py
+COPY ./onyx/utils/middleware.py /app/onyx/utils/middleware.py
 
 # Place to fetch version information
 COPY ./onyx/__init__.py /app/onyx/__init__.py

backend/alembic/README.md

@@ -1,4 +1,4 @@
-<!-- DANSWER_METADATA={"link": "https://github.com/onyx-dot-app/onyx/blob/main/backend/alembic/README.md"} -->
+<!-- ONYX_METADATA={"link": "https://github.com/onyx-dot-app/onyx/blob/main/backend/alembic/README.md"} -->
 
 # Alembic DB Migrations
 
@@ -20,3 +20,44 @@ To run all un-applied migrations:
 To undo migrations:
 `alembic downgrade -X`
 where X is the number of migrations you want to undo from the current state
+
+### Multi-tenant migrations
+
+For multi-tenant deployments, you can use additional options:
+
+**Upgrade all tenants:**
+```bash
+alembic -x upgrade_all_tenants=true upgrade head
+```
+
+**Upgrade specific schemas:**
+```bash
+# Single schema
+alembic -x schemas=tenant_12345678-1234-1234-1234-123456789012 upgrade head
+
+# Multiple schemas (comma-separated)
+alembic -x schemas=tenant_12345678-1234-1234-1234-123456789012,public,another_tenant upgrade head
+```
+
+**Upgrade tenants within an alphabetical range:**
+```bash
+# Upgrade tenants 100-200 when sorted alphabetically (positions 100 to 200)
+alembic -x upgrade_all_tenants=true -x tenant_range_start=100 -x tenant_range_end=200 upgrade head
+
+# Upgrade tenants starting from position 1000 alphabetically
+alembic -x upgrade_all_tenants=true -x tenant_range_start=1000 upgrade head
+
+# Upgrade first 500 tenants alphabetically
+alembic -x upgrade_all_tenants=true -x tenant_range_end=500 upgrade head
+```
+
+**Continue on error (for batch operations):**
+```bash
+alembic -x upgrade_all_tenants=true -x continue=true upgrade head
+```
+
+The tenant range filtering works by:
+1. Sorting tenant IDs alphabetically
+2. Using 1-based position numbers (1st, 2nd, 3rd tenant, etc.)
+3. Filtering to the specified range of positions
+4. Non-tenant schemas (like 'public') are always included

backend/alembic/env.py

@@ -1,12 +1,12 @@
 from typing import Any, Literal
-from onyx.db.engine import get_iam_auth_token
+from onyx.db.engine.iam_auth import get_iam_auth_token
 from onyx.configs.app_configs import USE_IAM_AUTH
 from onyx.configs.app_configs import POSTGRES_HOST
 from onyx.configs.app_configs import POSTGRES_PORT
 from onyx.configs.app_configs import POSTGRES_USER
 from onyx.configs.app_configs import AWS_REGION_NAME
-from onyx.db.engine import build_connection_string
-from onyx.db.engine import get_all_tenant_ids
+from onyx.db.engine.sql_engine import build_connection_string
+from onyx.db.engine.tenant_utils import get_all_tenant_ids
 from sqlalchemy import event
 from sqlalchemy import pool
 from sqlalchemy import text
@@ -21,9 +21,14 @@ from alembic import context
 from sqlalchemy.ext.asyncio import create_async_engine
 from sqlalchemy.sql.schema import SchemaItem
 from onyx.configs.constants import SSL_CERT_FILE
-from shared_configs.configs import MULTI_TENANT, POSTGRES_DEFAULT_SCHEMA
+from shared_configs.configs import (
+    MULTI_TENANT,
+    POSTGRES_DEFAULT_SCHEMA,
+    TENANT_ID_PREFIX,
+)
 from onyx.db.models import Base
 from celery.backends.database.session import ResultModelBase  # type: ignore
+from onyx.db.engine.sql_engine import SqlEngine
 
 # Make sure in alembic.ini [logger_root] level=INFO is set or most logging will be
 # hidden! (defaults to level=WARN)
@@ -68,15 +73,67 @@ def include_object(
     return True
 
 
-def get_schema_options() -> tuple[str, bool, bool, bool]:
+def filter_tenants_by_range(
+    tenant_ids: list[str], start_range: int | None = None, end_range: int | None = None
+) -> list[str]:
+    """
+    Filter tenant IDs by alphabetical position range.
+
+    Args:
+        tenant_ids: List of tenant IDs to filter
+        start_range: Starting position in alphabetically sorted list (1-based, inclusive)
+        end_range: Ending position in alphabetically sorted list (1-based, inclusive)
+
+    Returns:
+        Filtered list of tenant IDs in their original order
+    """
+    if start_range is None and end_range is None:
+        return tenant_ids
+
+    # Separate tenant IDs from non-tenant schemas
+    tenant_schemas = [tid for tid in tenant_ids if tid.startswith(TENANT_ID_PREFIX)]
+    non_tenant_schemas = [
+        tid for tid in tenant_ids if not tid.startswith(TENANT_ID_PREFIX)
+    ]
+
+    # Sort tenant schemas alphabetically.
+    # NOTE: can cause missed schemas if a schema is created in between workers
+    # fetching of all tenant IDs. We accept this risk for now. Just re-running
+    # the migration will fix the issue.
+    sorted_tenant_schemas = sorted(tenant_schemas)
+
+    # Apply range filtering (0-based indexing)
+    start_idx = start_range if start_range is not None else 0
+    end_idx = end_range if end_range is not None else len(sorted_tenant_schemas)
+
+    # Ensure indices are within bounds
+    start_idx = max(0, start_idx)
+    end_idx = min(len(sorted_tenant_schemas), end_idx)
+
+    # Get the filtered tenant schemas
+    filtered_tenant_schemas = sorted_tenant_schemas[start_idx:end_idx]
+
+    # Combine with non-tenant schemas and preserve original order
+    filtered_tenants = []
+    for tenant_id in tenant_ids:
+        if tenant_id in filtered_tenant_schemas or tenant_id in non_tenant_schemas:
+            filtered_tenants.append(tenant_id)
+
+    return filtered_tenants
+
+
+def get_schema_options() -> (
+    tuple[bool, bool, bool, int | None, int | None, list[str] | None]
+):
     x_args_raw = context.get_x_argument()
     x_args = {}
     for arg in x_args_raw:
-        for pair in arg.split(","):
-            if "=" in pair:
-                key, value = pair.split("=", 1)
-                x_args[key.strip()] = value.strip()
-    schema_name = x_args.get("schema", POSTGRES_DEFAULT_SCHEMA)
+        if "=" in arg:
+            key, value = arg.split("=", 1)
+            x_args[key.strip()] = value.strip()
+        else:
+            raise ValueError(f"Invalid argument: {arg}")
+
     create_schema = x_args.get("create_schema", "true").lower() == "true"
     upgrade_all_tenants = x_args.get("upgrade_all_tenants", "false").lower() == "true"
 
@@ -84,17 +141,81 @@ def get_schema_options() -> tuple[str, bool, bool, bool]:
     # only applies to online migrations
     continue_on_error = x_args.get("continue", "false").lower() == "true"
 
-    if (
-        MULTI_TENANT
-        and schema_name == POSTGRES_DEFAULT_SCHEMA
-        and not upgrade_all_tenants
-    ):
+    # Tenant range filtering
+    tenant_range_start = None
+    tenant_range_end = None
+
+    if "tenant_range_start" in x_args:
+        try:
+            tenant_range_start = int(x_args["tenant_range_start"])
+        except ValueError:
+            raise ValueError(
+                f"Invalid tenant_range_start value: {x_args['tenant_range_start']}. Must be an integer."
+            )
+
+    if "tenant_range_end" in x_args:
+        try:
+            tenant_range_end = int(x_args["tenant_range_end"])
+        except ValueError:
+            raise ValueError(
+                f"Invalid tenant_range_end value: {x_args['tenant_range_end']}. Must be an integer."
+            )
+
+    # Validate range
+    if tenant_range_start is not None and tenant_range_end is not None:
+        if tenant_range_start > tenant_range_end:
+            raise ValueError(
+                f"tenant_range_start ({tenant_range_start}) cannot be greater than tenant_range_end ({tenant_range_end})"
+            )
+
+    # Specific schema names filtering (replaces both schema_name and the old tenant_ids approach)
+    schemas = None
+    if "schemas" in x_args:
+        schema_names_str = x_args["schemas"].strip()
+        if schema_names_str:
+            # Split by comma and strip whitespace
+            schemas = [
+                name.strip() for name in schema_names_str.split(",") if name.strip()
+            ]
+            if schemas:
+                logger.info(f"Specific schema names specified: {schemas}")
+
+    # Validate that only one method is used at a time
+    range_filtering = tenant_range_start is not None or tenant_range_end is not None
+    specific_filtering = schemas is not None and len(schemas) > 0
+
+    if range_filtering and specific_filtering:
         raise ValueError(
-            "Cannot run default migrations in public schema when multi-tenancy is enabled. "
-            "Please specify a tenant-specific schema."
+            "Cannot use both tenant range filtering (tenant_range_start/tenant_range_end) "
+            "and specific schema filtering (schemas) at the same time. "
+            "Please use only one filtering method."
         )
 
-    return schema_name, create_schema, upgrade_all_tenants, continue_on_error
+    if upgrade_all_tenants and specific_filtering:
+        raise ValueError(
+            "Cannot use both upgrade_all_tenants=true and schemas at the same time. "
+            "Use either upgrade_all_tenants=true for all tenants, or schemas for specific schemas."
+        )
+
+    # If any filtering parameters are specified, we're not doing the default single schema migration
+    if range_filtering:
+        upgrade_all_tenants = True
+
+    # Validate multi-tenant requirements
+    if MULTI_TENANT and not upgrade_all_tenants and not specific_filtering:
+        raise ValueError(
+            "In multi-tenant mode, you must specify either upgrade_all_tenants=true "
+            "or provide schemas. Cannot run default migration."
+        )
+
+    return (
+        create_schema,
+        upgrade_all_tenants,
+        continue_on_error,
+        tenant_range_start,
+        tenant_range_end,
+        schemas,
+    )
 
 
 def do_run_migrations(
@@ -141,12 +262,20 @@ def provide_iam_token_for_alembic(
 
 async def run_async_migrations() -> None:
     (
-        schema_name,
         create_schema,
         upgrade_all_tenants,
         continue_on_error,
+        tenant_range_start,
+        tenant_range_end,
+        schemas,
     ) = get_schema_options()
 
+    if not schemas and not MULTI_TENANT:
+        schemas = [POSTGRES_DEFAULT_SCHEMA]
+
+    # without init_engine, subsequent engine calls fail hard intentionally
+    SqlEngine.init_engine(pool_size=20, max_overflow=5)
+
     engine = create_async_engine(
         build_connection_string(),
         poolclass=pool.NullPool,
@@ -160,12 +289,50 @@ async def run_async_migrations() -> None:
         ) -> None:
             provide_iam_token_for_alembic(dialect, conn_rec, cargs, cparams)
 
-    if upgrade_all_tenants:
+    if schemas:
+        # Use specific schema names directly without fetching all tenants
+        logger.info(f"Migrating specific schema names: {schemas}")
+
+        i_schema = 0
+        num_schemas = len(schemas)
+        for schema in schemas:
+            i_schema += 1
+            logger.info(
+                f"Migrating schema: index={i_schema} num_schemas={num_schemas} schema={schema}"
+            )
+            try:
+                async with engine.connect() as connection:
+                    await connection.run_sync(
+                        do_run_migrations,
+                        schema_name=schema,
+                        create_schema=create_schema,
+                    )
+            except Exception as e:
+                logger.error(f"Error migrating schema {schema}: {e}")
+                if not continue_on_error:
+                    logger.error("--continue=true is not set, raising exception!")
+                    raise
+
+                logger.warning("--continue=true is set, continuing to next schema.")
+
+    elif upgrade_all_tenants:
         tenant_schemas = get_all_tenant_ids()
 
+        filtered_tenant_schemas = filter_tenants_by_range(
+            tenant_schemas, tenant_range_start, tenant_range_end
+        )
+
+        if tenant_range_start is not None or tenant_range_end is not None:
+            logger.info(
+                f"Filtering tenants by range: start={tenant_range_start}, end={tenant_range_end}"
+            )
+            logger.info(
+                f"Total tenants: {len(tenant_schemas)}, Filtered tenants: {len(filtered_tenant_schemas)}"
+            )
+
         i_tenant = 0
-        num_tenants = len(tenant_schemas)
-        for schema in tenant_schemas:
+        num_tenants = len(filtered_tenant_schemas)
+        for schema in filtered_tenant_schemas:
             i_tenant += 1
             logger.info(
                 f"Migrating schema: index={i_tenant} num_tenants={num_tenants} schema={schema}"
@@ -180,36 +347,70 @@ async def run_async_migrations() -> None:
             except Exception as e:
                 logger.error(f"Error migrating schema {schema}: {e}")
                 if not continue_on_error:
-                    logger.error("--continue is not set, raising exception!")
+                    logger.error("--continue=true is not set, raising exception!")
                     raise
 
-                logger.warning("--continue is set, continuing to next schema.")
+                logger.warning("--continue=true is set, continuing to next schema.")
 
     else:
-        try:
-            logger.info(f"Migrating schema: {schema_name}")
-            async with engine.connect() as connection:
-                await connection.run_sync(
-                    do_run_migrations,
-                    schema_name=schema_name,
-                    create_schema=create_schema,
-                )
-        except Exception as e:
-            logger.error(f"Error migrating schema {schema_name}: {e}")
-            raise
+        # This should not happen in the new design since we require either
+        # upgrade_all_tenants=true or schemas in multi-tenant mode
+        # and for non-multi-tenant mode, we should use schemas with the default schema
+        raise ValueError(
+            "No migration target specified. Use either upgrade_all_tenants=true for all tenants "
+            "or schemas for specific schemas."
+        )
 
     await engine.dispose()
 
 
 def run_migrations_offline() -> None:
-    """This doesn't really get used when we migrate in the cloud."""
+    """
+    NOTE(rkuo): This generates a sql script that can be used to migrate the database ...
+    instead of migrating the db live via an open connection
+
+    Not clear on when this would be used by us or if it even works.
+
+    If it is offline, then why are there calls to the db engine?
+
+    This doesn't really get used when we migrate in the cloud."""
 
     logger.info("run_migrations_offline starting.")
 
-    schema_name, _, upgrade_all_tenants, continue_on_error = get_schema_options()
+    # without init_engine, subsequent engine calls fail hard intentionally
+    SqlEngine.init_engine(pool_size=20, max_overflow=5)
+
+    (
+        create_schema,
+        upgrade_all_tenants,
+        continue_on_error,
+        tenant_range_start,
+        tenant_range_end,
+        schemas,
+    ) = get_schema_options()
     url = build_connection_string()
 
-    if upgrade_all_tenants:
+    if schemas:
+        # Use specific schema names directly without fetching all tenants
+        logger.info(f"Migrating specific schema names: {schemas}")
+
+        for schema in schemas:
+            logger.info(f"Migrating schema: {schema}")
+            context.configure(
+                url=url,
+                target_metadata=target_metadata,  # type: ignore
+                literal_binds=True,
+                include_object=include_object,
+                version_table_schema=schema,
+                include_schemas=True,
+                script_location=config.get_main_option("script_location"),
+                dialect_opts={"paramstyle": "named"},
+            )
+
+            with context.begin_transaction():
+                context.run_migrations()
+
+    elif upgrade_all_tenants:
         engine = create_async_engine(url)
 
         if USE_IAM_AUTH:
@@ -223,7 +424,19 @@ def run_migrations_offline() -> None:
         tenant_schemas = get_all_tenant_ids()
         engine.sync_engine.dispose()
 
-        for schema in tenant_schemas:
+        filtered_tenant_schemas = filter_tenants_by_range(
+            tenant_schemas, tenant_range_start, tenant_range_end
+        )
+
+        if tenant_range_start is not None or tenant_range_end is not None:
+            logger.info(
+                f"Filtering tenants by range: start={tenant_range_start}, end={tenant_range_end}"
+            )
+            logger.info(
+                f"Total tenants: {len(tenant_schemas)}, Filtered tenants: {len(filtered_tenant_schemas)}"
+            )
+
+        for schema in filtered_tenant_schemas:
             logger.info(f"Migrating schema: {schema}")
             context.configure(
                 url=url,
@@ -239,21 +452,12 @@ def run_migrations_offline() -> None:
             with context.begin_transaction():
                 context.run_migrations()
     else:
-        logger.info(f"Migrating schema: {schema_name}")
-        context.configure(
-            url=url,
-            target_metadata=target_metadata,  # type: ignore
-            literal_binds=True,
-            include_object=include_object,
-            version_table_schema=schema_name,
-            include_schemas=True,
-            script_location=config.get_main_option("script_location"),
-            dialect_opts={"paramstyle": "named"},
+        # This should not happen in the new design
+        raise ValueError(
+            "No migration target specified. Use either upgrade_all_tenants=true for all tenants "
+            "or schemas for specific schemas."
         )
 
-        with context.begin_transaction():
-            context.run_migrations()
-
 
 def run_migrations_online() -> None:
     logger.info("run_migrations_online starting.")

backend/alembic/versions/027381bce97c_add_shortcut_option_for_users.py

@@ -5,6 +5,7 @@ Revises: 6fc7886d665d
 Create Date: 2025-01-14 12:14:00.814390
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 

backend/alembic/versions/03bf8be6b53a_rework_kg_config.py

@@ -0,0 +1,121 @@
+"""rework-kg-config
+
+Revision ID: 03bf8be6b53a
+Revises: 65bc6e0f8500
+Create Date: 2025-06-16 10:52:34.815335
+
+"""
+
+import json
+
+
+from datetime import datetime
+from datetime import timedelta
+from sqlalchemy.dialects import postgresql
+from sqlalchemy import text
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "03bf8be6b53a"
+down_revision = "65bc6e0f8500"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    # get current config
+    current_configs = (
+        op.get_bind()
+        .execute(text("SELECT kg_variable_name, kg_variable_values FROM kg_config"))
+        .all()
+    )
+    current_config_dict = {
+        config.kg_variable_name: (
+            config.kg_variable_values[0]
+            if config.kg_variable_name
+            not in ("KG_VENDOR_DOMAINS", "KG_IGNORE_EMAIL_DOMAINS")
+            else config.kg_variable_values
+        )
+        for config in current_configs
+        if config.kg_variable_values
+    }
+
+    # not using the KGConfigSettings model here in case it changes in the future
+    kg_config_settings = json.dumps(
+        {
+            "KG_EXPOSED": current_config_dict.get("KG_EXPOSED", False),
+            "KG_ENABLED": current_config_dict.get("KG_ENABLED", False),
+            "KG_VENDOR": current_config_dict.get("KG_VENDOR", None),
+            "KG_VENDOR_DOMAINS": current_config_dict.get("KG_VENDOR_DOMAINS", []),
+            "KG_IGNORE_EMAIL_DOMAINS": current_config_dict.get(
+                "KG_IGNORE_EMAIL_DOMAINS", []
+            ),
+            "KG_COVERAGE_START": current_config_dict.get(
+                "KG_COVERAGE_START",
+                (datetime.now() - timedelta(days=90)).strftime("%Y-%m-%d"),
+            ),
+            "KG_MAX_COVERAGE_DAYS": current_config_dict.get("KG_MAX_COVERAGE_DAYS", 90),
+            "KG_MAX_PARENT_RECURSION_DEPTH": current_config_dict.get(
+                "KG_MAX_PARENT_RECURSION_DEPTH", 2
+            ),
+            "KG_BETA_PERSONA_ID": current_config_dict.get("KG_BETA_PERSONA_ID", None),
+        }
+    )
+    op.execute(
+        f"INSERT INTO key_value_store (key, value) VALUES ('kg_config', '{kg_config_settings}')"
+    )
+
+    # drop kg config table
+    op.drop_table("kg_config")
+
+
+def downgrade() -> None:
+    # get current config
+    current_config_dict = {
+        "KG_EXPOSED": False,
+        "KG_ENABLED": False,
+        "KG_VENDOR": [],
+        "KG_VENDOR_DOMAINS": [],
+        "KG_IGNORE_EMAIL_DOMAINS": [],
+        "KG_COVERAGE_START": (datetime.now() - timedelta(days=90)).strftime("%Y-%m-%d"),
+        "KG_MAX_COVERAGE_DAYS": 90,
+        "KG_MAX_PARENT_RECURSION_DEPTH": 2,
+    }
+    current_configs = (
+        op.get_bind()
+        .execute(text("SELECT value FROM key_value_store WHERE key = 'kg_config'"))
+        .one_or_none()
+    )
+    if current_configs is not None:
+        current_config_dict.update(current_configs[0])
+    insert_values = [
+        {
+            "kg_variable_name": name,
+            "kg_variable_values": (
+                [str(val).lower() if isinstance(val, bool) else str(val)]
+                if not isinstance(val, list)
+                else val
+            ),
+        }
+        for name, val in current_config_dict.items()
+    ]
+
+    op.create_table(
+        "kg_config",
+        sa.Column("id", sa.Integer(), primary_key=True, nullable=False, index=True),
+        sa.Column("kg_variable_name", sa.String(), nullable=False, index=True),
+        sa.Column("kg_variable_values", postgresql.ARRAY(sa.String()), nullable=False),
+        sa.UniqueConstraint("kg_variable_name", name="uq_kg_config_variable_name"),
+    )
+    op.bulk_insert(
+        sa.table(
+            "kg_config",
+            sa.column("kg_variable_name", sa.String),
+            sa.column("kg_variable_values", postgresql.ARRAY(sa.String)),
+        ),
+        insert_values,
+    )
+
+    op.execute("DELETE FROM key_value_store WHERE key = 'kg_config'")

backend/alembic/versions/03d710ccf29c_add_permission_sync_attempt_tables.py

@@ -0,0 +1,153 @@
+"""add permission sync attempt tables
+
+Revision ID: 03d710ccf29c
+Revises: 96a5702df6aa
+Create Date: 2025-09-11 13:30:00.000000
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "03d710ccf29c"  # Generate a new unique ID
+down_revision = "96a5702df6aa"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    # Create the permission sync status enum
+    permission_sync_status_enum = sa.Enum(
+        "not_started",
+        "in_progress",
+        "success",
+        "canceled",
+        "failed",
+        "completed_with_errors",
+        name="permissionsyncstatus",
+        native_enum=False,
+    )
+
+    # Create doc_permission_sync_attempt table
+    op.create_table(
+        "doc_permission_sync_attempt",
+        sa.Column("id", sa.Integer(), nullable=False),
+        sa.Column("connector_credential_pair_id", sa.Integer(), nullable=False),
+        sa.Column("status", permission_sync_status_enum, nullable=False),
+        sa.Column("total_docs_synced", sa.Integer(), nullable=True),
+        sa.Column("docs_with_permission_errors", sa.Integer(), nullable=True),
+        sa.Column("error_message", sa.Text(), nullable=True),
+        sa.Column(
+            "time_created",
+            sa.DateTime(timezone=True),
+            server_default=sa.text("now()"),
+            nullable=False,
+        ),
+        sa.Column("time_started", sa.DateTime(timezone=True), nullable=True),
+        sa.Column("time_finished", sa.DateTime(timezone=True), nullable=True),
+        sa.ForeignKeyConstraint(
+            ["connector_credential_pair_id"],
+            ["connector_credential_pair.id"],
+        ),
+        sa.PrimaryKeyConstraint("id"),
+    )
+
+    # Create indexes for doc_permission_sync_attempt
+    op.create_index(
+        "ix_doc_permission_sync_attempt_time_created",
+        "doc_permission_sync_attempt",
+        ["time_created"],
+        unique=False,
+    )
+    op.create_index(
+        "ix_permission_sync_attempt_latest_for_cc_pair",
+        "doc_permission_sync_attempt",
+        ["connector_credential_pair_id", "time_created"],
+        unique=False,
+    )
+    op.create_index(
+        "ix_permission_sync_attempt_status_time",
+        "doc_permission_sync_attempt",
+        ["status", sa.text("time_finished DESC")],
+        unique=False,
+    )
+
+    # Create external_group_permission_sync_attempt table
+    # connector_credential_pair_id is nullable - group syncs can be global (e.g., Confluence)
+    op.create_table(
+        "external_group_permission_sync_attempt",
+        sa.Column("id", sa.Integer(), nullable=False),
+        sa.Column("connector_credential_pair_id", sa.Integer(), nullable=True),
+        sa.Column("status", permission_sync_status_enum, nullable=False),
+        sa.Column("total_users_processed", sa.Integer(), nullable=True),
+        sa.Column("total_groups_processed", sa.Integer(), nullable=True),
+        sa.Column("total_group_memberships_synced", sa.Integer(), nullable=True),
+        sa.Column("error_message", sa.Text(), nullable=True),
+        sa.Column(
+            "time_created",
+            sa.DateTime(timezone=True),
+            server_default=sa.text("now()"),
+            nullable=False,
+        ),
+        sa.Column("time_started", sa.DateTime(timezone=True), nullable=True),
+        sa.Column("time_finished", sa.DateTime(timezone=True), nullable=True),
+        sa.ForeignKeyConstraint(
+            ["connector_credential_pair_id"],
+            ["connector_credential_pair.id"],
+        ),
+        sa.PrimaryKeyConstraint("id"),
+    )
+
+    # Create indexes for external_group_permission_sync_attempt
+    op.create_index(
+        "ix_external_group_permission_sync_attempt_time_created",
+        "external_group_permission_sync_attempt",
+        ["time_created"],
+        unique=False,
+    )
+    op.create_index(
+        "ix_group_sync_attempt_cc_pair_time",
+        "external_group_permission_sync_attempt",
+        ["connector_credential_pair_id", "time_created"],
+        unique=False,
+    )
+    op.create_index(
+        "ix_group_sync_attempt_status_time",
+        "external_group_permission_sync_attempt",
+        ["status", sa.text("time_finished DESC")],
+        unique=False,
+    )
+
+
+def downgrade() -> None:
+    # Drop indexes
+    op.drop_index(
+        "ix_group_sync_attempt_status_time",
+        table_name="external_group_permission_sync_attempt",
+    )
+    op.drop_index(
+        "ix_group_sync_attempt_cc_pair_time",
+        table_name="external_group_permission_sync_attempt",
+    )
+    op.drop_index(
+        "ix_external_group_permission_sync_attempt_time_created",
+        table_name="external_group_permission_sync_attempt",
+    )
+    op.drop_index(
+        "ix_permission_sync_attempt_status_time",
+        table_name="doc_permission_sync_attempt",
+    )
+    op.drop_index(
+        "ix_permission_sync_attempt_latest_for_cc_pair",
+        table_name="doc_permission_sync_attempt",
+    )
+    op.drop_index(
+        "ix_doc_permission_sync_attempt_time_created",
+        table_name="doc_permission_sync_attempt",
+    )
+
+    # Drop tables
+    op.drop_table("external_group_permission_sync_attempt")
+    op.drop_table("doc_permission_sync_attempt")

backend/alembic/versions/0816326d83aa_add_federated_connector_tables.py

@@ -0,0 +1,72 @@
+"""add federated connector tables
+
+Revision ID: 0816326d83aa
+Revises: 12635f6655b7
+Create Date: 2025-06-29 14:09:45.109518
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy.dialects import postgresql
+
+
+# revision identifiers, used by Alembic.
+revision = "0816326d83aa"
+down_revision = "12635f6655b7"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    # Create federated_connector table
+    op.create_table(
+        "federated_connector",
+        sa.Column("id", sa.Integer(), nullable=False),
+        sa.Column("source", sa.String(), nullable=False),
+        sa.Column("credentials", sa.LargeBinary(), nullable=False),
+        sa.PrimaryKeyConstraint("id"),
+    )
+
+    # Create federated_connector_oauth_token table
+    op.create_table(
+        "federated_connector_oauth_token",
+        sa.Column("id", sa.Integer(), nullable=False),
+        sa.Column("federated_connector_id", sa.Integer(), nullable=False),
+        sa.Column("user_id", postgresql.UUID(as_uuid=True), nullable=False),
+        sa.Column("token", sa.LargeBinary(), nullable=False),
+        sa.Column("expires_at", sa.DateTime(), nullable=True),
+        sa.ForeignKeyConstraint(
+            ["federated_connector_id"], ["federated_connector.id"], ondelete="CASCADE"
+        ),
+        sa.ForeignKeyConstraint(["user_id"], ["user.id"], ondelete="CASCADE"),
+        sa.PrimaryKeyConstraint("id"),
+    )
+
+    # Create federated_connector__document_set table
+    op.create_table(
+        "federated_connector__document_set",
+        sa.Column("id", sa.Integer(), nullable=False),
+        sa.Column("federated_connector_id", sa.Integer(), nullable=False),
+        sa.Column("document_set_id", sa.Integer(), nullable=False),
+        sa.Column("entities", postgresql.JSONB(), nullable=False),
+        sa.ForeignKeyConstraint(
+            ["federated_connector_id"], ["federated_connector.id"], ondelete="CASCADE"
+        ),
+        sa.ForeignKeyConstraint(
+            ["document_set_id"], ["document_set.id"], ondelete="CASCADE"
+        ),
+        sa.PrimaryKeyConstraint("id"),
+        sa.UniqueConstraint(
+            "federated_connector_id",
+            "document_set_id",
+            name="uq_federated_connector_document_set",
+        ),
+    )
+
+
+def downgrade() -> None:
+    # Drop tables in reverse order due to foreign key dependencies
+    op.drop_table("federated_connector__document_set")
+    op.drop_table("federated_connector_oauth_token")
+    op.drop_table("federated_connector")

backend/alembic/versions/08a1eda20fe1_add_earliest_indexing_to_connector.py

@@ -5,6 +5,7 @@ Revises: 8a87bd6ec550
 Create Date: 2024-07-23 11:12:39.462397
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 

backend/alembic/versions/09995b8811eb_add_theme_preference_to_user.py

@@ -0,0 +1,33 @@
+"""add theme_preference to user
+
+Revision ID: 09995b8811eb
+Revises: 3d1cca026fe8
+Create Date: 2025-10-24 08:58:50.246949
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+from onyx.db.enums import ThemePreference
+
+
+# revision identifiers, used by Alembic.
+revision = "09995b8811eb"
+down_revision = "3d1cca026fe8"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.add_column(
+        "user",
+        sa.Column(
+            "theme_preference",
+            sa.Enum(ThemePreference, native_enum=False),
+            nullable=True,
+        ),
+    )
+
+
+def downgrade() -> None:
+    op.drop_column("user", "theme_preference")

backend/alembic/versions/0a2b51deb0b8_add_starter_prompts.py

@@ -5,6 +5,7 @@ Revises: 5f4b8568a221
 Create Date: 2024-03-02 23:23:49.960309
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 from sqlalchemy.dialects import postgresql

backend/alembic/versions/0a98909f2757_enable_encrypted_fields.py

@@ -5,6 +5,7 @@ Revises: 570282d33c49
 Create Date: 2024-05-05 19:30:34.317972
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 from sqlalchemy.sql import table

backend/alembic/versions/0cd424f32b1d_user_file_data_preparation_and_backfill.py

@@ -0,0 +1,389 @@
+"""Migration 2: User file data preparation and backfill
+
+Revision ID: 0cd424f32b1d
+Revises: 9b66d3156fc6
+Create Date: 2025-09-22 09:44:42.727034
+
+This migration populates the new columns added in migration 1.
+It prepares data for the UUID transition and relationship migration.
+"""
+
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy import text
+import logging
+
+logger = logging.getLogger("alembic.runtime.migration")
+
+# revision identifiers, used by Alembic.
+revision = "0cd424f32b1d"
+down_revision = "9b66d3156fc6"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    """Populate new columns with data."""
+
+    bind = op.get_bind()
+    inspector = sa.inspect(bind)
+
+    # === Step 1: Populate user_file.new_id ===
+    user_file_columns = [col["name"] for col in inspector.get_columns("user_file")]
+    has_new_id = "new_id" in user_file_columns
+
+    if has_new_id:
+        logger.info("Populating user_file.new_id with UUIDs...")
+
+        # Count rows needing UUIDs
+        null_count = bind.execute(
+            text("SELECT COUNT(*) FROM user_file WHERE new_id IS NULL")
+        ).scalar_one()
+
+        if null_count > 0:
+            logger.info(f"Generating UUIDs for {null_count} user_file records...")
+
+            # Populate in batches to avoid long locks
+            batch_size = 10000
+            total_updated = 0
+
+            while True:
+                result = bind.execute(
+                    text(
+                        """
+                    UPDATE user_file
+                    SET new_id = gen_random_uuid()
+                    WHERE new_id IS NULL
+                    AND id IN (
+                        SELECT id FROM user_file
+                        WHERE new_id IS NULL
+                        LIMIT :batch_size
+                    )
+                """
+                    ),
+                    {"batch_size": batch_size},
+                )
+
+                updated = result.rowcount
+                total_updated += updated
+
+                if updated < batch_size:
+                    break
+
+                logger.info(f"  Updated {total_updated}/{null_count} records...")
+
+            logger.info(f"Generated UUIDs for {total_updated} user_file records")
+
+        # Verify all records have UUIDs
+        remaining_null = bind.execute(
+            text("SELECT COUNT(*) FROM user_file WHERE new_id IS NULL")
+        ).scalar_one()
+
+        if remaining_null > 0:
+            raise Exception(
+                f"Failed to populate all user_file.new_id values ({remaining_null} NULL)"
+            )
+
+        # Lock down the column
+        op.alter_column("user_file", "new_id", nullable=False)
+        op.alter_column("user_file", "new_id", server_default=None)
+        logger.info("Locked down user_file.new_id column")
+
+    # === Step 2: Populate persona__user_file.user_file_id_uuid ===
+    persona_user_file_columns = [
+        col["name"] for col in inspector.get_columns("persona__user_file")
+    ]
+
+    if has_new_id and "user_file_id_uuid" in persona_user_file_columns:
+        logger.info("Populating persona__user_file.user_file_id_uuid...")
+
+        # Count rows needing update
+        null_count = bind.execute(
+            text(
+                """
+            SELECT COUNT(*) FROM persona__user_file
+            WHERE user_file_id IS NOT NULL AND user_file_id_uuid IS NULL
+        """
+            )
+        ).scalar_one()
+
+        if null_count > 0:
+            logger.info(f"Updating {null_count} persona__user_file records...")
+
+            # Update in batches
+            batch_size = 10000
+            total_updated = 0
+
+            while True:
+                result = bind.execute(
+                    text(
+                        """
+                    UPDATE persona__user_file p
+                    SET user_file_id_uuid = uf.new_id
+                    FROM user_file uf
+                    WHERE p.user_file_id = uf.id
+                    AND p.user_file_id_uuid IS NULL
+                    AND p.persona_id IN (
+                        SELECT persona_id
+                        FROM persona__user_file
+                        WHERE user_file_id_uuid IS NULL
+                        LIMIT :batch_size
+                    )
+                """
+                    ),
+                    {"batch_size": batch_size},
+                )
+
+                updated = result.rowcount
+                total_updated += updated
+
+                if updated < batch_size:
+                    break
+
+                logger.info(f"  Updated {total_updated}/{null_count} records...")
+
+            logger.info(f"Updated {total_updated} persona__user_file records")
+
+        # Verify all records are populated
+        remaining_null = bind.execute(
+            text(
+                """
+            SELECT COUNT(*) FROM persona__user_file
+            WHERE user_file_id IS NOT NULL AND user_file_id_uuid IS NULL
+        """
+            )
+        ).scalar_one()
+
+        if remaining_null > 0:
+            raise Exception(
+                f"Failed to populate all persona__user_file.user_file_id_uuid values ({remaining_null} NULL)"
+            )
+
+        op.alter_column("persona__user_file", "user_file_id_uuid", nullable=False)
+        logger.info("Locked down persona__user_file.user_file_id_uuid column")
+
+    # === Step 3: Create user_project records from chat_folder ===
+    if "chat_folder" in inspector.get_table_names():
+        logger.info("Creating user_project records from chat_folder...")
+
+        result = bind.execute(
+            text(
+                """
+            INSERT INTO user_project (user_id, name)
+            SELECT cf.user_id, cf.name
+            FROM chat_folder cf
+            WHERE NOT EXISTS (
+                SELECT 1
+                FROM user_project up
+                WHERE up.user_id = cf.user_id AND up.name = cf.name
+            )
+        """
+            )
+        )
+
+        logger.info(f"Created {result.rowcount} user_project records from chat_folder")
+
+    # === Step 4: Populate chat_session.project_id ===
+    chat_session_columns = [
+        col["name"] for col in inspector.get_columns("chat_session")
+    ]
+
+    if "folder_id" in chat_session_columns and "project_id" in chat_session_columns:
+        logger.info("Populating chat_session.project_id...")
+
+        # Count sessions needing update
+        null_count = bind.execute(
+            text(
+                """
+            SELECT COUNT(*) FROM chat_session
+            WHERE project_id IS NULL AND folder_id IS NOT NULL
+        """
+            )
+        ).scalar_one()
+
+        if null_count > 0:
+            logger.info(f"Updating {null_count} chat_session records...")
+
+            result = bind.execute(
+                text(
+                    """
+                UPDATE chat_session cs
+                SET project_id = up.id
+                FROM chat_folder cf
+                JOIN user_project up ON up.user_id = cf.user_id AND up.name = cf.name
+                WHERE cs.folder_id = cf.id AND cs.project_id IS NULL
+            """
+                )
+            )
+
+            logger.info(f"Updated {result.rowcount} chat_session records")
+
+        # Verify all records are populated
+        remaining_null = bind.execute(
+            text(
+                """
+            SELECT COUNT(*) FROM chat_session
+            WHERE project_id IS NULL AND folder_id IS NOT NULL
+        """
+            )
+        ).scalar_one()
+
+        if remaining_null > 0:
+            logger.warning(
+                f"Warning: {remaining_null} chat_session records could not be mapped to projects"
+            )
+
+    # === Step 5: Update plaintext FileRecord IDs/display names to UUID scheme ===
+    # Prior to UUID migration, plaintext cache files were stored with file_id like 'plain_text_<int_id>'.
+    # After migration, we use 'plaintext_<uuid>' (note the name change to 'plaintext_').
+    # This step remaps existing FileRecord rows to the new naming while preserving object_key/bucket.
+    logger.info("Updating plaintext FileRecord ids and display names to UUID scheme...")
+
+    # Count legacy plaintext records that can be mapped to UUID user_file ids
+    count_query = text(
+        """
+        SELECT COUNT(*)
+        FROM file_record fr
+        JOIN user_file uf ON fr.file_id = CONCAT('plaintext_', uf.id::text)
+        WHERE LOWER(fr.file_origin::text) = 'plaintext_cache'
+        """
+    )
+    legacy_count = bind.execute(count_query).scalar_one()
+
+    if legacy_count and legacy_count > 0:
+        logger.info(f"Found {legacy_count} legacy plaintext file records to update")
+
+        # Update display_name first for readability (safe regardless of rename)
+        bind.execute(
+            text(
+                """
+                UPDATE file_record fr
+                SET display_name = CONCAT('Plaintext for user file ', uf.new_id::text)
+                FROM user_file uf
+                WHERE LOWER(fr.file_origin::text) = 'plaintext_cache'
+                    AND fr.file_id = CONCAT('plaintext_', uf.id::text)
+                """
+            )
+        )
+
+        # Remap file_id from 'plaintext_<int>' -> 'plaintext_<uuid>' using transitional new_id
+        # Use a single UPDATE ... WHERE file_id LIKE 'plain_text_%'
+        # and ensure it aligns to existing user_file ids to avoid renaming unrelated rows
+        result = bind.execute(
+            text(
+                """
+                UPDATE file_record fr
+                SET file_id = CONCAT('plaintext_', uf.new_id::text)
+                FROM user_file uf
+                WHERE LOWER(fr.file_origin::text) = 'plaintext_cache'
+                    AND fr.file_id = CONCAT('plaintext_', uf.id::text)
+                """
+            )
+        )
+        logger.info(
+            f"Updated {result.rowcount} plaintext file_record ids to UUID scheme"
+        )
+
+    # === Step 6: Ensure document_id_migrated default TRUE and backfill existing FALSE ===
+    # New records should default to migrated=True so the migration task won't run for them.
+    # Existing rows that had a legacy document_id should be marked as not migrated to be processed.
+
+    # Backfill existing records: if document_id is not null, set to FALSE
+    bind.execute(
+        text(
+            """
+            UPDATE user_file
+            SET document_id_migrated = FALSE
+            WHERE document_id IS NOT NULL
+            """
+        )
+    )
+
+    # === Step 7: Backfill user_file.status from index_attempt ===
+    logger.info("Backfilling user_file.status from index_attempt...")
+
+    # Update user_file status based on latest index attempt
+    # Using CTEs instead of temp tables for asyncpg compatibility
+    result = bind.execute(
+        text(
+            """
+        WITH latest_attempt AS (
+            SELECT DISTINCT ON (ia.connector_credential_pair_id)
+                ia.connector_credential_pair_id,
+                ia.status
+            FROM index_attempt ia
+            ORDER BY ia.connector_credential_pair_id, ia.time_updated DESC
+        ),
+        uf_to_ccp AS (
+            SELECT DISTINCT uf.id AS uf_id, ccp.id AS cc_pair_id
+            FROM user_file uf
+            JOIN document_by_connector_credential_pair dcc
+                ON dcc.id = REPLACE(uf.document_id, 'USER_FILE_CONNECTOR__', 'FILE_CONNECTOR__')
+            JOIN connector_credential_pair ccp
+                ON ccp.connector_id = dcc.connector_id
+                AND ccp.credential_id = dcc.credential_id
+        )
+        UPDATE user_file uf
+        SET status = CASE
+            WHEN la.status IN ('NOT_STARTED', 'IN_PROGRESS') THEN 'PROCESSING'
+            WHEN la.status = 'SUCCESS' THEN 'COMPLETED'
+            ELSE 'FAILED'
+        END
+        FROM uf_to_ccp ufc
+        LEFT JOIN latest_attempt la
+            ON la.connector_credential_pair_id = ufc.cc_pair_id
+        WHERE uf.id = ufc.uf_id
+        AND uf.status = 'PROCESSING'
+    """
+        )
+    )
+
+    logger.info(f"Updated status for {result.rowcount} user_file records")
+
+    logger.info("Migration 2 (data preparation) completed successfully")
+
+
+def downgrade() -> None:
+    """Reset populated data to allow clean downgrade of schema."""
+
+    bind = op.get_bind()
+    inspector = sa.inspect(bind)
+
+    logger.info("Starting downgrade of data preparation...")
+
+    # Reset user_file columns to allow nulls before data removal
+    if "user_file" in inspector.get_table_names():
+        columns = [col["name"] for col in inspector.get_columns("user_file")]
+
+        if "new_id" in columns:
+            op.alter_column(
+                "user_file",
+                "new_id",
+                nullable=True,
+                server_default=sa.text("gen_random_uuid()"),
+            )
+            # Optionally clear the data
+            # bind.execute(text("UPDATE user_file SET new_id = NULL"))
+            logger.info("Reset user_file.new_id to nullable")
+
+    # Reset persona__user_file.user_file_id_uuid
+    if "persona__user_file" in inspector.get_table_names():
+        columns = [col["name"] for col in inspector.get_columns("persona__user_file")]
+
+        if "user_file_id_uuid" in columns:
+            op.alter_column("persona__user_file", "user_file_id_uuid", nullable=True)
+            # Optionally clear the data
+            # bind.execute(text("UPDATE persona__user_file SET user_file_id_uuid = NULL"))
+            logger.info("Reset persona__user_file.user_file_id_uuid to nullable")
+
+    # Note: We don't delete user_project records or reset chat_session.project_id
+    # as these might be in use and can be handled by the schema downgrade
+
+    # Reset user_file.status to default
+    if "user_file" in inspector.get_table_names():
+        columns = [col["name"] for col in inspector.get_columns("user_file")]
+        if "status" in columns:
+            bind.execute(text("UPDATE user_file SET status = 'PROCESSING'"))
+            logger.info("Reset user_file.status to default")
+
+    logger.info("Downgrade completed successfully")

backend/alembic/versions/0ebb1d516877_add_ccpair_deletion_failure_message.py

@@ -5,6 +5,7 @@ Revises: 52a219fb5233
 Create Date: 2024-09-10 15:03:48.233926
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 

backend/alembic/versions/0f7ff6d75b57_add_index_to_index_attempt_time_created.py

@@ -5,6 +5,7 @@ Revises: 369644546676
 Create Date: 2025-01-10 14:01:14.067144
 
 """
+
 from alembic import op
 
 # revision identifiers, used by Alembic.

backend/alembic/versions/12635f6655b7_drive_canonical_ids.py

@@ -0,0 +1,596 @@
+"""drive-canonical-ids
+
+Revision ID: 12635f6655b7
+Revises: 58c50ef19f08
+Create Date: 2025-06-20 14:44:54.241159
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+from urllib.parse import urlparse, urlunparse
+from httpx import HTTPStatusError
+import httpx
+from onyx.document_index.factory import get_default_document_index
+from onyx.db.search_settings import SearchSettings
+from onyx.document_index.vespa.shared_utils.utils import get_vespa_http_client
+from onyx.document_index.vespa.shared_utils.utils import (
+    replace_invalid_doc_id_characters,
+)
+from onyx.document_index.vespa_constants import DOCUMENT_ID_ENDPOINT
+from onyx.utils.logger import setup_logger
+import os
+
+logger = setup_logger()
+
+# revision identifiers, used by Alembic.
+revision = "12635f6655b7"
+down_revision = "58c50ef19f08"
+branch_labels = None
+depends_on = None
+
+SKIP_CANON_DRIVE_IDS = os.environ.get("SKIP_CANON_DRIVE_IDS", "true").lower() == "true"
+
+
+def active_search_settings() -> tuple[SearchSettings, SearchSettings | None]:
+    result = op.get_bind().execute(
+        sa.text(
+            """
+        SELECT * FROM search_settings WHERE status = 'PRESENT' ORDER BY id DESC LIMIT 1
+        """
+        )
+    )
+    search_settings_fetch = result.fetchall()
+    search_settings = (
+        SearchSettings(**search_settings_fetch[0]._asdict())
+        if search_settings_fetch
+        else None
+    )
+
+    result2 = op.get_bind().execute(
+        sa.text(
+            """
+        SELECT * FROM search_settings WHERE status = 'FUTURE' ORDER BY id DESC LIMIT 1
+        """
+        )
+    )
+    search_settings_future_fetch = result2.fetchall()
+    search_settings_future = (
+        SearchSettings(**search_settings_future_fetch[0]._asdict())
+        if search_settings_future_fetch
+        else None
+    )
+
+    if not isinstance(search_settings, SearchSettings):
+        raise RuntimeError(
+            "current search settings is of type " + str(type(search_settings))
+        )
+    if (
+        not isinstance(search_settings_future, SearchSettings)
+        and search_settings_future is not None
+    ):
+        raise RuntimeError(
+            "future search settings is of type " + str(type(search_settings_future))
+        )
+
+    return search_settings, search_settings_future
+
+
+def normalize_google_drive_url(url: str) -> str:
+    """Remove query parameters from Google Drive URLs to create canonical document IDs.
+    NOTE: copied from drive doc_conversion.py
+    """
+    parsed_url = urlparse(url)
+    parsed_url = parsed_url._replace(query="")
+    spl_path = parsed_url.path.split("/")
+    if spl_path and (spl_path[-1] in ["edit", "view", "preview"]):
+        spl_path.pop()
+        parsed_url = parsed_url._replace(path="/".join(spl_path))
+    # Remove query parameters and reconstruct URL
+    return urlunparse(parsed_url)
+
+
+def get_google_drive_documents_from_database() -> list[dict]:
+    """Get all Google Drive documents from the database."""
+    bind = op.get_bind()
+    result = bind.execute(
+        sa.text(
+            """
+            SELECT d.id
+            FROM document d
+            JOIN document_by_connector_credential_pair dcc ON d.id = dcc.id
+            JOIN connector_credential_pair cc ON dcc.connector_id = cc.connector_id
+                AND dcc.credential_id = cc.credential_id
+            JOIN connector c ON cc.connector_id = c.id
+            WHERE c.source = 'GOOGLE_DRIVE'
+        """
+        )
+    )
+
+    documents = []
+    for row in result:
+        documents.append({"document_id": row.id})
+
+    return documents
+
+
+def update_document_id_in_database(
+    old_doc_id: str, new_doc_id: str, index_name: str
+) -> None:
+    """Update document IDs in all relevant database tables using copy-and-swap approach."""
+    bind = op.get_bind()
+
+    # print(f"Updating database tables for document {old_doc_id} -> {new_doc_id}")
+
+    # Check if new document ID already exists
+    result = bind.execute(
+        sa.text("SELECT COUNT(*) FROM document WHERE id = :new_id"),
+        {"new_id": new_doc_id},
+    )
+    row = result.fetchone()
+    if row and row[0] > 0:
+        # print(f"Document with ID {new_doc_id} already exists, deleting old one")
+        delete_document_from_db(old_doc_id, index_name)
+        return
+
+    # Step 1: Create a new document row with the new ID (copy all fields from old row)
+    # Use a conservative approach to handle columns that might not exist in all installations
+    try:
+        bind.execute(
+            sa.text(
+                """
+                INSERT INTO document (id, from_ingestion_api, boost, hidden, semantic_id,
+                                    link, doc_updated_at, primary_owners, secondary_owners,
+                                    external_user_emails, external_user_group_ids, is_public,
+                                    chunk_count, last_modified, last_synced, kg_stage, kg_processing_time)
+                SELECT :new_id, from_ingestion_api, boost, hidden, semantic_id,
+                       link, doc_updated_at, primary_owners, secondary_owners,
+                       external_user_emails, external_user_group_ids, is_public,
+                       chunk_count, last_modified, last_synced, kg_stage, kg_processing_time
+                FROM document
+                WHERE id = :old_id
+            """
+            ),
+            {"new_id": new_doc_id, "old_id": old_doc_id},
+        )
+        # print(f"Successfully updated database tables for document {old_doc_id} -> {new_doc_id}")
+    except Exception as e:
+        # If the full INSERT fails, try a more basic version with only core columns
+        logger.warning(f"Full INSERT failed, trying basic version: {e}")
+        bind.execute(
+            sa.text(
+                """
+                INSERT INTO document (id, from_ingestion_api, boost, hidden, semantic_id,
+                                    link, doc_updated_at, primary_owners, secondary_owners)
+                SELECT :new_id, from_ingestion_api, boost, hidden, semantic_id,
+                       link, doc_updated_at, primary_owners, secondary_owners
+                FROM document
+                WHERE id = :old_id
+            """
+            ),
+            {"new_id": new_doc_id, "old_id": old_doc_id},
+        )
+
+    # Step 2: Update all foreign key references to point to the new ID
+
+    # Update document_by_connector_credential_pair table
+    bind.execute(
+        sa.text(
+            "UPDATE document_by_connector_credential_pair SET id = :new_id WHERE id = :old_id"
+        ),
+        {"new_id": new_doc_id, "old_id": old_doc_id},
+    )
+    # print(f"Successfully updated document_by_connector_credential_pair table for document {old_doc_id} -> {new_doc_id}")
+
+    # Update search_doc table (stores search results for chat replay)
+    # This is critical for agent functionality
+    bind.execute(
+        sa.text(
+            "UPDATE search_doc SET document_id = :new_id WHERE document_id = :old_id"
+        ),
+        {"new_id": new_doc_id, "old_id": old_doc_id},
+    )
+    # print(f"Successfully updated search_doc table for document {old_doc_id} -> {new_doc_id}")
+    # Update document_retrieval_feedback table (user feedback on documents)
+    bind.execute(
+        sa.text(
+            "UPDATE document_retrieval_feedback SET document_id = :new_id WHERE document_id = :old_id"
+        ),
+        {"new_id": new_doc_id, "old_id": old_doc_id},
+    )
+    # print(f"Successfully updated document_retrieval_feedback table for document {old_doc_id} -> {new_doc_id}")
+    # Update document__tag table (document-tag relationships)
+    bind.execute(
+        sa.text(
+            "UPDATE document__tag SET document_id = :new_id WHERE document_id = :old_id"
+        ),
+        {"new_id": new_doc_id, "old_id": old_doc_id},
+    )
+    # print(f"Successfully updated document__tag table for document {old_doc_id} -> {new_doc_id}")
+    # Update user_file table (user uploaded files linked to documents)
+    bind.execute(
+        sa.text(
+            "UPDATE user_file SET document_id = :new_id WHERE document_id = :old_id"
+        ),
+        {"new_id": new_doc_id, "old_id": old_doc_id},
+    )
+    # print(f"Successfully updated user_file table for document {old_doc_id} -> {new_doc_id}")
+    # Update KG and chunk_stats tables (these may not exist in all installations)
+    try:
+        # Update kg_entity table
+        bind.execute(
+            sa.text(
+                "UPDATE kg_entity SET document_id = :new_id WHERE document_id = :old_id"
+            ),
+            {"new_id": new_doc_id, "old_id": old_doc_id},
+        )
+        # print(f"Successfully updated kg_entity table for document {old_doc_id} -> {new_doc_id}")
+        # Update kg_entity_extraction_staging table
+        bind.execute(
+            sa.text(
+                "UPDATE kg_entity_extraction_staging SET document_id = :new_id WHERE document_id = :old_id"
+            ),
+            {"new_id": new_doc_id, "old_id": old_doc_id},
+        )
+        # print(f"Successfully updated kg_entity_extraction_staging table for document {old_doc_id} -> {new_doc_id}")
+        # Update kg_relationship table
+        bind.execute(
+            sa.text(
+                "UPDATE kg_relationship SET source_document = :new_id WHERE source_document = :old_id"
+            ),
+            {"new_id": new_doc_id, "old_id": old_doc_id},
+        )
+        # print(f"Successfully updated kg_relationship table for document {old_doc_id} -> {new_doc_id}")
+        # Update kg_relationship_extraction_staging table
+        bind.execute(
+            sa.text(
+                "UPDATE kg_relationship_extraction_staging SET source_document = :new_id WHERE source_document = :old_id"
+            ),
+            {"new_id": new_doc_id, "old_id": old_doc_id},
+        )
+        # print(f"Successfully updated kg_relationship_extraction_staging table for document {old_doc_id} -> {new_doc_id}")
+        # Update chunk_stats table
+        bind.execute(
+            sa.text(
+                "UPDATE chunk_stats SET document_id = :new_id WHERE document_id = :old_id"
+            ),
+            {"new_id": new_doc_id, "old_id": old_doc_id},
+        )
+        # print(f"Successfully updated chunk_stats table for document {old_doc_id} -> {new_doc_id}")
+        # Update chunk_stats ID field which includes document_id
+        bind.execute(
+            sa.text(
+                """
+                UPDATE chunk_stats
+                SET id = REPLACE(id, :old_id, :new_id)
+                WHERE id LIKE :old_id_pattern
+            """
+            ),
+            {
+                "new_id": new_doc_id,
+                "old_id": old_doc_id,
+                "old_id_pattern": f"{old_doc_id}__%",
+            },
+        )
+        # print(f"Successfully updated chunk_stats ID field for document {old_doc_id} -> {new_doc_id}")
+    except Exception as e:
+        logger.warning(f"Some KG/chunk tables may not exist or failed to update: {e}")
+
+    # Step 3: Delete the old document row (this should now be safe since all FKs point to new row)
+    bind.execute(
+        sa.text("DELETE FROM document WHERE id = :old_id"), {"old_id": old_doc_id}
+    )
+    # print(f"Successfully deleted document {old_doc_id} from database")
+
+
+def _visit_chunks(
+    *,
+    http_client: httpx.Client,
+    index_name: str,
+    selection: str,
+    continuation: str | None = None,
+) -> tuple[list[dict], str | None]:
+    """Helper that calls the /document/v1 visit API once and returns (docs, next_token)."""
+
+    # Use the same URL as the document API, but with visit-specific params
+    base_url = DOCUMENT_ID_ENDPOINT.format(index_name=index_name)
+
+    params: dict[str, str] = {
+        "selection": selection,
+        "wantedDocumentCount": "1000",
+    }
+    if continuation:
+        params["continuation"] = continuation
+
+    # print(f"Visiting chunks for selection '{selection}' with params {params}")
+    resp = http_client.get(base_url, params=params, timeout=None)
+    # print(f"Visited chunks for document {selection}")
+    resp.raise_for_status()
+
+    payload = resp.json()
+    return payload.get("documents", []), payload.get("continuation")
+
+
+def delete_document_chunks_from_vespa(index_name: str, doc_id: str) -> None:
+    """Delete all chunks for *doc_id* from Vespa using continuation-token paging (no offset)."""
+
+    total_deleted = 0
+    # Use exact match instead of contains - Document Selector Language doesn't support contains
+    selection = f'{index_name}.document_id=="{doc_id}"'
+
+    with get_vespa_http_client() as http_client:
+        continuation: str | None = None
+        while True:
+            docs, continuation = _visit_chunks(
+                http_client=http_client,
+                index_name=index_name,
+                selection=selection,
+                continuation=continuation,
+            )
+
+            if not docs:
+                break
+
+            for doc in docs:
+                vespa_full_id = doc.get("id")
+                if not vespa_full_id:
+                    continue
+
+                vespa_doc_uuid = vespa_full_id.split("::")[-1]
+                delete_url = f"{DOCUMENT_ID_ENDPOINT.format(index_name=index_name)}/{vespa_doc_uuid}"
+
+                try:
+                    resp = http_client.delete(delete_url)
+                    resp.raise_for_status()
+                    total_deleted += 1
+                except Exception as e:
+                    print(f"Failed to delete chunk {vespa_doc_uuid}: {e}")
+
+            if not continuation:
+                break
+
+
+def update_document_id_in_vespa(
+    index_name: str, old_doc_id: str, new_doc_id: str
+) -> None:
+    """Update all chunks' document_id field from *old_doc_id* to *new_doc_id* using continuation paging."""
+
+    clean_new_doc_id = replace_invalid_doc_id_characters(new_doc_id)
+
+    # Use exact match instead of contains - Document Selector Language doesn't support contains
+    selection = f'{index_name}.document_id=="{old_doc_id}"'
+
+    with get_vespa_http_client() as http_client:
+        continuation: str | None = None
+        while True:
+            # print(f"Visiting chunks for document {old_doc_id} -> {new_doc_id}")
+            docs, continuation = _visit_chunks(
+                http_client=http_client,
+                index_name=index_name,
+                selection=selection,
+                continuation=continuation,
+            )
+
+            if not docs:
+                break
+
+            for doc in docs:
+                vespa_full_id = doc.get("id")
+                if not vespa_full_id:
+                    continue
+
+                vespa_doc_uuid = vespa_full_id.split("::")[-1]
+                vespa_url = f"{DOCUMENT_ID_ENDPOINT.format(index_name=index_name)}/{vespa_doc_uuid}"
+
+                update_request = {
+                    "fields": {"document_id": {"assign": clean_new_doc_id}}
+                }
+
+                try:
+                    resp = http_client.put(vespa_url, json=update_request)
+                    resp.raise_for_status()
+                except Exception as e:
+                    print(f"Failed to update chunk {vespa_doc_uuid}: {e}")
+                    raise
+
+            if not continuation:
+                break
+
+
+def delete_document_from_db(current_doc_id: str, index_name: str) -> None:
+    # Delete all foreign key references first, then delete the document
+    try:
+        bind = op.get_bind()
+
+        # Delete from agent-related tables first (order matters due to foreign keys)
+        # Delete from agent__sub_query__search_doc first since it references search_doc
+        bind.execute(
+            sa.text(
+                """
+                DELETE FROM agent__sub_query__search_doc
+                WHERE search_doc_id IN (
+                    SELECT id FROM search_doc WHERE document_id = :doc_id
+                )
+                """
+            ),
+            {"doc_id": current_doc_id},
+        )
+
+        # Delete from chat_message__search_doc
+        bind.execute(
+            sa.text(
+                """
+                DELETE FROM chat_message__search_doc
+                WHERE search_doc_id IN (
+                    SELECT id FROM search_doc WHERE document_id = :doc_id
+                )
+                """
+            ),
+            {"doc_id": current_doc_id},
+        )
+
+        # Now we can safely delete from search_doc
+        bind.execute(
+            sa.text("DELETE FROM search_doc WHERE document_id = :doc_id"),
+            {"doc_id": current_doc_id},
+        )
+
+        # Delete from document_by_connector_credential_pair
+        bind.execute(
+            sa.text(
+                "DELETE FROM document_by_connector_credential_pair WHERE id = :doc_id"
+            ),
+            {"doc_id": current_doc_id},
+        )
+
+        # Delete from other tables that reference this document
+        bind.execute(
+            sa.text(
+                "DELETE FROM document_retrieval_feedback WHERE document_id = :doc_id"
+            ),
+            {"doc_id": current_doc_id},
+        )
+
+        bind.execute(
+            sa.text("DELETE FROM document__tag WHERE document_id = :doc_id"),
+            {"doc_id": current_doc_id},
+        )
+
+        bind.execute(
+            sa.text("DELETE FROM user_file WHERE document_id = :doc_id"),
+            {"doc_id": current_doc_id},
+        )
+
+        # Delete from KG tables if they exist
+        try:
+            bind.execute(
+                sa.text("DELETE FROM kg_entity WHERE document_id = :doc_id"),
+                {"doc_id": current_doc_id},
+            )
+
+            bind.execute(
+                sa.text(
+                    "DELETE FROM kg_entity_extraction_staging WHERE document_id = :doc_id"
+                ),
+                {"doc_id": current_doc_id},
+            )
+
+            bind.execute(
+                sa.text("DELETE FROM kg_relationship WHERE source_document = :doc_id"),
+                {"doc_id": current_doc_id},
+            )
+
+            bind.execute(
+                sa.text(
+                    "DELETE FROM kg_relationship_extraction_staging WHERE source_document = :doc_id"
+                ),
+                {"doc_id": current_doc_id},
+            )
+
+            bind.execute(
+                sa.text("DELETE FROM chunk_stats WHERE document_id = :doc_id"),
+                {"doc_id": current_doc_id},
+            )
+
+            bind.execute(
+                sa.text("DELETE FROM chunk_stats WHERE id LIKE :doc_id_pattern"),
+                {"doc_id_pattern": f"{current_doc_id}__%"},
+            )
+
+        except Exception as e:
+            logger.warning(
+                f"Some KG/chunk tables may not exist or failed to delete from: {e}"
+            )
+
+        # Finally delete the document itself
+        bind.execute(
+            sa.text("DELETE FROM document WHERE id = :doc_id"),
+            {"doc_id": current_doc_id},
+        )
+
+        # Delete chunks from vespa
+        delete_document_chunks_from_vespa(index_name, current_doc_id)
+
+    except Exception as e:
+        print(f"Failed to delete duplicate document {current_doc_id}: {e}")
+        # Continue with other documents instead of failing the entire migration
+
+
+def upgrade() -> None:
+    if SKIP_CANON_DRIVE_IDS:
+        return
+    current_search_settings, future_search_settings = active_search_settings()
+    document_index = get_default_document_index(
+        current_search_settings,
+        future_search_settings,
+    )
+
+    # Get the index name
+    if hasattr(document_index, "index_name"):
+        index_name = document_index.index_name
+    else:
+        # Default index name if we can't get it from the document_index
+        index_name = "danswer_index"
+
+    # Get all Google Drive documents from the database (this is faster and more reliable)
+    gdrive_documents = get_google_drive_documents_from_database()
+
+    if not gdrive_documents:
+        return
+
+    # Track normalized document IDs to detect duplicates
+    all_normalized_doc_ids = set()
+    updated_count = 0
+
+    for doc_info in gdrive_documents:
+        current_doc_id = doc_info["document_id"]
+        normalized_doc_id = normalize_google_drive_url(current_doc_id)
+
+        print(f"Processing document {current_doc_id} -> {normalized_doc_id}")
+        # Check for duplicates
+        if normalized_doc_id in all_normalized_doc_ids:
+            # print(f"Deleting duplicate document {current_doc_id}")
+            delete_document_from_db(current_doc_id, index_name)
+            continue
+
+        all_normalized_doc_ids.add(normalized_doc_id)
+
+        # If the document ID already doesn't have query parameters, skip it
+        if current_doc_id == normalized_doc_id:
+            # print(f"Skipping document {current_doc_id} -> {normalized_doc_id} because it already has no query parameters")
+            continue
+
+        try:
+            # Update both database and Vespa in order
+            # Database first to ensure consistency
+            update_document_id_in_database(
+                current_doc_id, normalized_doc_id, index_name
+            )
+
+            # For Vespa, we can now use the original document IDs since we're using contains matching
+            update_document_id_in_vespa(index_name, current_doc_id, normalized_doc_id)
+            updated_count += 1
+            # print(f"Finished updating document {current_doc_id} -> {normalized_doc_id}")
+        except Exception as e:
+            print(f"Failed to update document {current_doc_id}: {e}")
+
+            if isinstance(e, HTTPStatusError):
+                print(f"HTTPStatusError: {e}")
+                print(f"Response: {e.response.text}")
+                print(f"Status: {e.response.status_code}")
+                print(f"Headers: {e.response.headers}")
+                print(f"Request: {e.request.url}")
+                print(f"Request headers: {e.request.headers}")
+            # Note: Rollback is complex with copy-and-swap approach since the old document is already deleted
+            # In case of failure, manual intervention may be required
+            # Continue with other documents instead of failing the entire migration
+            continue
+
+    logger.info(f"Migration complete. Updated {updated_count} Google Drive documents")
+
+
+def downgrade() -> None:
+    # this is a one way migration, so no downgrade.
+    # It wouldn't make sense to store the extra query parameters
+    # and duplicate documents to allow a reversal.
+    pass

backend/alembic/versions/15326fcec57e_introduce_onyx_apis.py

@@ -5,6 +5,7 @@ Revises: 77d07dffae64
 Create Date: 2023-11-11 20:51:24.228999
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 

backend/alembic/versions/16c37a30adf2_user_file_relationship_migration.py

@@ -0,0 +1,261 @@
+"""Migration 3: User file relationship migration
+
+Revision ID: 16c37a30adf2
+Revises: 0cd424f32b1d
+Create Date: 2025-09-22 09:47:34.175596
+
+This migration converts folder-based relationships to project-based relationships.
+It migrates persona__user_folder to persona__user_file and populates project__user_file.
+"""
+
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy import text
+import logging
+
+logger = logging.getLogger("alembic.runtime.migration")
+
+# revision identifiers, used by Alembic.
+revision = "16c37a30adf2"
+down_revision = "0cd424f32b1d"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    """Migrate folder-based relationships to project-based relationships."""
+
+    bind = op.get_bind()
+    inspector = sa.inspect(bind)
+
+    # === Step 1: Migrate persona__user_folder to persona__user_file ===
+    table_names = inspector.get_table_names()
+
+    if "persona__user_folder" in table_names and "user_file" in table_names:
+        user_file_columns = [col["name"] for col in inspector.get_columns("user_file")]
+        has_new_id = "new_id" in user_file_columns
+
+        if has_new_id and "folder_id" in user_file_columns:
+            logger.info(
+                "Migrating persona__user_folder relationships to persona__user_file..."
+            )
+
+            # Count relationships to migrate (asyncpg-compatible)
+            count_query = text(
+                """
+                SELECT COUNT(*)
+                FROM (
+                    SELECT DISTINCT puf.persona_id, uf.id
+                    FROM persona__user_folder puf
+                    JOIN user_file uf ON uf.folder_id = puf.user_folder_id
+                    WHERE NOT EXISTS (
+                        SELECT 1
+                        FROM persona__user_file p2
+                        WHERE p2.persona_id = puf.persona_id
+                        AND p2.user_file_id = uf.id
+                    )
+                ) AS distinct_pairs
+            """
+            )
+            to_migrate = bind.execute(count_query).scalar_one()
+
+            if to_migrate > 0:
+                logger.info(f"Creating {to_migrate} persona-file relationships...")
+
+                # Migrate in batches to avoid memory issues
+                batch_size = 10000
+                total_inserted = 0
+
+                while True:
+                    # Insert batch directly using subquery (asyncpg compatible)
+                    result = bind.execute(
+                        text(
+                            """
+                        INSERT INTO persona__user_file (persona_id, user_file_id, user_file_id_uuid)
+                        SELECT DISTINCT puf.persona_id, uf.id as file_id, uf.new_id
+                        FROM persona__user_folder puf
+                        JOIN user_file uf ON uf.folder_id = puf.user_folder_id
+                        WHERE NOT EXISTS (
+                            SELECT 1
+                            FROM persona__user_file p2
+                            WHERE p2.persona_id = puf.persona_id
+                            AND p2.user_file_id = uf.id
+                        )
+                        LIMIT :batch_size
+                    """
+                        ),
+                        {"batch_size": batch_size},
+                    )
+
+                    inserted = result.rowcount
+                    total_inserted += inserted
+
+                    if inserted < batch_size:
+                        break
+
+                    logger.info(
+                        f"  Migrated {total_inserted}/{to_migrate} relationships..."
+                    )
+
+                logger.info(
+                    f"Created {total_inserted} persona__user_file relationships"
+                )
+
+    # === Step 2: Add foreign key for chat_session.project_id ===
+    chat_session_fks = inspector.get_foreign_keys("chat_session")
+    fk_exists = any(
+        fk["name"] == "fk_chat_session_project_id" for fk in chat_session_fks
+    )
+
+    if not fk_exists:
+        logger.info("Adding foreign key constraint for chat_session.project_id...")
+        op.create_foreign_key(
+            "fk_chat_session_project_id",
+            "chat_session",
+            "user_project",
+            ["project_id"],
+            ["id"],
+        )
+        logger.info("Added foreign key constraint")
+
+    # === Step 3: Populate project__user_file from user_file.folder_id ===
+    user_file_columns = [col["name"] for col in inspector.get_columns("user_file")]
+    has_new_id = "new_id" in user_file_columns
+
+    if has_new_id and "folder_id" in user_file_columns:
+        logger.info("Populating project__user_file from folder relationships...")
+
+        # Count relationships to create
+        count_query = text(
+            """
+            SELECT COUNT(*)
+            FROM user_file uf
+            WHERE uf.folder_id IS NOT NULL
+            AND NOT EXISTS (
+                SELECT 1
+                FROM project__user_file puf
+                WHERE puf.project_id = uf.folder_id
+                AND puf.user_file_id = uf.new_id
+            )
+        """
+        )
+        to_create = bind.execute(count_query).scalar_one()
+
+        if to_create > 0:
+            logger.info(f"Creating {to_create} project-file relationships...")
+
+            # Insert in batches
+            batch_size = 10000
+            total_inserted = 0
+
+            while True:
+                result = bind.execute(
+                    text(
+                        """
+                    INSERT INTO project__user_file (project_id, user_file_id)
+                    SELECT uf.folder_id, uf.new_id
+                    FROM user_file uf
+                    WHERE uf.folder_id IS NOT NULL
+                    AND NOT EXISTS (
+                        SELECT 1
+                        FROM project__user_file puf
+                        WHERE puf.project_id = uf.folder_id
+                        AND puf.user_file_id = uf.new_id
+                    )
+                    LIMIT :batch_size
+                    ON CONFLICT (project_id, user_file_id) DO NOTHING
+                """
+                    ),
+                    {"batch_size": batch_size},
+                )
+
+                inserted = result.rowcount
+                total_inserted += inserted
+
+                if inserted < batch_size:
+                    break
+
+                logger.info(f"  Created {total_inserted}/{to_create} relationships...")
+
+            logger.info(f"Created {total_inserted} project__user_file relationships")
+
+    # === Step 4: Create index on chat_session.project_id ===
+    try:
+        indexes = [ix.get("name") for ix in inspector.get_indexes("chat_session")]
+    except Exception:
+        indexes = []
+
+    if "ix_chat_session_project_id" not in indexes:
+        logger.info("Creating index on chat_session.project_id...")
+        op.create_index(
+            "ix_chat_session_project_id", "chat_session", ["project_id"], unique=False
+        )
+        logger.info("Created index")
+
+    logger.info("Migration 3 (relationship migration) completed successfully")
+
+
+def downgrade() -> None:
+    """Remove migrated relationships and constraints."""
+
+    bind = op.get_bind()
+    inspector = sa.inspect(bind)
+
+    logger.info("Starting downgrade of relationship migration...")
+
+    # Drop index on chat_session.project_id
+    try:
+        indexes = [ix.get("name") for ix in inspector.get_indexes("chat_session")]
+        if "ix_chat_session_project_id" in indexes:
+            op.drop_index("ix_chat_session_project_id", "chat_session")
+            logger.info("Dropped index on chat_session.project_id")
+    except Exception:
+        pass
+
+    # Drop foreign key constraint
+    try:
+        chat_session_fks = inspector.get_foreign_keys("chat_session")
+        fk_exists = any(
+            fk["name"] == "fk_chat_session_project_id" for fk in chat_session_fks
+        )
+        if fk_exists:
+            op.drop_constraint(
+                "fk_chat_session_project_id", "chat_session", type_="foreignkey"
+            )
+            logger.info("Dropped foreign key constraint on chat_session.project_id")
+    except Exception:
+        pass
+
+    # Clear project__user_file relationships (but keep the table for migration 1 to handle)
+    if "project__user_file" in inspector.get_table_names():
+        result = bind.execute(text("DELETE FROM project__user_file"))
+        logger.info(f"Cleared {result.rowcount} records from project__user_file")
+
+    # Remove migrated persona__user_file relationships
+    # Only remove those that came from folder relationships
+    if all(
+        table in inspector.get_table_names()
+        for table in ["persona__user_file", "persona__user_folder", "user_file"]
+    ):
+        user_file_columns = [col["name"] for col in inspector.get_columns("user_file")]
+        if "folder_id" in user_file_columns:
+            result = bind.execute(
+                text(
+                    """
+                DELETE FROM persona__user_file puf
+                WHERE EXISTS (
+                    SELECT 1
+                    FROM user_file uf
+                    JOIN persona__user_folder puf2
+                        ON puf2.user_folder_id = uf.folder_id
+                    WHERE puf.persona_id = puf2.persona_id
+                    AND puf.user_file_id = uf.id
+                )
+            """
+                )
+            )
+            logger.info(
+                f"Removed {result.rowcount} migrated persona__user_file relationships"
+            )
+
+    logger.info("Downgrade completed successfully")

backend/alembic/versions/173cae5bba26_port_config_store.py

@@ -5,6 +5,7 @@ Revises: e50154680a5c
 Create Date: 2024-03-19 15:30:44.425436
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 from sqlalchemy.dialects import postgresql

backend/alembic/versions/177de57c21c9_display_custom_llm_models.py

@@ -5,6 +5,7 @@ Revises: 4ee1287bd26a
 Create Date: 2024-11-21 11:49:04.488677
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 from sqlalchemy.dialects import postgresql

backend/alembic/versions/1a03d2c2856b_add_indexes_to_document__tag.py

@@ -5,6 +5,7 @@ Revises: 9c00a2bccb83
 Create Date: 2025-02-18 10:45:13.957807
 
 """
+
 from alembic import op
 
 # revision identifiers, used by Alembic.

backend/alembic/versions/1b10e1fda030_add_additional_data_to_notifications.py

@@ -5,6 +5,7 @@ Revises: 6756efa39ada
 Create Date: 2024-10-15 19:26:44.071259
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 from sqlalchemy.dialects import postgresql

backend/alembic/versions/1b8206b29c5d_add_user_delete_cascades.py

@@ -5,6 +5,7 @@ Revises: 35e6853a51d5
 Create Date: 2024-09-18 11:48:59.418726
 
 """
+
 from alembic import op
 
 

backend/alembic/versions/213fd978c6d8_notifications.py

@@ -5,6 +5,7 @@ Revises: 5fc1f54cc252
 Create Date: 2024-08-10 11:13:36.070790
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 

backend/alembic/versions/238b84885828_add_foreign_key_to_user__external_user_.py

@@ -0,0 +1,45 @@
+"""Add foreign key to user__external_user_group_id
+
+Revision ID: 238b84885828
+Revises: a7688ab35c45
+Create Date: 2025-05-19 17:15:33.424584
+
+"""
+
+from alembic import op
+
+
+# revision identifiers, used by Alembic.
+revision = "238b84885828"
+down_revision = "a7688ab35c45"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    # First, clean up any entries that don't have a valid cc_pair_id
+    op.execute(
+        """
+        DELETE FROM user__external_user_group_id
+        WHERE cc_pair_id NOT IN (SELECT id FROM connector_credential_pair)
+        """
+    )
+
+    # Add foreign key constraint with cascade delete
+    op.create_foreign_key(
+        "fk_user__external_user_group_id_cc_pair_id",
+        "user__external_user_group_id",
+        "connector_credential_pair",
+        ["cc_pair_id"],
+        ["id"],
+        ondelete="CASCADE",
+    )
+
+
+def downgrade() -> None:
+    # Drop the foreign key constraint
+    op.drop_constraint(
+        "fk_user__external_user_group_id_cc_pair_id",
+        "user__external_user_group_id",
+        type_="foreignkey",
+    )

backend/alembic/versions/23957775e5f5_remove_feedback_foreignkey_constraint.py

@@ -5,6 +5,7 @@ Revises: bc9771dccadf
 Create Date: 2024-06-27 16:04:51.480437
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 

backend/alembic/versions/2666d766cb9b_google_oauth2.py

@@ -5,6 +5,7 @@ Revises: 6d387b3196c2
 Create Date: 2023-05-05 15:49:35.716016
 
 """
+
 import fastapi_users_db_sqlalchemy
 import sqlalchemy as sa
 from alembic import op

backend/alembic/versions/26b931506ecb_default_chosen_assistants_to_none.py

@@ -5,6 +5,7 @@ Revises: 2daa494a0851
 Create Date: 2024-11-12 13:23:29.858995
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 from sqlalchemy.dialects import postgresql

backend/alembic/versions/27c6ecc08586_permission_framework.py

@@ -5,6 +5,7 @@ Revises: 2666d766cb9b
 Create Date: 2023-05-24 18:45:17.244495
 
 """
+
 import fastapi_users_db_sqlalchemy
 import sqlalchemy as sa
 from alembic import op
@@ -143,27 +144,34 @@ def upgrade() -> None:
 
 def downgrade() -> None:
     op.execute("TRUNCATE TABLE index_attempt")
-    op.add_column(
-        "index_attempt",
-        sa.Column("input_type", sa.VARCHAR(), autoincrement=False, nullable=False),
-    )
-    op.add_column(
-        "index_attempt",
-        sa.Column("source", sa.VARCHAR(), autoincrement=False, nullable=False),
-    )
-    op.add_column(
-        "index_attempt",
-        sa.Column(
-            "connector_specific_config",
-            postgresql.JSONB(astext_type=sa.Text()),
-            autoincrement=False,
-            nullable=False,
-        ),
-    )
-
-    # Check if the constraint exists before dropping
     conn = op.get_bind()
     inspector = sa.inspect(conn)
+    existing_columns = {col["name"] for col in inspector.get_columns("index_attempt")}
+
+    if "input_type" not in existing_columns:
+        op.add_column(
+            "index_attempt",
+            sa.Column("input_type", sa.VARCHAR(), autoincrement=False, nullable=False),
+        )
+
+    if "source" not in existing_columns:
+        op.add_column(
+            "index_attempt",
+            sa.Column("source", sa.VARCHAR(), autoincrement=False, nullable=False),
+        )
+
+    if "connector_specific_config" not in existing_columns:
+        op.add_column(
+            "index_attempt",
+            sa.Column(
+                "connector_specific_config",
+                postgresql.JSONB(astext_type=sa.Text()),
+                autoincrement=False,
+                nullable=False,
+            ),
+        )
+
+    # Check if the constraint exists before dropping
     constraints = inspector.get_foreign_keys("index_attempt")
 
     if any(
@@ -182,8 +190,12 @@ def downgrade() -> None:
             "fk_index_attempt_connector_id", "index_attempt", type_="foreignkey"
         )
 
-    op.drop_column("index_attempt", "credential_id")
-    op.drop_column("index_attempt", "connector_id")
-    op.drop_table("connector_credential_pair")
-    op.drop_table("credential")
-    op.drop_table("connector")
+    if "credential_id" in existing_columns:
+        op.drop_column("index_attempt", "credential_id")
+
+    if "connector_id" in existing_columns:
+        op.drop_column("index_attempt", "connector_id")
+
+    op.execute("DROP TABLE IF EXISTS connector_credential_pair CASCADE")
+    op.execute("DROP TABLE IF EXISTS credential CASCADE")
+    op.execute("DROP TABLE IF EXISTS connector CASCADE")

backend/alembic/versions/2955778aa44c_add_chunk_count_to_document.py

@@ -5,6 +5,7 @@ Revises: c0aab6edb6dd
 Create Date: 2025-01-04 11:39:43.268612
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 

backend/alembic/versions/2b75d0a8ffcb_user_file_schema_cleanup.py

@@ -0,0 +1,228 @@
+"""Migration 6: User file schema cleanup
+
+Revision ID: 2b75d0a8ffcb
+Revises: 3a78dba1080a
+Create Date: 2025-09-22 10:09:26.375377
+
+This migration removes legacy columns and tables after data migration is complete.
+It should only be run after verifying all data has been successfully migrated.
+"""
+
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy import text
+import logging
+import fastapi_users_db_sqlalchemy
+
+logger = logging.getLogger("alembic.runtime.migration")
+
+# revision identifiers, used by Alembic.
+revision = "2b75d0a8ffcb"
+down_revision = "3a78dba1080a"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    """Remove legacy columns and tables."""
+
+    bind = op.get_bind()
+    inspector = sa.inspect(bind)
+
+    logger.info("Starting schema cleanup...")
+
+    # === Step 1: Verify data migration is complete ===
+    logger.info("Verifying data migration completion...")
+
+    # Check if any chat sessions still have folder_id references
+    chat_session_columns = [
+        col["name"] for col in inspector.get_columns("chat_session")
+    ]
+    if "folder_id" in chat_session_columns:
+        orphaned_count = bind.execute(
+            text(
+                """
+            SELECT COUNT(*) FROM chat_session
+            WHERE folder_id IS NOT NULL AND project_id IS NULL
+        """
+            )
+        ).scalar_one()
+
+        if orphaned_count > 0:
+            logger.warning(
+                f"WARNING: {orphaned_count} chat_session records still have "
+                f"folder_id without project_id. Proceeding anyway."
+            )
+
+    # === Step 2: Drop chat_session.folder_id ===
+    if "folder_id" in chat_session_columns:
+        logger.info("Dropping chat_session.folder_id...")
+
+        # Drop foreign key constraint first
+        op.execute(
+            "ALTER TABLE chat_session DROP CONSTRAINT IF EXISTS chat_session_chat_folder_fk"
+        )
+        op.execute(
+            "ALTER TABLE chat_session DROP CONSTRAINT IF EXISTS chat_session_folder_fk"
+        )
+
+        # Drop the column
+        op.drop_column("chat_session", "folder_id")
+        logger.info("Dropped chat_session.folder_id")
+
+    # === Step 3: Drop persona__user_folder table ===
+    if "persona__user_folder" in inspector.get_table_names():
+        logger.info("Dropping persona__user_folder table...")
+
+        # Check for any remaining data
+        remaining = bind.execute(
+            text("SELECT COUNT(*) FROM persona__user_folder")
+        ).scalar_one()
+
+        if remaining > 0:
+            logger.warning(
+                f"WARNING: Dropping persona__user_folder with {remaining} records"
+            )
+
+        op.drop_table("persona__user_folder")
+        logger.info("Dropped persona__user_folder table")
+
+    # === Step 4: Drop chat_folder table ===
+    if "chat_folder" in inspector.get_table_names():
+        logger.info("Dropping chat_folder table...")
+
+        # Check for any remaining data
+        remaining = bind.execute(text("SELECT COUNT(*) FROM chat_folder")).scalar_one()
+
+        if remaining > 0:
+            logger.warning(f"WARNING: Dropping chat_folder with {remaining} records")
+
+        op.drop_table("chat_folder")
+        logger.info("Dropped chat_folder table")
+
+    # === Step 5: Drop user_file legacy columns ===
+    user_file_columns = [col["name"] for col in inspector.get_columns("user_file")]
+
+    # Drop folder_id
+    if "folder_id" in user_file_columns:
+        logger.info("Dropping user_file.folder_id...")
+        op.drop_column("user_file", "folder_id")
+        logger.info("Dropped user_file.folder_id")
+
+    # Drop cc_pair_id (already handled in migration 5, but be sure)
+    if "cc_pair_id" in user_file_columns:
+        logger.info("Dropping user_file.cc_pair_id...")
+
+        # Drop any remaining foreign key constraints
+        bind.execute(
+            text(
+                """
+            DO $$
+            DECLARE r RECORD;
+            BEGIN
+              FOR r IN (
+                SELECT conname
+                FROM pg_constraint c
+                JOIN pg_class t ON c.conrelid = t.oid
+                WHERE c.contype = 'f'
+                  AND t.relname = 'user_file'
+                  AND EXISTS (
+                    SELECT 1 FROM pg_attribute a
+                    WHERE a.attrelid = t.oid
+                    AND a.attname = 'cc_pair_id'
+                  )
+              ) LOOP
+                EXECUTE format('ALTER TABLE user_file DROP CONSTRAINT IF EXISTS %I', r.conname);
+              END LOOP;
+            END$$;
+        """
+            )
+        )
+
+        op.drop_column("user_file", "cc_pair_id")
+        logger.info("Dropped user_file.cc_pair_id")
+
+    # === Step 6: Clean up any remaining constraints ===
+    logger.info("Cleaning up remaining constraints...")
+
+    # Drop any unique constraints on removed columns
+    op.execute(
+        "ALTER TABLE user_file DROP CONSTRAINT IF EXISTS user_file_cc_pair_id_key"
+    )
+
+    logger.info("Migration 6 (schema cleanup) completed successfully")
+    logger.info("Legacy schema has been fully removed")
+
+
+def downgrade() -> None:
+    """Recreate dropped columns and tables (structure only, no data)."""
+
+    bind = op.get_bind()
+    inspector = sa.inspect(bind)
+
+    logger.warning("Downgrading schema cleanup - recreating structure only, no data!")
+
+    # Recreate user_file columns
+    if "user_file" in inspector.get_table_names():
+        columns = [col["name"] for col in inspector.get_columns("user_file")]
+
+        if "cc_pair_id" not in columns:
+            op.add_column(
+                "user_file", sa.Column("cc_pair_id", sa.Integer(), nullable=True)
+            )
+
+        if "folder_id" not in columns:
+            op.add_column(
+                "user_file", sa.Column("folder_id", sa.Integer(), nullable=True)
+            )
+
+    # Recreate persona__user_folder table
+    if "persona__user_folder" not in inspector.get_table_names():
+        op.create_table(
+            "persona__user_folder",
+            sa.Column("persona_id", sa.Integer(), nullable=False),
+            sa.Column("user_folder_id", sa.Integer(), nullable=False),
+            sa.PrimaryKeyConstraint("persona_id", "user_folder_id"),
+            sa.ForeignKeyConstraint(["persona_id"], ["persona.id"]),
+            sa.ForeignKeyConstraint(["user_folder_id"], ["user_project.id"]),
+        )
+
+    # Recreate chat_folder table and related structures
+    if "chat_folder" not in inspector.get_table_names():
+        op.create_table(
+            "chat_folder",
+            sa.Column("id", sa.Integer(), nullable=False),
+            sa.Column(
+                "user_id",
+                fastapi_users_db_sqlalchemy.generics.GUID(),
+                nullable=True,
+            ),
+            sa.Column("name", sa.String(), nullable=True),
+            sa.Column("display_priority", sa.Integer(), nullable=False),
+            sa.ForeignKeyConstraint(
+                ["user_id"],
+                ["user.id"],
+                name="chat_folder_user_id_fkey",
+            ),
+            sa.PrimaryKeyConstraint("id"),
+        )
+
+    # Add folder_id back to chat_session
+    if "chat_session" in inspector.get_table_names():
+        columns = [col["name"] for col in inspector.get_columns("chat_session")]
+        if "folder_id" not in columns:
+            op.add_column(
+                "chat_session", sa.Column("folder_id", sa.Integer(), nullable=True)
+            )
+
+            # Add foreign key if chat_folder exists
+            if "chat_folder" in inspector.get_table_names():
+                op.create_foreign_key(
+                    "chat_session_chat_folder_fk",
+                    "chat_session",
+                    "chat_folder",
+                    ["folder_id"],
+                    ["id"],
+                )
+
+    logger.info("Downgrade completed - structure recreated but data is lost")

backend/alembic/versions/2cdeff6d8c93_set_built_in_to_default.py

@@ -5,6 +5,7 @@ Revises: f5437cc136c5
 Create Date: 2025-02-11 14:57:51.308775
 
 """
+
 from alembic import op
 
 

backend/alembic/versions/2d2304e27d8c_add_above_below_to_persona.py

@@ -5,6 +5,7 @@ Revises: 4b08d97e175a
 Create Date: 2024-08-21 19:15:15.762948
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 

backend/alembic/versions/2daa494a0851_add_group_sync_time.py

@@ -5,6 +5,7 @@ Revises: c0fd6e4da83a
 Create Date: 2024-11-11 10:57:22.991157
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 

backend/alembic/versions/2f80c6a2550f_add_chat_session_specific_temperature_.py

@@ -5,6 +5,7 @@ Revises: 33ea50e88f24
 Create Date: 2025-01-31 10:30:27.289646
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 

backend/alembic/versions/2f95e36923e6_add_indexing_coordination.py

@@ -0,0 +1,115 @@
+"""add_indexing_coordination
+
+Revision ID: 2f95e36923e6
+Revises: 0816326d83aa
+Create Date: 2025-07-10 16:17:57.762182
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "2f95e36923e6"
+down_revision = "0816326d83aa"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    # Add database-based coordination fields (replacing Redis fencing)
+    op.add_column(
+        "index_attempt", sa.Column("celery_task_id", sa.String(), nullable=True)
+    )
+    op.add_column(
+        "index_attempt",
+        sa.Column(
+            "cancellation_requested",
+            sa.Boolean(),
+            nullable=False,
+            server_default="false",
+        ),
+    )
+
+    # Add batch coordination fields (replacing FileStore state)
+    op.add_column(
+        "index_attempt", sa.Column("total_batches", sa.Integer(), nullable=True)
+    )
+    op.add_column(
+        "index_attempt",
+        sa.Column(
+            "completed_batches", sa.Integer(), nullable=False, server_default="0"
+        ),
+    )
+    op.add_column(
+        "index_attempt",
+        sa.Column(
+            "total_failures_batch_level",
+            sa.Integer(),
+            nullable=False,
+            server_default="0",
+        ),
+    )
+    op.add_column(
+        "index_attempt",
+        sa.Column("total_chunks", sa.Integer(), nullable=False, server_default="0"),
+    )
+
+    # Progress tracking for stall detection
+    op.add_column(
+        "index_attempt",
+        sa.Column("last_progress_time", sa.DateTime(timezone=True), nullable=True),
+    )
+    op.add_column(
+        "index_attempt",
+        sa.Column(
+            "last_batches_completed_count",
+            sa.Integer(),
+            nullable=False,
+            server_default="0",
+        ),
+    )
+
+    # Heartbeat tracking for worker liveness detection
+    op.add_column(
+        "index_attempt",
+        sa.Column(
+            "heartbeat_counter", sa.Integer(), nullable=False, server_default="0"
+        ),
+    )
+    op.add_column(
+        "index_attempt",
+        sa.Column(
+            "last_heartbeat_value", sa.Integer(), nullable=False, server_default="0"
+        ),
+    )
+    op.add_column(
+        "index_attempt",
+        sa.Column("last_heartbeat_time", sa.DateTime(timezone=True), nullable=True),
+    )
+
+    # Add index for coordination queries
+    op.create_index(
+        "ix_index_attempt_active_coordination",
+        "index_attempt",
+        ["connector_credential_pair_id", "search_settings_id", "status"],
+    )
+
+
+def downgrade() -> None:
+    # Remove the new index
+    op.drop_index("ix_index_attempt_active_coordination", table_name="index_attempt")
+
+    # Remove the new columns
+    op.drop_column("index_attempt", "last_batches_completed_count")
+    op.drop_column("index_attempt", "last_progress_time")
+    op.drop_column("index_attempt", "last_heartbeat_time")
+    op.drop_column("index_attempt", "last_heartbeat_value")
+    op.drop_column("index_attempt", "heartbeat_counter")
+    op.drop_column("index_attempt", "total_chunks")
+    op.drop_column("index_attempt", "total_failures_batch_level")
+    op.drop_column("index_attempt", "completed_batches")
+    op.drop_column("index_attempt", "total_batches")
+    op.drop_column("index_attempt", "cancellation_requested")
+    op.drop_column("index_attempt", "celery_task_id")

backend/alembic/versions/30c1d5744104_persona_datetime_aware.py

@@ -5,6 +5,7 @@ Revises: 7f99be1cb9f5
 Create Date: 2023-10-16 23:21:01.283424
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 

backend/alembic/versions/325975216eb3_add_icon_color_and_icon_shape_to_persona.py

@@ -5,6 +5,7 @@ Revises: 91ffac7e65b3
 Create Date: 2024-07-24 21:29:31.784562
 
 """
+
 import random
 from alembic import op
 import sqlalchemy as sa

backend/alembic/versions/33cb72ea4d80_single_tool_call_per_message.py

@@ -5,6 +5,7 @@ Revises: 5b29123cd710
 Create Date: 2024-11-01 12:51:01.535003
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 

backend/alembic/versions/33ea50e88f24_foreign_key_input_prompts.py

@@ -5,6 +5,7 @@ Revises: a6df6b88ef81
 Create Date: 2025-01-29 10:54:22.141765
 
 """
+
 from alembic import op
 
 

backend/alembic/versions/351faebd379d_add_curator_fields.py

@@ -5,6 +5,7 @@ Revises: ee3f4b47fad5
 Create Date: 2024-08-15 22:37:08.397052
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 

backend/alembic/versions/35e518e0ddf4_properly_cascade.py

@@ -5,6 +5,7 @@ Revises: 91a0a4d62b14
 Create Date: 2024-09-20 21:24:04.891018
 
 """
+
 from alembic import op
 
 

backend/alembic/versions/35e6853a51d5_server_default_chosen_assistants.py

@@ -5,6 +5,7 @@ Revises: c99d76fcd298
 Create Date: 2024-09-13 13:20:32.885317
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 from sqlalchemy.dialects import postgresql

backend/alembic/versions/369644546676_add_composite_index_for_index_attempt_.py

@@ -5,6 +5,7 @@ Revises: 2955778aa44c
 Create Date: 2025-01-08 15:38:17.224380
 
 """
+
 from alembic import op
 from sqlalchemy import text
 

backend/alembic/versions/36e9220ab794_update_kg_trigger_functions.py

@@ -0,0 +1,136 @@
+"""update_kg_trigger_functions
+
+Revision ID: 36e9220ab794
+Revises: c9e2cd766c29
+Create Date: 2025-06-22 17:33:25.833733
+
+"""
+
+from alembic import op
+from sqlalchemy.orm import Session
+from sqlalchemy import text
+from shared_configs.configs import POSTGRES_DEFAULT_SCHEMA
+
+# revision identifiers, used by Alembic.
+revision = "36e9220ab794"
+down_revision = "c9e2cd766c29"
+branch_labels = None
+depends_on = None
+
+
+def _get_tenant_contextvar(session: Session) -> str:
+    """Get the current schema for the migration"""
+    current_tenant = session.execute(text("SELECT current_schema()")).scalar()
+    if isinstance(current_tenant, str):
+        return current_tenant
+    else:
+        raise ValueError("Current tenant is not a string")
+
+
+def upgrade() -> None:
+
+    bind = op.get_bind()
+    session = Session(bind=bind)
+
+    # Create kg_entity trigger to update kg_entity.name and its trigrams
+    tenant_id = _get_tenant_contextvar(session)
+    alphanum_pattern = r"[^a-z0-9]+"
+    truncate_length = 1000
+    function = "update_kg_entity_name"
+    op.execute(
+        text(
+            f"""
+            CREATE OR REPLACE FUNCTION "{tenant_id}".{function}()
+            RETURNS TRIGGER AS $$
+            DECLARE
+                name text;
+                cleaned_name text;
+            BEGIN
+                -- Set name to semantic_id if document_id is not NULL
+                IF NEW.document_id IS NOT NULL THEN
+                    SELECT lower(semantic_id) INTO name
+                    FROM "{tenant_id}".document
+                    WHERE id = NEW.document_id;
+                ELSE
+                    name = lower(NEW.name);
+                END IF;
+
+                -- Clean name and truncate if too long
+                cleaned_name = regexp_replace(
+                    name,
+                    '{alphanum_pattern}', '', 'g'
+                );
+                IF length(cleaned_name) > {truncate_length} THEN
+                    cleaned_name = left(cleaned_name, {truncate_length});
+                END IF;
+
+                -- Set name and name trigrams
+                NEW.name = name;
+                NEW.name_trigrams = {POSTGRES_DEFAULT_SCHEMA}.show_trgm(cleaned_name);
+                RETURN NEW;
+            END;
+            $$ LANGUAGE plpgsql;
+            """
+        )
+    )
+    trigger = f"{function}_trigger"
+    op.execute(f'DROP TRIGGER IF EXISTS {trigger} ON "{tenant_id}".kg_entity')
+    op.execute(
+        f"""
+        CREATE TRIGGER {trigger}
+            BEFORE INSERT OR UPDATE OF name
+            ON "{tenant_id}".kg_entity
+            FOR EACH ROW
+            EXECUTE FUNCTION "{tenant_id}".{function}();
+        """
+    )
+
+    # Create kg_entity trigger to update kg_entity.name and its trigrams
+    function = "update_kg_entity_name_from_doc"
+    op.execute(
+        text(
+            f"""
+            CREATE OR REPLACE FUNCTION "{tenant_id}".{function}()
+            RETURNS TRIGGER AS $$
+            DECLARE
+                doc_name text;
+                cleaned_name text;
+            BEGIN
+                doc_name = lower(NEW.semantic_id);
+
+                -- Clean name and truncate if too long
+                cleaned_name = regexp_replace(
+                    doc_name,
+                    '{alphanum_pattern}', '', 'g'
+                );
+                IF length(cleaned_name) > {truncate_length} THEN
+                    cleaned_name = left(cleaned_name, {truncate_length});
+                END IF;
+
+                -- Set name and name trigrams for all entities referencing this document
+                UPDATE "{tenant_id}".kg_entity
+                SET
+                    name = doc_name,
+                    name_trigrams = {POSTGRES_DEFAULT_SCHEMA}.show_trgm(cleaned_name)
+                WHERE document_id = NEW.id;
+                RETURN NEW;
+            END;
+            $$ LANGUAGE plpgsql;
+            """
+        )
+    )
+    trigger = f"{function}_trigger"
+    op.execute(f'DROP TRIGGER IF EXISTS {trigger} ON "{tenant_id}".document')
+    op.execute(
+        f"""
+        CREATE TRIGGER {trigger}
+            AFTER UPDATE OF semantic_id
+            ON "{tenant_id}".document
+            FOR EACH ROW
+            EXECUTE FUNCTION "{tenant_id}".{function}();
+        """
+    )
+
+
+def downgrade() -> None:
+    pass

backend/alembic/versions/3781a5eb12cb_add_chunk_stats_table.py

@@ -5,6 +5,7 @@ Revises: df46c75b714e
 Create Date: 2025-03-10 10:02:30.586666
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 

backend/alembic/versions/3879338f8ba1_add_tool_table.py

@@ -5,6 +5,7 @@ Revises: f1c6478c3fd8
 Create Date: 2024-05-11 16:11:23.718084
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 

backend/alembic/versions/38eda64af7fe_add_chat_session_sharing.py

@@ -5,6 +5,7 @@ Revises: 776b3bbe9092
 Create Date: 2024-03-27 19:41:29.073594
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 

backend/alembic/versions/3934b1bc7b62_update_github_connector_repo_name_to_.py

@@ -5,6 +5,7 @@ Revises: b7c2b63c4a03
 Create Date: 2025-03-05 10:50:30.516962
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 import json

backend/alembic/versions/3a78dba1080a_user_file_legacy_data_cleanup.py

@@ -0,0 +1,298 @@
+"""Migration 5: User file legacy data cleanup
+
+Revision ID: 3a78dba1080a
+Revises: 7cc3fcc116c1
+Create Date: 2025-09-22 10:04:27.986294
+
+This migration removes legacy user-file documents and connector_credential_pairs.
+It performs bulk deletions of obsolete data after the UUID migration.
+"""
+
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy.dialects import postgresql as psql
+from sqlalchemy import text
+import logging
+from typing import List
+import uuid
+
+logger = logging.getLogger("alembic.runtime.migration")
+
+# revision identifiers, used by Alembic.
+revision = "3a78dba1080a"
+down_revision = "7cc3fcc116c1"
+branch_labels = None
+depends_on = None
+
+
+def batch_delete(
+    bind: sa.engine.Connection,
+    table_name: str,
+    id_column: str,
+    ids: List[str | int | uuid.UUID],
+    batch_size: int = 1000,
+    id_type: str = "int",
+) -> int:
+    """Delete records in batches to avoid memory issues and timeouts."""
+    total_count = len(ids)
+    if total_count == 0:
+        return 0
+
+    logger.info(
+        f"Starting batch deletion of {total_count} records from {table_name}..."
+    )
+
+    # Determine appropriate ARRAY type
+    if id_type == "uuid":
+        array_type = psql.ARRAY(psql.UUID(as_uuid=True))
+    elif id_type == "int":
+        array_type = psql.ARRAY(sa.Integer())
+    else:
+        array_type = psql.ARRAY(sa.String())
+
+    total_deleted = 0
+    failed_batches = []
+
+    for i in range(0, total_count, batch_size):
+        batch_ids = ids[i : i + batch_size]
+        try:
+            stmt = text(
+                f"DELETE FROM {table_name} WHERE {id_column} = ANY(:ids)"
+            ).bindparams(sa.bindparam("ids", value=batch_ids, type_=array_type))
+            result = bind.execute(stmt)
+            total_deleted += result.rowcount
+
+            # Log progress every 10 batches or at completion
+            batch_num = (i // batch_size) + 1
+            if batch_num % 10 == 0 or i + batch_size >= total_count:
+                logger.info(
+                    f"  Deleted {min(i + batch_size, total_count)}/{total_count} records "
+                    f"({total_deleted} actual) from {table_name}"
+                )
+        except Exception as e:
+            logger.error(f"Failed to delete batch {(i // batch_size) + 1}: {e}")
+            failed_batches.append((i, min(i + batch_size, total_count)))
+
+    if failed_batches:
+        logger.warning(
+            f"Failed to delete {len(failed_batches)} batches from {table_name}. "
+            f"Total deleted: {total_deleted}/{total_count}"
+        )
+        # Fail the migration to avoid silently succeeding on partial cleanup
+        raise RuntimeError(
+            f"Batch deletion failed for {table_name}: "
+            f"{len(failed_batches)} failed batches out of "
+            f"{(total_count + batch_size - 1) // batch_size}."
+        )
+
+    return total_deleted
+
+
+def upgrade() -> None:
+    """Remove legacy user-file documents and connector_credential_pairs."""
+
+    bind = op.get_bind()
+    inspector = sa.inspect(bind)
+
+    logger.info("Starting legacy data cleanup...")
+
+    # === Step 1: Identify and delete user-file documents ===
+    logger.info("Identifying user-file documents to delete...")
+
+    # Get document IDs to delete
+    doc_rows = bind.execute(
+        text(
+            """
+        SELECT DISTINCT dcc.id AS document_id
+        FROM document_by_connector_credential_pair dcc
+        JOIN connector_credential_pair u
+          ON u.connector_id = dcc.connector_id
+         AND u.credential_id = dcc.credential_id
+        WHERE u.is_user_file IS TRUE
+    """
+        )
+    ).fetchall()
+
+    doc_ids = [r[0] for r in doc_rows]
+
+    if doc_ids:
+        logger.info(f"Found {len(doc_ids)} user-file documents to delete")
+
+        # Delete dependent rows first
+        tables_to_clean = [
+            ("document_retrieval_feedback", "document_id"),
+            ("document__tag", "document_id"),
+            ("chunk_stats", "document_id"),
+        ]
+
+        for table_name, column_name in tables_to_clean:
+            if table_name in inspector.get_table_names():
+                # document_id is a string in these tables
+                deleted = batch_delete(
+                    bind, table_name, column_name, doc_ids, id_type="str"
+                )
+                logger.info(f"Deleted {deleted} records from {table_name}")
+
+        # Delete document_by_connector_credential_pair entries
+        deleted = batch_delete(
+            bind, "document_by_connector_credential_pair", "id", doc_ids, id_type="str"
+        )
+        logger.info(f"Deleted {deleted} document_by_connector_credential_pair records")
+
+        # Delete documents themselves
+        deleted = batch_delete(bind, "document", "id", doc_ids, id_type="str")
+        logger.info(f"Deleted {deleted} document records")
+    else:
+        logger.info("No user-file documents found to delete")
+
+    # === Step 2: Clean up user-file connector_credential_pairs ===
+    logger.info("Cleaning up user-file connector_credential_pairs...")
+
+    # Get cc_pair IDs
+    cc_pair_rows = bind.execute(
+        text(
+            """
+        SELECT id AS cc_pair_id
+        FROM connector_credential_pair
+        WHERE is_user_file IS TRUE
+    """
+        )
+    ).fetchall()
+
+    cc_pair_ids = [r[0] for r in cc_pair_rows]
+
+    if cc_pair_ids:
+        logger.info(
+            f"Found {len(cc_pair_ids)} user-file connector_credential_pairs to clean up"
+        )
+
+        # Delete related records
+        # Clean child tables first to satisfy foreign key constraints,
+        # then the parent tables
+        tables_to_clean = [
+            ("index_attempt_errors", "connector_credential_pair_id"),
+            ("index_attempt", "connector_credential_pair_id"),
+            ("background_error", "cc_pair_id"),
+            ("document_set__connector_credential_pair", "connector_credential_pair_id"),
+            ("user_group__connector_credential_pair", "cc_pair_id"),
+        ]
+
+        for table_name, column_name in tables_to_clean:
+            if table_name in inspector.get_table_names():
+                deleted = batch_delete(
+                    bind, table_name, column_name, cc_pair_ids, id_type="int"
+                )
+                logger.info(f"Deleted {deleted} records from {table_name}")
+
+    # === Step 3: Identify connectors and credentials to delete ===
+    logger.info("Identifying orphaned connectors and credentials...")
+
+    # Get connectors used only by user-file cc_pairs
+    connector_rows = bind.execute(
+        text(
+            """
+        SELECT DISTINCT ccp.connector_id
+        FROM connector_credential_pair ccp
+        WHERE ccp.is_user_file IS TRUE
+          AND ccp.connector_id != 0  -- Exclude system default
+          AND NOT EXISTS (
+            SELECT 1
+            FROM connector_credential_pair c2
+            WHERE c2.connector_id = ccp.connector_id
+              AND c2.is_user_file IS NOT TRUE
+          )
+    """
+        )
+    ).fetchall()
+
+    userfile_only_connector_ids = [r[0] for r in connector_rows]
+
+    # Get credentials used only by user-file cc_pairs
+    credential_rows = bind.execute(
+        text(
+            """
+        SELECT DISTINCT ccp.credential_id
+        FROM connector_credential_pair ccp
+        WHERE ccp.is_user_file IS TRUE
+          AND ccp.credential_id != 0  -- Exclude public/default
+          AND NOT EXISTS (
+            SELECT 1
+            FROM connector_credential_pair c2
+            WHERE c2.credential_id = ccp.credential_id
+              AND c2.is_user_file IS NOT TRUE
+          )
+    """
+        )
+    ).fetchall()
+
+    userfile_only_credential_ids = [r[0] for r in credential_rows]
+
+    # === Step 4: Delete the cc_pairs themselves ===
+    if cc_pair_ids:
+        # Remove FK dependency from user_file first
+        bind.execute(
+            text(
+                """
+            DO $$
+            DECLARE r RECORD;
+            BEGIN
+              FOR r IN (
+                SELECT conname
+                FROM pg_constraint c
+                JOIN pg_class t ON c.conrelid = t.oid
+                JOIN pg_class ft ON c.confrelid = ft.oid
+                WHERE c.contype = 'f'
+                  AND t.relname = 'user_file'
+                  AND ft.relname = 'connector_credential_pair'
+              ) LOOP
+                EXECUTE format('ALTER TABLE user_file DROP CONSTRAINT IF EXISTS %I', r.conname);
+              END LOOP;
+            END$$;
+        """
+            )
+        )
+
+        # Delete cc_pairs
+        deleted = batch_delete(
+            bind, "connector_credential_pair", "id", cc_pair_ids, id_type="int"
+        )
+        logger.info(f"Deleted {deleted} connector_credential_pair records")
+
+    # === Step 5: Delete orphaned connectors ===
+    if userfile_only_connector_ids:
+        deleted = batch_delete(
+            bind, "connector", "id", userfile_only_connector_ids, id_type="int"
+        )
+        logger.info(f"Deleted {deleted} orphaned connector records")
+
+    # === Step 6: Delete orphaned credentials ===
+    if userfile_only_credential_ids:
+        # Clean up credential__user_group mappings first
+        deleted = batch_delete(
+            bind,
+            "credential__user_group",
+            "credential_id",
+            userfile_only_credential_ids,
+            id_type="int",
+        )
+        logger.info(f"Deleted {deleted} credential__user_group records")
+
+        # Delete credentials
+        deleted = batch_delete(
+            bind, "credential", "id", userfile_only_credential_ids, id_type="int"
+        )
+        logger.info(f"Deleted {deleted} orphaned credential records")
+
+    logger.info("Migration 5 (legacy data cleanup) completed successfully")
+
+
+def downgrade() -> None:
+    """Cannot restore deleted data - requires backup restoration."""
+
+    logger.error("CRITICAL: Downgrading data cleanup cannot restore deleted data!")
+    logger.error("Data restoration requires backup files or database backup.")
+
+    # raise NotImplementedError(
+    #     "Downgrade of legacy data cleanup is not supported. "
+    #     "Deleted data must be restored from backups."
+    # )

backend/alembic/versions/3b25685ff73c_move_is_public_to_cc_pair.py

@@ -5,6 +5,7 @@ Revises: e0a68a81d434
 Create Date: 2023-10-05 18:47:09.582849
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 

backend/alembic/versions/3bd4c84fe72f_improved_index.py

@@ -5,6 +5,7 @@ Revises: 8f43500ee275
 Create Date: 2025-02-26 13:07:56.217791
 
 """
+
 from alembic import op
 
 
@@ -20,22 +21,14 @@ depends_on = None
 # an outage by creating an index without using CONCURRENTLY. This migration:
 #
 # 1. Creates more efficient full-text search capabilities using tsvector columns and GIN indexes
-# 2. Uses CONCURRENTLY for all index creation to prevent table locking
-# 3. Explicitly manages transactions with COMMIT statements to allow CONCURRENTLY to work
-# (see: https://www.postgresql.org/docs/9.4/sql-createindex.html#SQL-CREATEINDEX-CONCURRENTLY)
-# (see: https://github.com/sqlalchemy/alembic/issues/277)
-# 4. Adds indexes to both chat_message and chat_session tables for comprehensive search
+# 2. Adds indexes to both chat_message and chat_session tables for comprehensive search
+# 3. Note: CONCURRENTLY was removed due to operational issues
 
 
 def upgrade() -> None:
     # First, drop any existing indexes to avoid conflicts
-    op.execute("COMMIT")
-    op.execute("DROP INDEX CONCURRENTLY IF EXISTS idx_chat_message_tsv;")
-
-    op.execute("COMMIT")
-    op.execute("DROP INDEX CONCURRENTLY IF EXISTS idx_chat_session_desc_tsv;")
-
-    op.execute("COMMIT")
+    op.execute("DROP INDEX IF EXISTS idx_chat_message_tsv;")
+    op.execute("DROP INDEX IF EXISTS idx_chat_session_desc_tsv;")
     op.execute("DROP INDEX IF EXISTS idx_chat_message_message_lower;")
 
     # Drop existing columns if they exist
@@ -51,12 +44,9 @@ def upgrade() -> None:
         """
     )
 
-    # Commit the current transaction before creating concurrent indexes
-    op.execute("COMMIT")
-
     op.execute(
         """
-        CREATE INDEX CONCURRENTLY IF NOT EXISTS idx_chat_message_tsv
+        CREATE INDEX IF NOT EXISTS idx_chat_message_tsv
         ON chat_message
         USING GIN (message_tsv)
         """
@@ -71,12 +61,9 @@ def upgrade() -> None:
         """
     )
 
-    # Commit again before creating the second concurrent index
-    op.execute("COMMIT")
-
     op.execute(
         """
-        CREATE INDEX CONCURRENTLY IF NOT EXISTS idx_chat_session_desc_tsv
+        CREATE INDEX IF NOT EXISTS idx_chat_session_desc_tsv
         ON chat_session
         USING GIN (description_tsv)
         """
@@ -84,12 +71,9 @@ def upgrade() -> None:
 
 
 def downgrade() -> None:
-    # Drop the indexes first (use CONCURRENTLY for dropping too)
-    op.execute("COMMIT")
-    op.execute("DROP INDEX CONCURRENTLY IF EXISTS idx_chat_message_tsv;")
-
-    op.execute("COMMIT")
-    op.execute("DROP INDEX CONCURRENTLY IF EXISTS idx_chat_session_desc_tsv;")
+    # Drop the indexes first
+    op.execute("DROP INDEX IF EXISTS idx_chat_message_tsv;")
+    op.execute("DROP INDEX IF EXISTS idx_chat_session_desc_tsv;")
 
     # Then drop the columns
     op.execute("ALTER TABLE chat_message DROP COLUMN IF EXISTS message_tsv;")

backend/alembic/versions/3c5e35aa9af0_polling_document_count.py

@@ -5,6 +5,7 @@ Revises: 27c6ecc08586
 Create Date: 2023-06-14 23:45:51.760440
 
 """
+
 import sqlalchemy as sa
 from alembic import op
 

backend/alembic/versions/3c6531f32351_add_back_input_prompts.py

@@ -5,6 +5,7 @@ Revises: aeda5f2df4f6
 Create Date: 2025-01-13 12:49:51.705235
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 import fastapi_users_db_sqlalchemy

backend/alembic/versions/3d1cca026fe8_add_oauth_config_and_user_tokens.py

@@ -0,0 +1,121 @@
+"""add_oauth_config_and_user_tokens
+
+Revision ID: 3d1cca026fe8
+Revises: c8a93a2af083
+Create Date: 2025-10-21 13:27:34.274721
+
+"""
+
+from alembic import op
+import fastapi_users_db_sqlalchemy
+import sqlalchemy as sa
+from sqlalchemy.dialects import postgresql
+
+# revision identifiers, used by Alembic.
+revision = "3d1cca026fe8"
+down_revision = "c8a93a2af083"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    # Create oauth_config table
+    op.create_table(
+        "oauth_config",
+        sa.Column("id", sa.Integer(), nullable=False),
+        sa.Column("name", sa.String(), nullable=False),
+        sa.Column("authorization_url", sa.Text(), nullable=False),
+        sa.Column("token_url", sa.Text(), nullable=False),
+        sa.Column("client_id", sa.LargeBinary(), nullable=False),
+        sa.Column("client_secret", sa.LargeBinary(), nullable=False),
+        sa.Column("scopes", postgresql.JSONB(astext_type=sa.Text()), nullable=True),
+        sa.Column(
+            "additional_params",
+            postgresql.JSONB(astext_type=sa.Text()),
+            nullable=True,
+        ),
+        sa.Column(
+            "created_at",
+            sa.DateTime(timezone=True),
+            server_default=sa.text("now()"),
+            nullable=False,
+        ),
+        sa.Column(
+            "updated_at",
+            sa.DateTime(timezone=True),
+            server_default=sa.text("now()"),
+            nullable=False,
+        ),
+        sa.PrimaryKeyConstraint("id"),
+        sa.UniqueConstraint("name"),
+    )
+
+    # Create oauth_user_token table
+    op.create_table(
+        "oauth_user_token",
+        sa.Column("id", sa.Integer(), nullable=False),
+        sa.Column("oauth_config_id", sa.Integer(), nullable=False),
+        sa.Column(
+            "user_id",
+            fastapi_users_db_sqlalchemy.generics.GUID(),
+            nullable=False,
+        ),
+        sa.Column("token_data", sa.LargeBinary(), nullable=False),
+        sa.Column(
+            "created_at",
+            sa.DateTime(timezone=True),
+            server_default=sa.text("now()"),
+            nullable=False,
+        ),
+        sa.Column(
+            "updated_at",
+            sa.DateTime(timezone=True),
+            server_default=sa.text("now()"),
+            nullable=False,
+        ),
+        sa.ForeignKeyConstraint(
+            ["oauth_config_id"], ["oauth_config.id"], ondelete="CASCADE"
+        ),
+        sa.ForeignKeyConstraint(["user_id"], ["user.id"], ondelete="CASCADE"),
+        sa.PrimaryKeyConstraint("id"),
+        sa.UniqueConstraint("oauth_config_id", "user_id", name="uq_oauth_user_token"),
+    )
+
+    # Create index on user_id for efficient user-based token lookups
+    # Note: unique constraint on (oauth_config_id, user_id) already creates
+    # an index for config-based lookups
+    op.create_index(
+        "ix_oauth_user_token_user_id",
+        "oauth_user_token",
+        ["user_id"],
+    )
+
+    # Add oauth_config_id column to tool table
+    op.add_column("tool", sa.Column("oauth_config_id", sa.Integer(), nullable=True))
+
+    # Create foreign key from tool to oauth_config
+    op.create_foreign_key(
+        "tool_oauth_config_fk",
+        "tool",
+        "oauth_config",
+        ["oauth_config_id"],
+        ["id"],
+        ondelete="SET NULL",
+    )
+
+
+def downgrade() -> None:
+    # Drop foreign key from tool to oauth_config
+    op.drop_constraint("tool_oauth_config_fk", "tool", type_="foreignkey")
+
+    # Drop oauth_config_id column from tool table
+    op.drop_column("tool", "oauth_config_id")
+
+    # Drop index on user_id
+    op.drop_index("ix_oauth_user_token_user_id", table_name="oauth_user_token")
+
+    # Drop oauth_user_token table (will cascade delete tokens)
+    op.drop_table("oauth_user_token")
+
+    # Drop oauth_config table
+    op.drop_table("oauth_config")

backend/alembic/versions/3fc5d75723b3_add_doc_metadata_field_in_document_model.py

@@ -0,0 +1,30 @@
+"""add_doc_metadata_field_in_document_model
+
+Revision ID: 3fc5d75723b3
+Revises: 2f95e36923e6
+Create Date: 2025-07-28 18:45:37.985406
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+from sqlalchemy.dialects import postgresql
+
+# revision identifiers, used by Alembic.
+revision = "3fc5d75723b3"
+down_revision = "2f95e36923e6"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.add_column(
+        "document",
+        sa.Column(
+            "doc_metadata", postgresql.JSONB(astext_type=sa.Text()), nullable=True
+        ),
+    )
+
+
+def downgrade() -> None:
+    op.drop_column("document", "doc_metadata")

backend/alembic/versions/401c1ac29467_add_tables_for_ui_based_llm_.py

@@ -5,6 +5,7 @@ Revises: 703313b75876
 Create Date: 2024-04-13 18:07:29.153817
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 from sqlalchemy.dialects import postgresql

backend/alembic/versions/40926a4dab77_reset_userfile_document_id_migrated_.py

@@ -0,0 +1,28 @@
+"""reset userfile document_id_migrated field
+
+Revision ID: 40926a4dab77
+Revises: 64bd5677aeb6
+Create Date: 2025-10-06 16:10:32.898668
+
+"""
+
+from alembic import op
+
+# revision identifiers, used by Alembic.
+revision = "40926a4dab77"
+down_revision = "64bd5677aeb6"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    # Set all existing records to not migrated
+    op.execute(
+        "UPDATE user_file SET document_id_migrated = FALSE "
+        "WHERE document_id_migrated IS DISTINCT FROM FALSE;"
+    )
+
+
+def downgrade() -> None:
+    # No-op
+    pass

backend/alembic/versions/43cbbb3f5e6a_rename_index_origin_to_index_recursively.py

@@ -5,6 +5,7 @@ Revises: e1392f05e840
 Create Date: 2024-08-01 12:38:54.466081
 
 """
+
 from alembic import op
 
 # revision identifiers, used by Alembic.

backend/alembic/versions/44f856ae2a4a_add_cloud_embedding_model.py

@@ -5,6 +5,7 @@ Revises: d716b0791ddd
 Create Date: 2024-06-28 20:01:05.927647
 
 """
+
 from alembic import op
 import sqlalchemy as sa
 

backend/alembic/versions/4505fd7302e1_added_is_internet_to_dbdoc.py

@@ -5,6 +5,7 @@ Revises: c18cdf4b497e
 Create Date: 2024-06-18 20:46:09.095034
 
 """
+
 from alembic import op
 import sqlalchemy as sa