fix(be): replace HTTPException with OnyxError in DB layer

Migrate 10 DB layer files from HTTPException to OnyxError for
consistent structured error responses. Also fixes several incorrect
status codes:

- input_prompt: 401 → 403 for authorization errors, 422 → 404 for
  not-found
- connector_credential_pair: 401 → 404 for credential not found
- feedback: 400 → 403 for authorization errors
- chat: 400 → 404 for session not found
- user_group: 400 → 403 for permission errors

Files: onyx/db/{input_prompt,users,connector_credential_pair,feedback,
persona,chat,projects}.py, onyx/db/engine/{sql_engine,async_sql_engine}.py,
ee/onyx/db/user_group.py

.github/actions/build-integration-image/action.yml

@@ -54,6 +54,7 @@ runs:
       shell: bash
       env:
         RUNS_ON_ECR_CACHE: ${{ inputs.runs-on-ecr-cache }}
+        INTEGRATION_REPOSITORY: ${{ inputs.runs-on-ecr-cache }}
         TAG: nightly-llm-it-${{ inputs.run-id }}
         CACHE_SUFFIX: ${{ steps.format-branch.outputs.cache-suffix }}
         HEAD_SHA: ${{ inputs.github-sha }}

.github/workflows/deployment.yml

@@ -426,8 +426,9 @@ jobs:
             ONYX_VERSION=${{ github.ref_name }}
             NODE_OPTIONS=--max-old-space-size=8192
           cache-from: |
-            type=registry,ref=${{ env.REGISTRY_IMAGE }}:latest
             type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:web-cache-amd64
+            type=registry,ref=${{ env.REGISTRY_IMAGE }}:edge
+            type=registry,ref=${{ env.REGISTRY_IMAGE }}:latest
           cache-to: |
             type=inline
             type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:web-cache-amd64,mode=max
@@ -499,8 +500,9 @@ jobs:
             ONYX_VERSION=${{ github.ref_name }}
             NODE_OPTIONS=--max-old-space-size=8192
           cache-from: |
-            type=registry,ref=${{ env.REGISTRY_IMAGE }}:latest
             type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:web-cache-arm64
+            type=registry,ref=${{ env.REGISTRY_IMAGE }}:edge
+            type=registry,ref=${{ env.REGISTRY_IMAGE }}:latest
           cache-to: |
             type=inline
             type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:web-cache-arm64,mode=max
@@ -646,8 +648,8 @@ jobs:
             NEXT_PUBLIC_INCLUDE_ERROR_POPUP_SUPPORT_LINK=true
             NODE_OPTIONS=--max-old-space-size=8192
           cache-from: |
-            type=registry,ref=${{ env.REGISTRY_IMAGE }}:latest
             type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:cloudweb-cache-amd64
+            type=registry,ref=${{ env.REGISTRY_IMAGE }}:latest
           cache-to: |
             type=inline
             type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:cloudweb-cache-amd64,mode=max
@@ -728,8 +730,8 @@ jobs:
             NEXT_PUBLIC_INCLUDE_ERROR_POPUP_SUPPORT_LINK=true
             NODE_OPTIONS=--max-old-space-size=8192
           cache-from: |
-            type=registry,ref=${{ env.REGISTRY_IMAGE }}:latest
             type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:cloudweb-cache-arm64
+            type=registry,ref=${{ env.REGISTRY_IMAGE }}:latest
           cache-to: |
             type=inline
             type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:cloudweb-cache-arm64,mode=max
@@ -862,8 +864,9 @@ jobs:
           build-args: |
             ONYX_VERSION=${{ github.ref_name }}
           cache-from: |
-            type=registry,ref=${{ env.REGISTRY_IMAGE }}:latest
             type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-cache-amd64
+            type=registry,ref=${{ env.REGISTRY_IMAGE }}:edge
+            type=registry,ref=${{ env.REGISTRY_IMAGE }}:latest
           cache-to: |
             type=inline
             type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-cache-amd64,mode=max
@@ -934,8 +937,9 @@ jobs:
           build-args: |
             ONYX_VERSION=${{ github.ref_name }}
           cache-from: |
-            type=registry,ref=${{ env.REGISTRY_IMAGE }}:latest
             type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-cache-arm64
+            type=registry,ref=${{ env.REGISTRY_IMAGE }}:edge
+            type=registry,ref=${{ env.REGISTRY_IMAGE }}:latest
           cache-to: |
             type=inline
             type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-cache-arm64,mode=max
@@ -1072,8 +1076,8 @@ jobs:
             ONYX_VERSION=${{ github.ref_name }}
             ENABLE_CRAFT=true
           cache-from: |
-            type=registry,ref=${{ env.REGISTRY_IMAGE }}:latest
             type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-craft-cache-amd64
+            type=registry,ref=${{ env.REGISTRY_IMAGE }}:latest
           cache-to: |
             type=inline
             type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-craft-cache-amd64,mode=max
@@ -1145,8 +1149,8 @@ jobs:
             ONYX_VERSION=${{ github.ref_name }}
             ENABLE_CRAFT=true
           cache-from: |
-            type=registry,ref=${{ env.REGISTRY_IMAGE }}:latest
             type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-craft-cache-arm64
+            type=registry,ref=${{ env.REGISTRY_IMAGE }}:latest
           cache-to: |
             type=inline
             type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:backend-craft-cache-arm64,mode=max
@@ -1287,8 +1291,9 @@ jobs:
           build-args: |
             ONYX_VERSION=${{ github.ref_name }}
           cache-from: |
-            type=registry,ref=${{ env.REGISTRY_IMAGE }}:latest
             type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:model-server-cache-amd64
+            type=registry,ref=${{ env.REGISTRY_IMAGE }}:edge
+            type=registry,ref=${{ env.REGISTRY_IMAGE }}:latest
           cache-to: |
             type=inline
             type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:model-server-cache-amd64,mode=max
@@ -1366,8 +1371,9 @@ jobs:
           build-args: |
             ONYX_VERSION=${{ github.ref_name }}
           cache-from: |
-            type=registry,ref=${{ env.REGISTRY_IMAGE }}:latest
             type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:model-server-cache-arm64
+            type=registry,ref=${{ env.REGISTRY_IMAGE }}:edge
+            type=registry,ref=${{ env.REGISTRY_IMAGE }}:latest
           cache-to: |
             type=inline
             type=registry,ref=${{ env.RUNS_ON_ECR_CACHE }}:model-server-cache-arm64,mode=max

.github/workflows/nightly-llm-provider-chat.yml

@@ -15,6 +15,10 @@ permissions:
 jobs:
   provider-chat-test:
     uses: ./.github/workflows/reusable-nightly-llm-provider-chat.yml
+    secrets: inherit
+    permissions:
+      contents: read
+      id-token: write
     with:
       openai_models: ${{ vars.NIGHTLY_LLM_OPENAI_MODELS }}
       anthropic_models: ${{ vars.NIGHTLY_LLM_ANTHROPIC_MODELS }}
@@ -25,16 +29,6 @@ jobs:
       ollama_models: ${{ vars.NIGHTLY_LLM_OLLAMA_MODELS }}
       openrouter_models: ${{ vars.NIGHTLY_LLM_OPENROUTER_MODELS }}
       strict: true
-    secrets:
-      openai_api_key: ${{ secrets.OPENAI_API_KEY }}
-      anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}
-      bedrock_api_key: ${{ secrets.BEDROCK_API_KEY }}
-      vertex_ai_custom_config_json: ${{ secrets.NIGHTLY_LLM_VERTEX_AI_CUSTOM_CONFIG_JSON }}
-      azure_api_key: ${{ secrets.AZURE_API_KEY }}
-      ollama_api_key: ${{ secrets.OLLAMA_API_KEY }}
-      openrouter_api_key: ${{ secrets.OPENROUTER_API_KEY }}
-      DOCKER_USERNAME: ${{ secrets.DOCKER_USERNAME }}
-      DOCKER_TOKEN: ${{ secrets.DOCKER_TOKEN }}
 
   notify-slack-on-failure:
     needs: [provider-chat-test]

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

@@ -160,7 +160,7 @@ jobs:
           cd deployment/docker_compose
 
           # Get list of running containers
-          containers=$(docker compose -f docker-compose.yml -f docker-compose.dev.yml -f docker-compose.opensearch.yml ps -q)
+          containers=$(docker compose -f docker-compose.yml -f docker-compose.dev.yml ps -q)
 
           # Collect logs from each container
           for container in $containers; do

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

@@ -335,7 +335,6 @@ jobs:
           # TODO(Nik): https://linear.app/onyx-app/issue/ENG-1/update-test-infra-to-use-test-license
           LICENSE_ENFORCEMENT_ENABLED=false
           CHECK_TTL_MANAGEMENT_TASK_FREQUENCY_IN_HOURS=0.001
-          USE_LIGHTWEIGHT_BACKGROUND_WORKER=false
           EOF
           fi
 
@@ -471,13 +470,13 @@ jobs:
           path: ${{ github.workspace }}/docker-compose.log
       # ------------------------------------------------------------
 
-  no-vectordb-tests:
+  onyx-lite-tests:
     needs: [build-backend-image, build-integration-image]
     runs-on:
       [
         runs-on,
         runner=4cpu-linux-arm64,
-        "run-id=${{ github.run_id }}-no-vectordb-tests",
+        "run-id=${{ github.run_id }}-onyx-lite-tests",
         "extras=ecr-cache",
       ]
     timeout-minutes: 45
@@ -495,13 +494,12 @@ jobs:
           username: ${{ secrets.DOCKER_USERNAME }}
           password: ${{ secrets.DOCKER_TOKEN }}
 
-      - name: Create .env file for no-vectordb Docker Compose
+      - name: Create .env file for Onyx Lite Docker Compose
         env:
           ECR_CACHE: ${{ env.RUNS_ON_ECR_CACHE }}
           RUN_ID: ${{ github.run_id }}
         run: |
           cat <<EOF > deployment/docker_compose/.env
-          COMPOSE_PROFILES=s3-filestore
           ENABLE_PAID_ENTERPRISE_EDITION_FEATURES=true
           LICENSE_ENFORCEMENT_ENABLED=false
           AUTH_TYPE=basic
@@ -509,28 +507,23 @@ jobs:
           POSTGRES_USE_NULL_POOL=true
           REQUIRE_EMAIL_VERIFICATION=false
           DISABLE_TELEMETRY=true
-          DISABLE_VECTOR_DB=true
           ONYX_BACKEND_IMAGE=${ECR_CACHE}:integration-test-backend-test-${RUN_ID}
           INTEGRATION_TESTS_MODE=true
-          USE_LIGHTWEIGHT_BACKGROUND_WORKER=true
           EOF
 
-      # Start only the services needed for no-vectordb mode (no Vespa, no model servers)
-      - name: Start Docker containers (no-vectordb)
+      # Start only the services needed for Onyx Lite (Postgres + API server)
+      - name: Start Docker containers (onyx-lite)
         run: |
           cd deployment/docker_compose
-          docker compose -f docker-compose.yml -f docker-compose.no-vectordb.yml -f docker-compose.dev.yml up \
+          docker compose -f docker-compose.yml -f docker-compose.onyx-lite.yml -f docker-compose.dev.yml up \
             relational_db \
-            cache \
-            minio \
             api_server \
-            background \
             -d
-        id: start_docker_no_vectordb
+        id: start_docker_onyx_lite
 
       - name: Wait for services to be ready
         run: |
-          echo "Starting wait-for-service script (no-vectordb)..."
+          echo "Starting wait-for-service script (onyx-lite)..."
           start_time=$(date +%s)
           timeout=300
           while true; do
@@ -552,14 +545,14 @@ jobs:
             sleep 5
           done
 
-      - name: Run No-VectorDB Integration Tests
+      - name: Run Onyx Lite Integration Tests
         uses: nick-fields/retry@ce71cc2ab81d554ebbe88c79ab5975992d79ba08 # ratchet:nick-fields/retry@v3
         with:
           timeout_minutes: 20
           max_attempts: 3
           retry_wait_seconds: 10
           command: |
-            echo "Running no-vectordb integration tests..."
+            echo "Running onyx-lite integration tests..."
             docker run --rm --network onyx_default \
               --name test-runner \
               -e POSTGRES_HOST=relational_db \
@@ -570,39 +563,38 @@ jobs:
               -e DB_READONLY_PASSWORD=password \
               -e POSTGRES_POOL_PRE_PING=true \
               -e POSTGRES_USE_NULL_POOL=true \
-              -e REDIS_HOST=cache \
               -e API_SERVER_HOST=api_server \
               -e OPENAI_API_KEY=${OPENAI_API_KEY} \
               -e TEST_WEB_HOSTNAME=test-runner \
               ${{ env.RUNS_ON_ECR_CACHE }}:integration-test-${{ github.run_id }} \
               /app/tests/integration/tests/no_vectordb
 
-      - name: Dump API server logs (no-vectordb)
+      - name: Dump API server logs (onyx-lite)
         if: always()
         run: |
           cd deployment/docker_compose
-          docker compose -f docker-compose.yml -f docker-compose.no-vectordb.yml -f docker-compose.dev.yml \
-            logs --no-color api_server > $GITHUB_WORKSPACE/api_server_no_vectordb.log || true
+          docker compose -f docker-compose.yml -f docker-compose.onyx-lite.yml -f docker-compose.dev.yml \
+            logs --no-color api_server > $GITHUB_WORKSPACE/api_server_onyx_lite.log || true
 
-      - name: Dump all-container logs (no-vectordb)
+      - name: Dump all-container logs (onyx-lite)
         if: always()
         run: |
           cd deployment/docker_compose
-          docker compose -f docker-compose.yml -f docker-compose.no-vectordb.yml -f docker-compose.dev.yml \
-            logs --no-color > $GITHUB_WORKSPACE/docker-compose-no-vectordb.log || true
+          docker compose -f docker-compose.yml -f docker-compose.onyx-lite.yml -f docker-compose.dev.yml \
+            logs --no-color > $GITHUB_WORKSPACE/docker-compose-onyx-lite.log || true
 
-      - name: Upload logs (no-vectordb)
+      - name: Upload logs (onyx-lite)
         if: always()
         uses: actions/upload-artifact@b7c566a772e6b6bfb58ed0dc250532a479d7789f
         with:
-          name: docker-all-logs-no-vectordb
-          path: ${{ github.workspace }}/docker-compose-no-vectordb.log
+          name: docker-all-logs-onyx-lite
+          path: ${{ github.workspace }}/docker-compose-onyx-lite.log
 
-      - name: Stop Docker containers (no-vectordb)
+      - name: Stop Docker containers (onyx-lite)
         if: always()
         run: |
           cd deployment/docker_compose
-          docker compose -f docker-compose.yml -f docker-compose.no-vectordb.yml -f docker-compose.dev.yml down -v
+          docker compose -f docker-compose.yml -f docker-compose.onyx-lite.yml -f docker-compose.dev.yml down -v
 
   multitenant-tests:
     needs:
@@ -744,7 +736,7 @@ jobs:
     # NOTE: Github-hosted runners have about 20s faster queue times and are preferred here.
     runs-on: ubuntu-slim
     timeout-minutes: 45
-    needs: [integration-tests, no-vectordb-tests, multitenant-tests]
+    needs: [integration-tests, onyx-lite-tests, multitenant-tests]
     if: ${{ always() }}
     steps:
       - name: Check job status

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

@@ -268,10 +268,11 @@ jobs:
           persist-credentials: false
 
       - name: Setup node
+        # zizmor: ignore[cache-poisoning] ephemeral runners; no release artifacts
         uses: actions/setup-node@6044e13b5dc448c55e2357c09f80417699197238 # ratchet:actions/setup-node@v4
         with:
           node-version: 22
-          cache: "npm"
+          cache: "npm" # zizmor: ignore[cache-poisoning]
           cache-dependency-path: ./web/package-lock.json
 
       - name: Install node dependencies
@@ -279,6 +280,7 @@ jobs:
         run: npm ci
 
       - name: Cache playwright cache
+        # zizmor: ignore[cache-poisoning] ephemeral runners; no release artifacts
         uses: runs-on/cache@50350ad4242587b6c8c2baa2e740b1bc11285ff4 # ratchet:runs-on/cache@v4
         with:
           path: ~/.cache/ms-playwright
@@ -590,6 +592,108 @@ jobs:
           name: docker-logs-${{ matrix.project }}-${{ github.run_id }}
           path: ${{ github.workspace }}/docker-compose.log
 
+  playwright-tests-lite:
+    needs: [build-web-image, build-backend-image]
+    name: Playwright Tests (lite)
+    runs-on:
+      - runs-on
+      - runner=4cpu-linux-arm64
+      - "run-id=${{ github.run_id }}-playwright-tests-lite"
+      - "extras=ecr-cache"
+    timeout-minutes: 30
+    steps:
+      - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
+
+      - name: Checkout code
+        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # ratchet:actions/checkout@v6
+        with:
+          persist-credentials: false
+
+      - name: Setup node
+        # zizmor: ignore[cache-poisoning] ephemeral runners; no release artifacts
+        uses: actions/setup-node@6044e13b5dc448c55e2357c09f80417699197238 # ratchet:actions/setup-node@v4
+        with:
+          node-version: 22
+          cache: "npm" # zizmor: ignore[cache-poisoning]
+          cache-dependency-path: ./web/package-lock.json
+
+      - name: Install node dependencies
+        working-directory: ./web
+        run: npm ci
+
+      - name: Cache playwright cache
+        # zizmor: ignore[cache-poisoning] ephemeral runners; no release artifacts
+        uses: runs-on/cache@50350ad4242587b6c8c2baa2e740b1bc11285ff4 # ratchet:runs-on/cache@v4
+        with:
+          path: ~/.cache/ms-playwright
+          key: ${{ runner.os }}-playwright-npm-${{ hashFiles('web/package-lock.json') }}
+          restore-keys: |
+            ${{ runner.os }}-playwright-npm-
+
+      - name: Install playwright browsers
+        working-directory: ./web
+        run: npx playwright install --with-deps
+
+      - name: Create .env file for Docker Compose
+        env:
+          OPENAI_API_KEY_VALUE: ${{ env.OPENAI_API_KEY }}
+          ECR_CACHE: ${{ env.RUNS_ON_ECR_CACHE }}
+          RUN_ID: ${{ github.run_id }}
+        run: |
+          cat <<EOF > deployment/docker_compose/.env
+          ENABLE_PAID_ENTERPRISE_EDITION_FEATURES=true
+          LICENSE_ENFORCEMENT_ENABLED=false
+          AUTH_TYPE=basic
+          INTEGRATION_TESTS_MODE=true
+          GEN_AI_API_KEY=${OPENAI_API_KEY_VALUE}
+          MOCK_LLM_RESPONSE=true
+          REQUIRE_EMAIL_VERIFICATION=false
+          DISABLE_TELEMETRY=true
+          ONYX_BACKEND_IMAGE=${ECR_CACHE}:playwright-test-backend-${RUN_ID}
+          ONYX_WEB_SERVER_IMAGE=${ECR_CACHE}:playwright-test-web-${RUN_ID}
+          EOF
+
+      # needed for pulling external images otherwise, we hit the "Unauthenticated users" limit
+      # https://docs.docker.com/docker-hub/usage/
+      - name: Login to Docker Hub
+        uses: docker/login-action@c94ce9fb468520275223c153574b00df6fe4bcc9 # ratchet:docker/login-action@v3
+        with:
+          username: ${{ secrets.DOCKER_USERNAME }}
+          password: ${{ secrets.DOCKER_TOKEN }}
+
+      - name: Start Docker containers (lite)
+        run: |
+          cd deployment/docker_compose
+          docker compose -f docker-compose.yml -f docker-compose.onyx-lite.yml -f docker-compose.dev.yml up -d
+        id: start_docker
+
+      - name: Run Playwright tests (lite)
+        working-directory: ./web
+        run: npx playwright test --project lite
+
+      - uses: actions/upload-artifact@b7c566a772e6b6bfb58ed0dc250532a479d7789f
+        if: always()
+        with:
+          name: playwright-test-results-lite-${{ github.run_id }}
+          path: ./web/output/playwright/
+          retention-days: 30
+
+      - name: Save Docker logs
+        if: success() || failure()
+        env:
+          WORKSPACE: ${{ github.workspace }}
+        run: |
+          cd deployment/docker_compose
+          docker compose logs > docker-compose.log
+          mv docker-compose.log ${WORKSPACE}/docker-compose.log
+
+      - name: Upload logs
+        if: success() || failure()
+        uses: actions/upload-artifact@b7c566a772e6b6bfb58ed0dc250532a479d7789f
+        with:
+          name: docker-logs-lite-${{ github.run_id }}
+          path: ${{ github.workspace }}/docker-compose.log
+
   # Post a single combined visual regression comment after all matrix jobs finish
   visual-regression-comment:
     needs: [playwright-tests]
@@ -686,7 +790,7 @@ jobs:
     # NOTE: Github-hosted runners have about 20s faster queue times and are preferred here.
     runs-on: ubuntu-slim
     timeout-minutes: 45
-    needs: [playwright-tests]
+    needs: [playwright-tests, playwright-tests-lite]
     if: ${{ always() }}
     steps:
       - name: Check job status

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

@@ -8,7 +8,7 @@ on:
   pull_request:
     branches:
       - main
-      - 'release/**'
+      - "release/**"
   push:
     tags:
       - "v*.*.*"
@@ -21,7 +21,13 @@ jobs:
     # See https://runs-on.com/runners/linux/
     # Note: Mypy seems quite optimized for x64 compared to arm64.
     # Similarly, mypy is single-threaded and incremental, so 2cpu is sufficient.
-    runs-on: [runs-on, runner=2cpu-linux-x64, "run-id=${{ github.run_id }}-mypy-check", "extras=s3-cache"]
+    runs-on:
+      [
+        runs-on,
+        runner=2cpu-linux-x64,
+        "run-id=${{ github.run_id }}-mypy-check",
+        "extras=s3-cache",
+      ]
     timeout-minutes: 45
 
     steps:
@@ -52,21 +58,14 @@ jobs:
         if: ${{ vars.DISABLE_MYPY_CACHE != 'true' }}
         uses: runs-on/cache@50350ad4242587b6c8c2baa2e740b1bc11285ff4 # ratchet:runs-on/cache@v4
         with:
-          path: backend/.mypy_cache
-          key: mypy-${{ runner.os }}-${{ github.base_ref || github.event.merge_group.base_ref || 'main' }}-${{ hashFiles('**/*.py', '**/*.pyi', 'backend/pyproject.toml') }}
+          path: .mypy_cache
+          key: mypy-${{ runner.os }}-${{ github.base_ref || github.event.merge_group.base_ref || 'main' }}-${{ hashFiles('**/*.py', '**/*.pyi', 'pyproject.toml') }}
           restore-keys: |
             mypy-${{ runner.os }}-${{ github.base_ref || github.event.merge_group.base_ref || 'main' }}-
             mypy-${{ runner.os }}-
 
       - name: Run MyPy
-        working-directory: ./backend
         env:
           MYPY_FORCE_COLOR: 1
           TERM: xterm-256color
         run: mypy .
-
-      - name: Run MyPy (tools/)
-        env:
-          MYPY_FORCE_COLOR: 1
-          TERM: xterm-256color
-        run: mypy tools/

.github/workflows/reusable-nightly-llm-provider-chat.yml

@@ -48,28 +48,10 @@ on:
         required: false
         default: true
         type: boolean
-    secrets:
-      openai_api_key:
-        required: false
-      anthropic_api_key:
-        required: false
-      bedrock_api_key:
-        required: false
-      vertex_ai_custom_config_json:
-        required: false
-      azure_api_key:
-        required: false
-      ollama_api_key:
-        required: false
-      openrouter_api_key:
-        required: false
-      DOCKER_USERNAME:
-        required: true
-      DOCKER_TOKEN:
-        required: true
 
 permissions:
   contents: read
+  id-token: write
 
 jobs:
   build-backend-image:
@@ -81,6 +63,7 @@ jobs:
         "extras=ecr-cache",
       ]
     timeout-minutes: 45
+    environment: ci-protected
     steps:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
 
@@ -89,6 +72,19 @@ jobs:
         with:
           persist-credentials: false
 
+      - name: Configure AWS credentials
+        uses: aws-actions/configure-aws-credentials@61815dcd50bd041e203e49132bacad1fd04d2708
+        with:
+          role-to-assume: ${{ secrets.AWS_OIDC_ROLE_ARN }}
+          aws-region: us-east-2
+
+      - name: Get AWS Secrets
+        uses: aws-actions/aws-secretsmanager-get-secrets@a9a7eb4e2f2871d30dc5b892576fde60a2ecc802
+        with:
+          secret-ids: |
+            DOCKER_USERNAME, test/docker-username
+            DOCKER_TOKEN, test/docker-token
+
       - name: Build backend image
         uses: ./.github/actions/build-backend-image
         with:
@@ -97,8 +93,8 @@ jobs:
           pr-number: ${{ github.event.pull_request.number }}
           github-sha: ${{ github.sha }}
           run-id: ${{ github.run_id }}
-          docker-username: ${{ secrets.DOCKER_USERNAME }}
-          docker-token: ${{ secrets.DOCKER_TOKEN }}
+          docker-username: ${{ env.DOCKER_USERNAME }}
+          docker-token: ${{ env.DOCKER_TOKEN }}
           docker-no-cache: ${{ vars.DOCKER_NO_CACHE == 'true' && 'true' || 'false' }}
 
   build-model-server-image:
@@ -110,6 +106,7 @@ jobs:
         "extras=ecr-cache",
       ]
     timeout-minutes: 45
+    environment: ci-protected
     steps:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
 
@@ -118,6 +115,19 @@ jobs:
         with:
           persist-credentials: false
 
+      - name: Configure AWS credentials
+        uses: aws-actions/configure-aws-credentials@61815dcd50bd041e203e49132bacad1fd04d2708
+        with:
+          role-to-assume: ${{ secrets.AWS_OIDC_ROLE_ARN }}
+          aws-region: us-east-2
+
+      - name: Get AWS Secrets
+        uses: aws-actions/aws-secretsmanager-get-secrets@a9a7eb4e2f2871d30dc5b892576fde60a2ecc802
+        with:
+          secret-ids: |
+            DOCKER_USERNAME, test/docker-username
+            DOCKER_TOKEN, test/docker-token
+
       - name: Build model server image
         uses: ./.github/actions/build-model-server-image
         with:
@@ -126,8 +136,8 @@ jobs:
           pr-number: ${{ github.event.pull_request.number }}
           github-sha: ${{ github.sha }}
           run-id: ${{ github.run_id }}
-          docker-username: ${{ secrets.DOCKER_USERNAME }}
-          docker-token: ${{ secrets.DOCKER_TOKEN }}
+          docker-username: ${{ env.DOCKER_USERNAME }}
+          docker-token: ${{ env.DOCKER_TOKEN }}
 
   build-integration-image:
     runs-on:
@@ -138,6 +148,7 @@ jobs:
         "extras=ecr-cache",
       ]
     timeout-minutes: 45
+    environment: ci-protected
     steps:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
 
@@ -146,6 +157,19 @@ jobs:
         with:
           persist-credentials: false
 
+      - name: Configure AWS credentials
+        uses: aws-actions/configure-aws-credentials@61815dcd50bd041e203e49132bacad1fd04d2708
+        with:
+          role-to-assume: ${{ secrets.AWS_OIDC_ROLE_ARN }}
+          aws-region: us-east-2
+
+      - name: Get AWS Secrets
+        uses: aws-actions/aws-secretsmanager-get-secrets@a9a7eb4e2f2871d30dc5b892576fde60a2ecc802
+        with:
+          secret-ids: |
+            DOCKER_USERNAME, test/docker-username
+            DOCKER_TOKEN, test/docker-token
+
       - name: Build integration image
         uses: ./.github/actions/build-integration-image
         with:
@@ -154,8 +178,8 @@ jobs:
           pr-number: ${{ github.event.pull_request.number }}
           github-sha: ${{ github.sha }}
           run-id: ${{ github.run_id }}
-          docker-username: ${{ secrets.DOCKER_USERNAME }}
-          docker-token: ${{ secrets.DOCKER_TOKEN }}
+          docker-username: ${{ env.DOCKER_USERNAME }}
+          docker-token: ${{ env.DOCKER_TOKEN }}
 
   provider-chat-test:
     needs:
@@ -170,56 +194,56 @@ jobs:
         include:
           - provider: openai
             models: ${{ inputs.openai_models }}
-            api_key_secret: openai_api_key
-            custom_config_secret: ""
+            api_key_env: OPENAI_API_KEY
+            custom_config_env: ""
             api_base: ""
             api_version: ""
             deployment_name: ""
             required: true
           - provider: anthropic
             models: ${{ inputs.anthropic_models }}
-            api_key_secret: anthropic_api_key
-            custom_config_secret: ""
+            api_key_env: ANTHROPIC_API_KEY
+            custom_config_env: ""
             api_base: ""
             api_version: ""
             deployment_name: ""
             required: true
           - provider: bedrock
             models: ${{ inputs.bedrock_models }}
-            api_key_secret: bedrock_api_key
-            custom_config_secret: ""
+            api_key_env: BEDROCK_API_KEY
+            custom_config_env: ""
             api_base: ""
             api_version: ""
             deployment_name: ""
             required: false
           - provider: vertex_ai
             models: ${{ inputs.vertex_ai_models }}
-            api_key_secret: ""
-            custom_config_secret: vertex_ai_custom_config_json
+            api_key_env: ""
+            custom_config_env: NIGHTLY_LLM_VERTEX_AI_CUSTOM_CONFIG_JSON
             api_base: ""
             api_version: ""
             deployment_name: ""
             required: false
           - provider: azure
             models: ${{ inputs.azure_models }}
-            api_key_secret: azure_api_key
-            custom_config_secret: ""
+            api_key_env: AZURE_API_KEY
+            custom_config_env: ""
             api_base: ${{ inputs.azure_api_base }}
             api_version: "2025-04-01-preview"
             deployment_name: ""
             required: false
           - provider: ollama_chat
             models: ${{ inputs.ollama_models }}
-            api_key_secret: ollama_api_key
-            custom_config_secret: ""
+            api_key_env: OLLAMA_API_KEY
+            custom_config_env: ""
             api_base: "https://ollama.com"
             api_version: ""
             deployment_name: ""
             required: false
           - provider: openrouter
             models: ${{ inputs.openrouter_models }}
-            api_key_secret: openrouter_api_key
-            custom_config_secret: ""
+            api_key_env: OPENROUTER_API_KEY
+            custom_config_env: ""
             api_base: "https://openrouter.ai/api/v1"
             api_version: ""
             deployment_name: ""
@@ -230,6 +254,7 @@ jobs:
       - "run-id=${{ github.run_id }}-nightly-${{ matrix.provider }}-provider-chat-test"
       - extras=ecr-cache
     timeout-minutes: 45
+    environment: ci-protected
     steps:
       - uses: runs-on/action@cd2b598b0515d39d78c38a02d529db87d2196d1e # ratchet:runs-on/action@v2
 
@@ -238,21 +263,43 @@ jobs:
         with:
           persist-credentials: false
 
+      - name: Configure AWS credentials
+        uses: aws-actions/configure-aws-credentials@61815dcd50bd041e203e49132bacad1fd04d2708
+        with:
+          role-to-assume: ${{ secrets.AWS_OIDC_ROLE_ARN }}
+          aws-region: us-east-2
+
+      - name: Get AWS Secrets
+        uses: aws-actions/aws-secretsmanager-get-secrets@a9a7eb4e2f2871d30dc5b892576fde60a2ecc802
+        with:
+          # Keep JSON values unparsed so vertex custom config is passed as raw JSON.
+          parse-json-secrets: false
+          secret-ids: |
+            DOCKER_USERNAME, test/docker-username
+            DOCKER_TOKEN, test/docker-token
+            OPENAI_API_KEY, test/openai-api-key
+            ANTHROPIC_API_KEY, test/anthropic-api-key
+            BEDROCK_API_KEY, test/bedrock-api-key
+            NIGHTLY_LLM_VERTEX_AI_CUSTOM_CONFIG_JSON, test/nightly-llm-vertex-ai-custom-config-json
+            AZURE_API_KEY, test/azure-api-key
+            OLLAMA_API_KEY, test/ollama-api-key
+            OPENROUTER_API_KEY, test/openrouter-api-key
+
       - name: Run nightly provider chat test
         uses: ./.github/actions/run-nightly-provider-chat-test
         with:
           provider: ${{ matrix.provider }}
           models: ${{ matrix.models }}
-          provider-api-key: ${{ matrix.api_key_secret && secrets[matrix.api_key_secret] || '' }}
+          provider-api-key: ${{ matrix.api_key_env && env[matrix.api_key_env] || '' }}
           strict: ${{ inputs.strict && 'true' || 'false' }}
           api-base: ${{ matrix.api_base }}
           api-version: ${{ matrix.api_version }}
           deployment-name: ${{ matrix.deployment_name }}
-          custom-config-json: ${{ matrix.custom_config_secret && secrets[matrix.custom_config_secret] || '' }}
+          custom-config-json: ${{ matrix.custom_config_env && env[matrix.custom_config_env] || '' }}
           runs-on-ecr-cache: ${{ env.RUNS_ON_ECR_CACHE }}
           run-id: ${{ github.run_id }}
-          docker-username: ${{ secrets.DOCKER_USERNAME }}
-          docker-token: ${{ secrets.DOCKER_TOKEN }}
+          docker-username: ${{ env.DOCKER_USERNAME }}
+          docker-token: ${{ env.DOCKER_TOKEN }}
 
       - name: Dump API server logs
         if: always()

.vscode/launch.json

@@ -40,19 +40,7 @@
       }
     },
     {
-      "name": "Celery (lightweight mode)",
-      "configurations": [
-        "Celery primary",
-        "Celery background",
-        "Celery beat"
-      ],
-      "presentation": {
-        "group": "1"
-      },
-      "stopAll": true
-    },
-    {
-      "name": "Celery (standard mode)",
+      "name": "Celery",
       "configurations": [
         "Celery primary",
         "Celery light",
@@ -253,35 +241,6 @@
       },
       "consoleTitle": "Celery light Console"
     },
-    {
-      "name": "Celery background",
-      "type": "debugpy",
-      "request": "launch",
-      "module": "celery",
-      "cwd": "${workspaceFolder}/backend",
-      "envFile": "${workspaceFolder}/.vscode/.env",
-      "env": {
-        "LOG_LEVEL": "INFO",
-        "PYTHONUNBUFFERED": "1",
-        "PYTHONPATH": "."
-      },
-      "args": [
-        "-A",
-        "onyx.background.celery.versioned_apps.background",
-        "worker",
-        "--pool=threads",
-        "--concurrency=20",
-        "--prefetch-multiplier=4",
-        "--loglevel=INFO",
-        "--hostname=background@%n",
-        "-Q",
-        "vespa_metadata_sync,connector_deletion,doc_permissions_upsert,checkpoint_cleanup,index_attempt_cleanup,docprocessing,connector_doc_fetching,connector_pruning,connector_doc_permissions_sync,connector_external_group_sync,csv_generation,kg_processing,monitoring,user_file_processing,user_file_project_sync,user_file_delete,opensearch_migration"
-      ],
-      "presentation": {
-        "group": "2"
-      },
-      "consoleTitle": "Celery background Console"
-    },
     {
       "name": "Celery heavy",
       "type": "debugpy",

AGENTS.md

@@ -86,37 +86,6 @@ Onyx uses Celery for asynchronous task processing with multiple specialized work
      - 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
@@ -617,6 +586,45 @@ 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.
 
+## Error Handling
+
+**Always raise `OnyxError` from `onyx.error_handling.exceptions` instead of `HTTPException`.
+Never hardcode status codes or use `starlette.status` / `fastapi.status` constants directly.**
+
+A global FastAPI exception handler converts `OnyxError` into a JSON response with the standard
+`{"error_code": "...", "message": "..."}` shape. This eliminates boilerplate and keeps error
+handling consistent across the entire backend.
+
+```python
+from onyx.error_handling.error_codes import OnyxErrorCode
+from onyx.error_handling.exceptions import OnyxError
+
+# ✅ Good
+raise OnyxError(OnyxErrorCode.NOT_FOUND, "Session not found")
+
+# ✅ Good — no extra message needed
+raise OnyxError(OnyxErrorCode.UNAUTHENTICATED)
+
+# ✅ Good — upstream service with dynamic status code
+raise OnyxError(OnyxErrorCode.BAD_GATEWAY, detail, status_code_override=upstream_status)
+
+# ❌ Bad — using HTTPException directly
+raise HTTPException(status_code=404, detail="Session not found")
+
+# ❌ Bad — starlette constant
+raise HTTPException(status_code=status.HTTP_403_FORBIDDEN, detail="Access denied")
+```
+
+Available error codes are defined in `backend/onyx/error_handling/error_codes.py`. If a new error
+category is needed, add it there first — do not invent ad-hoc codes.
+
+**Upstream service errors:** When forwarding errors from an upstream service where the HTTP
+status code is dynamic (comes from the upstream response), use `status_code_override`:
+
+```python
+raise OnyxError(OnyxErrorCode.BAD_GATEWAY, detail, status_code_override=e.response.status_code)
+```
+
 ## Best Practices
 
 In addition to the other content in this file, best practices for contributing

backend/alembic/versions/2664261bfaab_add_cache_store_table.py

@@ -0,0 +1,37 @@
+"""add cache_store table
+
+Revision ID: 2664261bfaab
+Revises: 4a1e4b1c89d2
+Create Date: 2026-02-27 00:00:00.000000
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+# revision identifiers, used by Alembic.
+revision = "2664261bfaab"
+down_revision = "4a1e4b1c89d2"
+branch_labels: None = None
+depends_on: None = None
+
+
+def upgrade() -> None:
+    op.create_table(
+        "cache_store",
+        sa.Column("key", sa.String(), nullable=False),
+        sa.Column("value", sa.LargeBinary(), nullable=True),
+        sa.Column("expires_at", sa.DateTime(timezone=True), nullable=True),
+        sa.PrimaryKeyConstraint("key"),
+    )
+    op.create_index(
+        "ix_cache_store_expires",
+        "cache_store",
+        ["expires_at"],
+        postgresql_where=sa.text("expires_at IS NOT NULL"),
+    )
+
+
+def downgrade() -> None:
+    op.drop_index("ix_cache_store_expires", table_name="cache_store")
+    op.drop_table("cache_store")

backend/alembic/versions/4a1e4b1c89d2_add_indexing_to_userfilestatus.py

@@ -0,0 +1,51 @@
+"""Add INDEXING to UserFileStatus
+
+Revision ID: 4a1e4b1c89d2
+Revises: 6b3b4083c5aa
+Create Date: 2026-02-28 00:00:00.000000
+
+"""
+
+import sqlalchemy as sa
+from alembic import op
+
+revision = "4a1e4b1c89d2"
+down_revision = "6b3b4083c5aa"
+branch_labels = None
+depends_on = None
+
+TABLE = "user_file"
+COLUMN = "status"
+CONSTRAINT_NAME = "ck_user_file_status"
+
+OLD_VALUES = ("PROCESSING", "COMPLETED", "FAILED", "CANCELED", "DELETING")
+NEW_VALUES = ("PROCESSING", "INDEXING", "COMPLETED", "FAILED", "CANCELED", "DELETING")
+
+
+def _drop_status_check_constraint() -> None:
+    """Drop the existing CHECK constraint on user_file.status.
+
+    The constraint name is auto-generated by SQLAlchemy and unknown,
+    so we look it up via the inspector.
+    """
+    inspector = sa.inspect(op.get_bind())
+    for constraint in inspector.get_check_constraints(TABLE):
+        if COLUMN in constraint.get("sqltext", ""):
+            constraint_name = constraint["name"]
+            if constraint_name is not None:
+                op.drop_constraint(constraint_name, TABLE, type_="check")
+
+
+def upgrade() -> None:
+    _drop_status_check_constraint()
+    in_clause = ", ".join(f"'{v}'" for v in NEW_VALUES)
+    op.create_check_constraint(CONSTRAINT_NAME, TABLE, f"{COLUMN} IN ({in_clause})")
+
+
+def downgrade() -> None:
+    op.execute(
+        f"UPDATE {TABLE} SET {COLUMN} = 'PROCESSING' WHERE {COLUMN} = 'INDEXING'"
+    )
+    op.drop_constraint(CONSTRAINT_NAME, TABLE, type_="check")
+    in_clause = ", ".join(f"'{v}'" for v in OLD_VALUES)
+    op.create_check_constraint(CONSTRAINT_NAME, TABLE, f"{COLUMN} IN ({in_clause})")

backend/alembic/versions/6b3b4083c5aa_persona_cleanup_and_featured.py

@@ -0,0 +1,112 @@
+"""persona cleanup and featured
+
+Revision ID: 6b3b4083c5aa
+Revises: 57122d037335
+Create Date: 2026-02-26 12:00:00.000000
+
+"""
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "6b3b4083c5aa"
+down_revision = "57122d037335"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    # Add featured column with nullable=True first
+    op.add_column("persona", sa.Column("featured", sa.Boolean(), nullable=True))
+
+    # Migrate data from is_default_persona to featured
+    op.execute("UPDATE persona SET featured = is_default_persona")
+
+    # Make featured non-nullable with default=False
+    op.alter_column(
+        "persona",
+        "featured",
+        existing_type=sa.Boolean(),
+        nullable=False,
+        server_default=sa.false(),
+    )
+
+    # Drop is_default_persona column
+    op.drop_column("persona", "is_default_persona")
+
+    # Drop unused columns
+    op.drop_column("persona", "num_chunks")
+    op.drop_column("persona", "chunks_above")
+    op.drop_column("persona", "chunks_below")
+    op.drop_column("persona", "llm_relevance_filter")
+    op.drop_column("persona", "llm_filter_extraction")
+    op.drop_column("persona", "recency_bias")
+
+
+def downgrade() -> None:
+    # Add back recency_bias column
+    op.add_column(
+        "persona",
+        sa.Column(
+            "recency_bias",
+            sa.VARCHAR(),
+            nullable=False,
+            server_default="base_decay",
+        ),
+    )
+
+    # Add back llm_filter_extraction column
+    op.add_column(
+        "persona",
+        sa.Column(
+            "llm_filter_extraction",
+            sa.Boolean(),
+            nullable=False,
+            server_default=sa.false(),
+        ),
+    )
+
+    # Add back llm_relevance_filter column
+    op.add_column(
+        "persona",
+        sa.Column(
+            "llm_relevance_filter",
+            sa.Boolean(),
+            nullable=False,
+            server_default=sa.false(),
+        ),
+    )
+
+    # Add back chunks_below column
+    op.add_column(
+        "persona",
+        sa.Column("chunks_below", sa.Integer(), nullable=False, server_default="0"),
+    )
+
+    # Add back chunks_above column
+    op.add_column(
+        "persona",
+        sa.Column("chunks_above", sa.Integer(), nullable=False, server_default="0"),
+    )
+
+    # Add back num_chunks column
+    op.add_column("persona", sa.Column("num_chunks", sa.Float(), nullable=True))
+
+    # Add back is_default_persona column
+    op.add_column(
+        "persona",
+        sa.Column(
+            "is_default_persona",
+            sa.Boolean(),
+            nullable=False,
+            server_default=sa.false(),
+        ),
+    )
+
+    # Migrate data from featured to is_default_persona
+    op.execute("UPDATE persona SET is_default_persona = featured")
+
+    # Drop featured column
+    op.drop_column("persona", "featured")

backend/alembic/versions/a3b8d9e2f1c4_make_scim_external_id_nullable.py

@@ -0,0 +1,34 @@
+"""make scim_user_mapping.external_id nullable
+
+Revision ID: a3b8d9e2f1c4
+Revises: 2664261bfaab
+Create Date: 2026-03-02
+
+"""
+
+from alembic import op
+
+
+# revision identifiers, used by Alembic.
+revision = "a3b8d9e2f1c4"
+down_revision = "2664261bfaab"
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    op.alter_column(
+        "scim_user_mapping",
+        "external_id",
+        nullable=True,
+    )
+
+
+def downgrade() -> None:
+    # Delete any rows where external_id is NULL before re-applying NOT NULL
+    op.execute("DELETE FROM scim_user_mapping WHERE external_id IS NULL")
+    op.alter_column(
+        "scim_user_mapping",
+        "external_id",
+        nullable=False,
+    )

backend/ee/onyx/background/celery/apps/background.py

@@ -1,15 +0,0 @@
-from onyx.background.celery.apps import app_base
-from onyx.background.celery.apps.background import celery_app
-
-
-celery_app.autodiscover_tasks(
-    app_base.filter_task_modules(
-        [
-            "ee.onyx.background.celery.tasks.doc_permission_syncing",
-            "ee.onyx.background.celery.tasks.external_group_syncing",
-            "ee.onyx.background.celery.tasks.cleanup",
-            "ee.onyx.background.celery.tasks.tenant_provisioning",
-            "ee.onyx.background.celery.tasks.query_history",
-        ]
-    )
-)

backend/ee/onyx/db/license.py

@@ -11,11 +11,10 @@ from ee.onyx.server.license.models import LicenseMetadata
 from ee.onyx.server.license.models import LicensePayload
 from ee.onyx.server.license.models import LicenseSource
 from onyx.auth.schemas import UserRole
+from onyx.cache.factory import get_cache_backend
 from onyx.configs.constants import ANONYMOUS_USER_EMAIL
 from onyx.db.models import License
 from onyx.db.models import User
-from onyx.redis.redis_pool import get_redis_client
-from onyx.redis.redis_pool import get_redis_replica_client
 from onyx.utils.logger import setup_logger
 from shared_configs.configs import MULTI_TENANT
 from shared_configs.contextvars import get_current_tenant_id
@@ -142,7 +141,7 @@ def get_used_seats(tenant_id: str | None = None) -> int:
 
 def get_cached_license_metadata(tenant_id: str | None = None) -> LicenseMetadata | None:
     """
-    Get license metadata from Redis cache.
+    Get license metadata from cache.
 
     Args:
         tenant_id: Tenant ID (for multi-tenant deployments)
@@ -150,38 +149,34 @@ def get_cached_license_metadata(tenant_id: str | None = None) -> LicenseMetadata
     Returns:
         LicenseMetadata if cached, None otherwise
     """
-    tenant = tenant_id or get_current_tenant_id()
-    redis_client = get_redis_replica_client(tenant_id=tenant)
+    cache = get_cache_backend(tenant_id=tenant_id)
+    cached = cache.get(LICENSE_METADATA_KEY)
+    if not cached:
+        return None
 
-    cached = redis_client.get(LICENSE_METADATA_KEY)
-    if cached:
-        try:
-            cached_str: str
-            if isinstance(cached, bytes):
-                cached_str = cached.decode("utf-8")
-            else:
-                cached_str = str(cached)
-            return LicenseMetadata.model_validate_json(cached_str)
-        except Exception as e:
-            logger.warning(f"Failed to parse cached license metadata: {e}")
-            return None
-    return None
+    try:
+        cached_str = (
+            cached.decode("utf-8") if isinstance(cached, bytes) else str(cached)
+        )
+        return LicenseMetadata.model_validate_json(cached_str)
+    except Exception as e:
+        logger.warning(f"Failed to parse cached license metadata: {e}")
+        return None
 
 
 def invalidate_license_cache(tenant_id: str | None = None) -> None:
     """
     Invalidate the license metadata cache (not the license itself).
 
-    This deletes the cached LicenseMetadata from Redis. The actual license
-    in the database is not affected. Redis delete is idempotent - if the
-    key doesn't exist, this is a no-op.
+    Deletes the cached LicenseMetadata. The actual license in the database
+    is not affected. Delete is idempotent — if the key doesn't exist, this
+    is a no-op.
 
     Args:
         tenant_id: Tenant ID (for multi-tenant deployments)
     """
-    tenant = tenant_id or get_current_tenant_id()
-    redis_client = get_redis_client(tenant_id=tenant)
-    redis_client.delete(LICENSE_METADATA_KEY)
+    cache = get_cache_backend(tenant_id=tenant_id)
+    cache.delete(LICENSE_METADATA_KEY)
     logger.info("License cache invalidated")
 
 
@@ -192,7 +187,7 @@ def update_license_cache(
     tenant_id: str | None = None,
 ) -> LicenseMetadata:
     """
-    Update the Redis cache with license metadata.
+    Update the cache with license metadata.
 
     We cache all license statuses (ACTIVE, GRACE_PERIOD, GATED_ACCESS) because:
     1. Frontend needs status to show appropriate UI/banners
@@ -211,7 +206,7 @@ def update_license_cache(
     from ee.onyx.utils.license import get_license_status
 
     tenant = tenant_id or get_current_tenant_id()
-    redis_client = get_redis_client(tenant_id=tenant)
+    cache = get_cache_backend(tenant_id=tenant_id)
 
     used_seats = get_used_seats(tenant)
     status = get_license_status(payload, grace_period_end)
@@ -230,7 +225,7 @@ def update_license_cache(
         stripe_subscription_id=payload.stripe_subscription_id,
     )
 
-    redis_client.set(
+    cache.set(
         LICENSE_METADATA_KEY,
         metadata.model_dump_json(),
         ex=LICENSE_CACHE_TTL_SECONDS,

backend/ee/onyx/db/scim.py

@@ -126,12 +126,16 @@ class ScimDAL(DAL):
 
     def create_user_mapping(
         self,
-        external_id: str,
+        external_id: str | None,
         user_id: UUID,
         scim_username: str | None = None,
         fields: ScimMappingFields | None = None,
     ) -> ScimUserMapping:
-        """Create a mapping between a SCIM externalId and an Onyx user."""
+        """Create a SCIM mapping for a user.
+
+        ``external_id`` may be ``None`` when the IdP omits it (RFC 7643
+        allows this). The mapping still marks the user as SCIM-managed.
+        """
         f = fields or ScimMappingFields()
         mapping = ScimUserMapping(
             external_id=external_id,
@@ -270,8 +274,13 @@ class ScimDAL(DAL):
         Raises:
             ValueError: If the filter uses an unsupported attribute.
         """
-        query = select(User).where(
-            User.role.notin_([UserRole.SLACK_USER, UserRole.EXT_PERM_USER])
+        # Inner-join with ScimUserMapping so only SCIM-managed users appear.
+        # Pre-existing system accounts (anonymous, admin, etc.) are excluded
+        # unless they were explicitly linked via SCIM provisioning.
+        query = (
+            select(User)
+            .join(ScimUserMapping, ScimUserMapping.user_id == User.id)
+            .where(User.role.notin_([UserRole.SLACK_USER, UserRole.EXT_PERM_USER]))
         )
 
         if scim_filter:
@@ -321,34 +330,37 @@ class ScimDAL(DAL):
         scim_username: str | None = None,
         fields: ScimMappingFields | None = None,
     ) -> None:
-        """Create, update, or delete the external ID mapping for a user.
+        """Sync the SCIM mapping for a user.
+
+        If a mapping already exists, its fields are updated (including
+        setting ``external_id`` to ``None`` when the IdP omits it).
+        If no mapping exists and ``new_external_id`` is provided, a new
+        mapping is created.  A mapping is never deleted here — SCIM-managed
+        users must retain their mapping to remain visible in ``GET /Users``.
 
         When *fields* is provided, all mapping fields are written
         unconditionally — including ``None`` values — so that a caller can
         clear a previously-set field (e.g. removing a department).
         """
         mapping = self.get_user_mapping_by_user_id(user_id)
-        if new_external_id:
-            if mapping:
-                if mapping.external_id != new_external_id:
-                    mapping.external_id = new_external_id
-                if scim_username is not None:
-                    mapping.scim_username = scim_username
-                if fields is not None:
-                    mapping.department = fields.department
-                    mapping.manager = fields.manager
-                    mapping.given_name = fields.given_name
-                    mapping.family_name = fields.family_name
-                    mapping.scim_emails_json = fields.scim_emails_json
-            else:
-                self.create_user_mapping(
-                    external_id=new_external_id,
-                    user_id=user_id,
-                    scim_username=scim_username,
-                    fields=fields,
-                )
-        elif mapping:
-            self.delete_user_mapping(mapping.id)
+        if mapping:
+            if mapping.external_id != new_external_id:
+                mapping.external_id = new_external_id
+            if scim_username is not None:
+                mapping.scim_username = scim_username
+            if fields is not None:
+                mapping.department = fields.department
+                mapping.manager = fields.manager
+                mapping.given_name = fields.given_name
+                mapping.family_name = fields.family_name
+                mapping.scim_emails_json = fields.scim_emails_json
+        elif new_external_id:
+            self.create_user_mapping(
+                external_id=new_external_id,
+                user_id=user_id,
+                scim_username=scim_username,
+                fields=fields,
+            )
 
     def _get_user_mappings_batch(
         self, user_ids: list[UUID]

backend/ee/onyx/db/user_group.py

@@ -2,7 +2,6 @@ from collections.abc import Sequence
 from operator import and_
 from uuid import UUID
 
-from fastapi import HTTPException
 from sqlalchemy import delete
 from sqlalchemy import func
 from sqlalchemy import Select
@@ -15,6 +14,7 @@ from sqlalchemy.orm import Session
 from ee.onyx.server.user_group.models import SetCuratorRequest
 from ee.onyx.server.user_group.models import UserGroupCreate
 from ee.onyx.server.user_group.models import UserGroupUpdate
+from onyx.configs.app_configs import DISABLE_VECTOR_DB
 from onyx.db.connector_credential_pair import get_connector_credential_pair_from_id
 from onyx.db.enums import AccessType
 from onyx.db.enums import ConnectorCredentialPairStatus
@@ -36,6 +36,8 @@ from onyx.db.models import UserGroup
 from onyx.db.models import UserGroup__ConnectorCredentialPair
 from onyx.db.models import UserRole
 from onyx.db.users import fetch_user_by_id
+from onyx.error_handling.error_codes import OnyxErrorCode
+from onyx.error_handling.exceptions import OnyxError
 from onyx.utils.logger import setup_logger
 
 logger = setup_logger()
@@ -165,18 +167,12 @@ def validate_object_creation_for_user(
     if object_is_public and user.role == UserRole.BASIC:
         detail = "User does not have permission to create public objects"
         logger.error(detail)
-        raise HTTPException(
-            status_code=400,
-            detail=detail,
-        )
+        raise OnyxError(OnyxErrorCode.INSUFFICIENT_PERMISSIONS, detail)
 
     if not target_group_ids:
         detail = "Curators must specify 1+ groups"
         logger.error(detail)
-        raise HTTPException(
-            status_code=400,
-            detail=detail,
-        )
+        raise OnyxError(OnyxErrorCode.VALIDATION_ERROR, detail)
 
     user_curated_groups = fetch_user_groups_for_user(
         db_session=db_session,
@@ -189,10 +185,7 @@ def validate_object_creation_for_user(
     if not target_group_ids_set.issubset(user_curated_group_ids):
         detail = "Curators cannot control groups they don't curate"
         logger.error(detail)
-        raise HTTPException(
-            status_code=400,
-            detail=detail,
-        )
+        raise OnyxError(OnyxErrorCode.INSUFFICIENT_PERMISSIONS, detail)
 
 
 def fetch_user_group(db_session: Session, user_group_id: int) -> UserGroup | None:
@@ -471,7 +464,9 @@ def _add_user_group__cc_pair_relationships__no_commit(
 
 def insert_user_group(db_session: Session, user_group: UserGroupCreate) -> UserGroup:
     db_user_group = UserGroup(
-        name=user_group.name, time_last_modified_by_user=func.now()
+        name=user_group.name,
+        time_last_modified_by_user=func.now(),
+        is_up_to_date=DISABLE_VECTOR_DB,
     )
     db_session.add(db_user_group)
     db_session.flush()  # give the group an ID
@@ -774,8 +769,7 @@ def update_user_group(
             cc_pair_ids=user_group_update.cc_pair_ids,
         )
 
-    # only needs to sync with Vespa if the cc_pairs have been updated
-    if cc_pairs_updated:
+    if cc_pairs_updated and not DISABLE_VECTOR_DB:
         db_user_group.is_up_to_date = False
 
     removed_users = db_session.scalars(

backend/ee/onyx/main.py

@@ -31,6 +31,7 @@ from ee.onyx.server.query_and_chat.query_backend import (
 from ee.onyx.server.query_and_chat.search_backend import router as search_router
 from ee.onyx.server.query_history.api import router as query_history_router
 from ee.onyx.server.reporting.usage_export_api import router as usage_export_router
+from ee.onyx.server.scim.api import register_scim_exception_handlers
 from ee.onyx.server.scim.api import scim_router
 from ee.onyx.server.seeding import seed_db
 from ee.onyx.server.tenants.api import router as tenants_router
@@ -167,6 +168,7 @@ def get_application() -> FastAPI:
     # they use their own SCIM bearer token auth).
     # Not behind APP_API_PREFIX because IdPs expect /scim/v2/... directly.
     application.include_router(scim_router)
+    register_scim_exception_handlers(application)
 
     # Ensure all routes have auth enabled or are explicitly marked as public
     check_ee_router_auth(application)

backend/ee/onyx/server/billing/api.py

@@ -26,7 +26,6 @@ import asyncio
 import httpx
 from fastapi import APIRouter
 from fastapi import Depends
-from fastapi import HTTPException
 from pydantic import BaseModel
 from sqlalchemy.orm import Session
 
@@ -42,7 +41,6 @@ from ee.onyx.server.billing.models import SeatUpdateRequest
 from ee.onyx.server.billing.models import SeatUpdateResponse
 from ee.onyx.server.billing.models import StripePublishableKeyResponse
 from ee.onyx.server.billing.models import SubscriptionStatusResponse
-from ee.onyx.server.billing.service import BillingServiceError
 from ee.onyx.server.billing.service import (
     create_checkout_session as create_checkout_service,
 )
@@ -58,6 +56,8 @@ from onyx.configs.app_configs import STRIPE_PUBLISHABLE_KEY_OVERRIDE
 from onyx.configs.app_configs import STRIPE_PUBLISHABLE_KEY_URL
 from onyx.configs.app_configs import WEB_DOMAIN
 from onyx.db.engine.sql_engine import get_session
+from onyx.error_handling.error_codes import OnyxErrorCode
+from onyx.error_handling.exceptions import OnyxError
 from onyx.redis.redis_pool import get_shared_redis_client
 from onyx.utils.logger import setup_logger
 from shared_configs.configs import MULTI_TENANT
@@ -169,26 +169,23 @@ async def create_checkout_session(
     if seats is not None:
         used_seats = get_used_seats(tenant_id)
         if seats < used_seats:
-            raise HTTPException(
-                status_code=400,
-                detail=f"Cannot subscribe with fewer seats than current usage. "
+            raise OnyxError(
+                OnyxErrorCode.VALIDATION_ERROR,
+                f"Cannot subscribe with fewer seats than current usage. "
                 f"You have {used_seats} active users/integrations but requested {seats} seats.",
             )
 
     # Build redirect URL for after checkout completion
     redirect_url = f"{WEB_DOMAIN}/admin/billing?checkout=success"
 
-    try:
-        return await create_checkout_service(
-            billing_period=billing_period,
-            seats=seats,
-            email=email,
-            license_data=license_data,
-            redirect_url=redirect_url,
-            tenant_id=tenant_id,
-        )
-    except BillingServiceError as e:
-        raise HTTPException(status_code=e.status_code, detail=e.message)
+    return await create_checkout_service(
+        billing_period=billing_period,
+        seats=seats,
+        email=email,
+        license_data=license_data,
+        redirect_url=redirect_url,
+        tenant_id=tenant_id,
+    )
 
 
 @router.post("/create-customer-portal-session")
@@ -206,18 +203,15 @@ async def create_customer_portal_session(
 
     # Self-hosted requires license
     if not MULTI_TENANT and not license_data:
-        raise HTTPException(status_code=400, detail="No license found")
+        raise OnyxError(OnyxErrorCode.VALIDATION_ERROR, "No license found")
 
     return_url = request.return_url if request else f"{WEB_DOMAIN}/admin/billing"
 
-    try:
-        return await create_portal_service(
-            license_data=license_data,
-            return_url=return_url,
-            tenant_id=tenant_id,
-        )
-    except BillingServiceError as e:
-        raise HTTPException(status_code=e.status_code, detail=e.message)
+    return await create_portal_service(
+        license_data=license_data,
+        return_url=return_url,
+        tenant_id=tenant_id,
+    )
 
 
 @router.get("/billing-information")
@@ -240,9 +234,9 @@ async def get_billing_information(
 
     # Check circuit breaker (self-hosted only)
     if _is_billing_circuit_open():
-        raise HTTPException(
-            status_code=503,
-            detail="Stripe connection temporarily disabled. Click 'Connect to Stripe' to retry.",
+        raise OnyxError(
+            OnyxErrorCode.SERVICE_UNAVAILABLE,
+            "Stripe connection temporarily disabled. Click 'Connect to Stripe' to retry.",
         )
 
     try:
@@ -250,11 +244,11 @@ async def get_billing_information(
             license_data=license_data,
             tenant_id=tenant_id,
         )
-    except BillingServiceError as e:
+    except OnyxError as e:
         # Open circuit breaker on connection failures (self-hosted only)
         if e.status_code in (502, 503, 504):
             _open_billing_circuit()
-        raise HTTPException(status_code=e.status_code, detail=e.message)
+        raise
 
 
 @router.post("/seats/update")
@@ -274,31 +268,25 @@ async def update_seats(
 
     # Self-hosted requires license
     if not MULTI_TENANT and not license_data:
-        raise HTTPException(status_code=400, detail="No license found")
+        raise OnyxError(OnyxErrorCode.VALIDATION_ERROR, "No license found")
 
     # Validate that new seat count is not less than current used seats
     used_seats = get_used_seats(tenant_id)
     if request.new_seat_count < used_seats:
-        raise HTTPException(
-            status_code=400,
-            detail=f"Cannot reduce seats below current usage. "
+        raise OnyxError(
+            OnyxErrorCode.VALIDATION_ERROR,
+            f"Cannot reduce seats below current usage. "
             f"You have {used_seats} active users/integrations but requested {request.new_seat_count} seats.",
         )
 
-    try:
-        result = await update_seat_service(
-            new_seat_count=request.new_seat_count,
-            license_data=license_data,
-            tenant_id=tenant_id,
-        )
-
-        # Note: Don't store license here - the control plane may still be processing
-        # the subscription update. The frontend should call /license/claim after a
-        # short delay to get the freshly generated license.
-
-        return result
-    except BillingServiceError as e:
-        raise HTTPException(status_code=e.status_code, detail=e.message)
+    # Note: Don't store license here - the control plane may still be processing
+    # the subscription update. The frontend should call /license/claim after a
+    # short delay to get the freshly generated license.
+    return await update_seat_service(
+        new_seat_count=request.new_seat_count,
+        license_data=license_data,
+        tenant_id=tenant_id,
+    )
 
 
 @router.get("/stripe-publishable-key")
@@ -329,18 +317,18 @@ async def get_stripe_publishable_key() -> StripePublishableKeyResponse:
         if STRIPE_PUBLISHABLE_KEY_OVERRIDE:
             key = STRIPE_PUBLISHABLE_KEY_OVERRIDE.strip()
             if not key.startswith("pk_"):
-                raise HTTPException(
-                    status_code=500,
-                    detail="Invalid Stripe publishable key format",
+                raise OnyxError(
+                    OnyxErrorCode.INTERNAL_ERROR,
+                    "Invalid Stripe publishable key format",
                 )
             _stripe_publishable_key_cache = key
             return StripePublishableKeyResponse(publishable_key=key)
 
         # Fall back to S3 bucket
         if not STRIPE_PUBLISHABLE_KEY_URL:
-            raise HTTPException(
-                status_code=500,
-                detail="Stripe publishable key is not configured",
+            raise OnyxError(
+                OnyxErrorCode.INTERNAL_ERROR,
+                "Stripe publishable key is not configured",
             )
 
         try:
@@ -351,17 +339,17 @@ async def get_stripe_publishable_key() -> StripePublishableKeyResponse:
 
                 # Validate key format
                 if not key.startswith("pk_"):
-                    raise HTTPException(
-                        status_code=500,
-                        detail="Invalid Stripe publishable key format",
+                    raise OnyxError(
+                        OnyxErrorCode.INTERNAL_ERROR,
+                        "Invalid Stripe publishable key format",
                     )
 
                 _stripe_publishable_key_cache = key
                 return StripePublishableKeyResponse(publishable_key=key)
         except httpx.HTTPError:
-            raise HTTPException(
-                status_code=500,
-                detail="Failed to fetch Stripe publishable key",
+            raise OnyxError(
+                OnyxErrorCode.INTERNAL_ERROR,
+                "Failed to fetch Stripe publishable key",
             )
 
 

backend/ee/onyx/server/billing/service.py

@@ -22,6 +22,8 @@ from ee.onyx.server.billing.models import SeatUpdateResponse
 from ee.onyx.server.billing.models import SubscriptionStatusResponse
 from ee.onyx.server.tenants.access import generate_data_plane_token
 from onyx.configs.app_configs import CONTROL_PLANE_API_BASE_URL
+from onyx.error_handling.error_codes import OnyxErrorCode
+from onyx.error_handling.exceptions import OnyxError
 from onyx.utils.logger import setup_logger
 from shared_configs.configs import MULTI_TENANT
 
@@ -31,15 +33,6 @@ logger = setup_logger()
 _REQUEST_TIMEOUT = 30.0
 
 
-class BillingServiceError(Exception):
-    """Exception raised for billing service errors."""
-
-    def __init__(self, message: str, status_code: int = 500):
-        self.message = message
-        self.status_code = status_code
-        super().__init__(self.message)
-
-
 def _get_proxy_headers(license_data: str | None) -> dict[str, str]:
     """Build headers for proxy requests (self-hosted).
 
@@ -101,7 +94,7 @@ async def _make_billing_request(
         Response JSON as dict
 
     Raises:
-        BillingServiceError: If request fails
+        OnyxError: If request fails
     """
 
     base_url = _get_base_url()
@@ -128,11 +121,17 @@ async def _make_billing_request(
         except Exception:
             pass
         logger.error(f"{error_message}: {e.response.status_code} - {detail}")
-        raise BillingServiceError(detail, e.response.status_code)
+        raise OnyxError(
+            OnyxErrorCode.BAD_GATEWAY,
+            detail,
+            status_code_override=e.response.status_code,
+        )
 
     except httpx.RequestError:
         logger.exception("Failed to connect to billing service")
-        raise BillingServiceError("Failed to connect to billing service", 502)
+        raise OnyxError(
+            OnyxErrorCode.BAD_GATEWAY, "Failed to connect to billing service"
+        )
 
 
 async def create_checkout_session(

backend/ee/onyx/server/enterprise_settings/api.py

@@ -223,6 +223,15 @@ def get_active_scim_token(
     token = dal.get_active_token()
     if not token:
         raise HTTPException(status_code=404, detail="No active SCIM token")
+
+    # Derive the IdP domain from the first synced user as a heuristic.
+    idp_domain: str | None = None
+    mappings, _total = dal.list_user_mappings(start_index=1, count=1)
+    if mappings:
+        user = dal.get_user(mappings[0].user_id)
+        if user and "@" in user.email:
+            idp_domain = user.email.rsplit("@", 1)[1]
+
     return ScimTokenResponse(
         id=token.id,
         name=token.name,
@@ -230,6 +239,7 @@ def get_active_scim_token(
         is_active=token.is_active,
         created_at=token.created_at,
         last_used_at=token.last_used_at,
+        idp_domain=idp_domain,
     )
 
 

backend/ee/onyx/server/license/api.py

@@ -14,7 +14,6 @@ import requests
 from fastapi import APIRouter
 from fastapi import Depends
 from fastapi import File
-from fastapi import HTTPException
 from fastapi import UploadFile
 from sqlalchemy.orm import Session
 
@@ -35,6 +34,8 @@ from ee.onyx.server.license.models import SeatUsageResponse
 from ee.onyx.utils.license import verify_license_signature
 from onyx.auth.users import User
 from onyx.db.engine.sql_engine import get_session
+from onyx.error_handling.error_codes import OnyxErrorCode
+from onyx.error_handling.exceptions import OnyxError
 from onyx.utils.logger import setup_logger
 from shared_configs.configs import MULTI_TENANT
 
@@ -127,9 +128,9 @@ async def claim_license(
     2. Without session_id: Re-claim using existing license for auth
     """
     if MULTI_TENANT:
-        raise HTTPException(
-            status_code=400,
-            detail="License claiming is only available for self-hosted deployments",
+        raise OnyxError(
+            OnyxErrorCode.VALIDATION_ERROR,
+            "License claiming is only available for self-hosted deployments",
         )
 
     try:
@@ -146,15 +147,16 @@ async def claim_license(
             # Re-claim using existing license for auth
             metadata = get_license_metadata(db_session)
             if not metadata or not metadata.tenant_id:
-                raise HTTPException(
-                    status_code=400,
-                    detail="No license found. Provide session_id after checkout.",
+                raise OnyxError(
+                    OnyxErrorCode.VALIDATION_ERROR,
+                    "No license found. Provide session_id after checkout.",
                 )
 
             license_row = get_license(db_session)
             if not license_row or not license_row.license_data:
-                raise HTTPException(
-                    status_code=400, detail="No license found in database"
+                raise OnyxError(
+                    OnyxErrorCode.VALIDATION_ERROR,
+                    "No license found in database",
                 )
 
             url = f"{CLOUD_DATA_PLANE_URL}/proxy/license/{metadata.tenant_id}"
@@ -173,7 +175,7 @@ async def claim_license(
         license_data = data.get("license")
 
         if not license_data:
-            raise HTTPException(status_code=404, detail="No license in response")
+            raise OnyxError(OnyxErrorCode.NOT_FOUND, "No license in response")
 
         # Verify signature before persisting
         payload = verify_license_signature(license_data)
@@ -199,12 +201,14 @@ async def claim_license(
             detail = error_data.get("detail", detail)
         except Exception:
             pass
-        raise HTTPException(status_code=status_code, detail=detail)
+        raise OnyxError(
+            OnyxErrorCode.BAD_GATEWAY, detail, status_code_override=status_code
+        )
     except ValueError as e:
-        raise HTTPException(status_code=400, detail=str(e))
+        raise OnyxError(OnyxErrorCode.VALIDATION_ERROR, str(e))
     except requests.RequestException:
-        raise HTTPException(
-            status_code=502, detail="Failed to connect to license server"
+        raise OnyxError(
+            OnyxErrorCode.BAD_GATEWAY, "Failed to connect to license server"
         )
 
 
@@ -221,9 +225,9 @@ async def upload_license(
     The license file must be cryptographically signed by Onyx.
     """
     if MULTI_TENANT:
-        raise HTTPException(
-            status_code=400,
-            detail="License upload is only available for self-hosted deployments",
+        raise OnyxError(
+            OnyxErrorCode.VALIDATION_ERROR,
+            "License upload is only available for self-hosted deployments",
         )
 
     try:
@@ -234,14 +238,14 @@ async def upload_license(
         # Remove any stray whitespace/newlines from user input
         license_data = license_data.strip()
     except UnicodeDecodeError:
-        raise HTTPException(status_code=400, detail="Invalid license file format")
+        raise OnyxError(OnyxErrorCode.INVALID_INPUT, "Invalid license file format")
 
     # Verify cryptographic signature - this is the only validation needed
     # The license's tenant_id identifies the customer in control plane, not locally
     try:
         payload = verify_license_signature(license_data)
     except ValueError as e:
-        raise HTTPException(status_code=400, detail=str(e))
+        raise OnyxError(OnyxErrorCode.VALIDATION_ERROR, str(e))
 
     # Persist to DB and update cache
     upsert_license(db_session, license_data)
@@ -297,9 +301,9 @@ async def delete_license(
     Admin only - removes license from database and invalidates cache.
     """
     if MULTI_TENANT:
-        raise HTTPException(
-            status_code=400,
-            detail="License deletion is only available for self-hosted deployments",
+        raise OnyxError(
+            OnyxErrorCode.VALIDATION_ERROR,
+            "License deletion is only available for self-hosted deployments",
         )
 
     try:

backend/ee/onyx/server/middleware/license_enforcement.py

@@ -46,7 +46,6 @@ from fastapi import FastAPI
 from fastapi import Request
 from fastapi import Response
 from fastapi.responses import JSONResponse
-from redis.exceptions import RedisError
 from sqlalchemy.exc import SQLAlchemyError
 
 from ee.onyx.configs.app_configs import LICENSE_ENFORCEMENT_ENABLED
@@ -56,6 +55,7 @@ from ee.onyx.configs.license_enforcement_config import (
 )
 from ee.onyx.db.license import get_cached_license_metadata
 from ee.onyx.db.license import refresh_license_cache
+from onyx.cache.interface import CACHE_TRANSIENT_ERRORS
 from onyx.db.engine.sql_engine import get_session_with_current_tenant
 from onyx.server.settings.models import ApplicationStatus
 from shared_configs.contextvars import get_current_tenant_id
@@ -164,9 +164,9 @@ def add_license_enforcement_middleware(
                     "[license_enforcement] No license, allowing community features"
                 )
                 is_gated = False
-        except RedisError as e:
+        except CACHE_TRANSIENT_ERRORS as e:
             logger.warning(f"Failed to check license metadata: {e}")
-            # Fail open - don't block users due to Redis connectivity issues
+            # Fail open - don't block users due to cache connectivity issues
             is_gated = False
 
         if is_gated:

backend/ee/onyx/server/scim/api.py

@@ -15,7 +15,9 @@ from uuid import UUID
 
 from fastapi import APIRouter
 from fastapi import Depends
+from fastapi import FastAPI
 from fastapi import Query
+from fastapi import Request
 from fastapi import Response
 from fastapi.responses import JSONResponse
 from fastapi_users.password import PasswordHelper
@@ -24,6 +26,7 @@ from sqlalchemy.exc import IntegrityError
 from sqlalchemy.orm import Session
 
 from ee.onyx.db.scim import ScimDAL
+from ee.onyx.server.scim.auth import ScimAuthError
 from ee.onyx.server.scim.auth import verify_scim_token
 from ee.onyx.server.scim.filtering import parse_scim_filter
 from ee.onyx.server.scim.models import SCIM_LIST_RESPONSE_SCHEMA
@@ -77,6 +80,22 @@ scim_router = APIRouter(prefix="/scim/v2", tags=["SCIM"])
 _pw_helper = PasswordHelper()
 
 
+def register_scim_exception_handlers(app: FastAPI) -> None:
+    """Register SCIM-specific exception handlers on the FastAPI app.
+
+    Call this after ``app.include_router(scim_router)`` so that auth
+    failures from ``verify_scim_token`` return RFC 7644 §3.12 error
+    envelopes (with ``schemas`` and ``status`` fields) instead of
+    FastAPI's default ``{"detail": "..."}`` format.
+    """
+
+    @app.exception_handler(ScimAuthError)
+    async def _handle_scim_auth_error(
+        _request: Request, exc: ScimAuthError
+    ) -> ScimJSONResponse:
+        return _scim_error_response(exc.status_code, exc.detail)
+
+
 def _get_provider(
     _token: ScimToken = Depends(verify_scim_token),
 ) -> ScimProvider:
@@ -404,21 +423,63 @@ def create_user(
 
     email = user_resource.userName.strip()
 
-    # externalId is how the IdP correlates this user on subsequent requests.
-    # Without it, the IdP can't find the user and will try to re-create,
-    # hitting a 409 conflict — so we require it up front.
-    if not user_resource.externalId:
-        return _scim_error_response(400, "externalId is required")
+    # Check for existing user — if they exist but aren't SCIM-managed yet,
+    # link them to the IdP rather than rejecting with 409.
+    external_id: str | None = user_resource.externalId
+    scim_username: str = user_resource.userName.strip()
+    fields: ScimMappingFields = _fields_from_resource(user_resource)
 
-    # Enforce seat limit
+    existing_user = dal.get_user_by_email(email)
+    if existing_user:
+        existing_mapping = dal.get_user_mapping_by_user_id(existing_user.id)
+        if existing_mapping:
+            return _scim_error_response(409, f"User with email {email} already exists")
+
+        # Adopt pre-existing user into SCIM management.
+        # Reactivating a deactivated user consumes a seat, so enforce the
+        # seat limit the same way replace_user does.
+        if user_resource.active and not existing_user.is_active:
+            seat_error = _check_seat_availability(dal)
+            if seat_error:
+                return _scim_error_response(403, seat_error)
+
+        personal_name = _scim_name_to_str(user_resource.name)
+        dal.update_user(
+            existing_user,
+            is_active=user_resource.active,
+            **({"personal_name": personal_name} if personal_name else {}),
+        )
+
+        try:
+            dal.create_user_mapping(
+                external_id=external_id,
+                user_id=existing_user.id,
+                scim_username=scim_username,
+                fields=fields,
+            )
+            dal.commit()
+        except IntegrityError:
+            dal.rollback()
+            return _scim_error_response(
+                409, f"User with email {email} already has a SCIM mapping"
+            )
+
+        return _scim_resource_response(
+            provider.build_user_resource(
+                existing_user,
+                external_id,
+                scim_username=scim_username,
+                fields=fields,
+            ),
+            status_code=201,
+        )
+
+    # Only enforce seat limit for net-new users — adopting a pre-existing
+    # user doesn't consume a new seat.
     seat_error = _check_seat_availability(dal)
     if seat_error:
         return _scim_error_response(403, seat_error)
 
-    # Check for existing user
-    if dal.get_user_by_email(email):
-        return _scim_error_response(409, f"User with email {email} already exists")
-
     # Create user with a random password (SCIM users authenticate via IdP)
     personal_name = _scim_name_to_str(user_resource.name)
     user = User(
@@ -436,18 +497,21 @@ def create_user(
         dal.rollback()
         return _scim_error_response(409, f"User with email {email} already exists")
 
-    # Create SCIM mapping (externalId is validated above, always present)
-    external_id = user_resource.externalId
-    scim_username = user_resource.userName.strip()
-    fields = _fields_from_resource(user_resource)
-    dal.create_user_mapping(
-        external_id=external_id,
-        user_id=user.id,
-        scim_username=scim_username,
-        fields=fields,
-    )
-
-    dal.commit()
+    # Always create a SCIM mapping so that the user is marked as
+    # SCIM-managed. externalId may be None (RFC 7643 says it's optional).
+    try:
+        dal.create_user_mapping(
+            external_id=external_id,
+            user_id=user.id,
+            scim_username=scim_username,
+            fields=fields,
+        )
+        dal.commit()
+    except IntegrityError:
+        dal.rollback()
+        return _scim_error_response(
+            409, f"User with email {email} already has a SCIM mapping"
+        )
 
     return _scim_resource_response(
         provider.build_user_resource(

backend/ee/onyx/server/scim/auth.py

@@ -19,7 +19,6 @@ import hashlib
 import secrets
 
 from fastapi import Depends
-from fastapi import HTTPException
 from fastapi import Request
 from sqlalchemy.orm import Session
 
@@ -28,6 +27,21 @@ from onyx.auth.utils import get_hashed_bearer_token_from_request
 from onyx.db.engine.sql_engine import get_session
 from onyx.db.models import ScimToken
 
+
+class ScimAuthError(Exception):
+    """Raised when SCIM bearer token authentication fails.
+
+    Unlike HTTPException, this carries the status and detail so the SCIM
+    exception handler can wrap them in an RFC 7644 §3.12 error envelope
+    with ``schemas`` and ``status`` fields.
+    """
+
+    def __init__(self, status_code: int, detail: str) -> None:
+        self.status_code = status_code
+        self.detail = detail
+        super().__init__(detail)
+
+
 SCIM_TOKEN_PREFIX = "onyx_scim_"
 SCIM_TOKEN_LENGTH = 48
 
@@ -82,23 +96,14 @@ def verify_scim_token(
     """
     hashed = _get_hashed_scim_token_from_request(request)
     if not hashed:
-        raise HTTPException(
-            status_code=401,
-            detail="Missing or invalid SCIM bearer token",
-        )
+        raise ScimAuthError(401, "Missing or invalid SCIM bearer token")
 
     token = dal.get_token_by_hash(hashed)
 
     if not token:
-        raise HTTPException(
-            status_code=401,
-            detail="Invalid SCIM bearer token",
-        )
+        raise ScimAuthError(401, "Invalid SCIM bearer token")
 
     if not token.is_active:
-        raise HTTPException(
-            status_code=401,
-            detail="SCIM token has been revoked",
-        )
+        raise ScimAuthError(401, "SCIM token has been revoked")
 
     return token

backend/ee/onyx/server/scim/models.py

@@ -365,6 +365,7 @@ class ScimTokenResponse(BaseModel):
     is_active: bool
     created_at: datetime
     last_used_at: datetime | None = None
+    idp_domain: str | None = None
 
 
 class ScimTokenCreatedResponse(ScimTokenResponse):

backend/ee/onyx/server/scim/providers/base.py

@@ -153,26 +153,31 @@ class ScimProvider(ABC):
         self,
         user: User,
         fields: ScimMappingFields,
-    ) -> ScimName | None:
+    ) -> ScimName:
         """Build SCIM name components for the response.
 
         Round-trips stored ``given_name``/``family_name`` when available (so
         the IdP gets back what it sent). Falls back to splitting
         ``personal_name`` for users provisioned before we stored components.
+        Always returns a ScimName — Okta's spec tests expect ``name``
+        (with ``givenName``/``familyName``) on every user resource.
         Providers may override for custom behavior.
         """
         if fields.given_name is not None or fields.family_name is not None:
             return ScimName(
-                givenName=fields.given_name,
-                familyName=fields.family_name,
-                formatted=user.personal_name,
+                givenName=fields.given_name or "",
+                familyName=fields.family_name or "",
+                formatted=user.personal_name or "",
             )
         if not user.personal_name:
-            return None
+            # Derive a reasonable name from the email so that SCIM spec tests
+            # see non-empty givenName / familyName for every user resource.
+            local = user.email.split("@")[0] if user.email else ""
+            return ScimName(givenName=local, familyName="", formatted=local)
         parts = user.personal_name.split(" ", 1)
         return ScimName(
             givenName=parts[0],
-            familyName=parts[1] if len(parts) > 1 else None,
+            familyName=parts[1] if len(parts) > 1 else "",
             formatted=user.personal_name,
         )
 

backend/ee/onyx/server/seeding.py

@@ -18,7 +18,6 @@ from ee.onyx.server.enterprise_settings.store import (
     store_settings as store_ee_settings,
 )
 from ee.onyx.server.enterprise_settings.store import upload_logo
-from onyx.context.search.enums import RecencyBiasSetting
 from onyx.db.engine.sql_engine import get_session_with_current_tenant
 from onyx.db.llm import fetch_existing_llm_provider
 from onyx.db.llm import update_default_provider
@@ -161,12 +160,6 @@ def _seed_personas(db_session: Session, personas: list[PersonaUpsertRequest]) ->
                     user=None,  # Seeding is done as admin
                     name=persona.name,
                     description=persona.description,
-                    num_chunks=(
-                        persona.num_chunks if persona.num_chunks is not None else 0.0
-                    ),
-                    llm_relevance_filter=persona.llm_relevance_filter,
-                    llm_filter_extraction=persona.llm_filter_extraction,
-                    recency_bias=RecencyBiasSetting.AUTO,
                     document_set_ids=persona.document_set_ids,
                     llm_model_provider_override=persona.llm_model_provider_override,
                     llm_model_version_override=persona.llm_model_version_override,
@@ -178,6 +171,7 @@ def _seed_personas(db_session: Session, personas: list[PersonaUpsertRequest]) ->
                     system_prompt=persona.system_prompt,
                     task_prompt=persona.task_prompt,
                     datetime_aware=persona.datetime_aware,
+                    featured=persona.featured,
                     commit=False,
                 )
             db_session.commit()

backend/ee/onyx/server/settings/api.py

@@ -6,6 +6,7 @@ from sqlalchemy.exc import SQLAlchemyError
 from ee.onyx.configs.app_configs import LICENSE_ENFORCEMENT_ENABLED
 from ee.onyx.db.license import get_cached_license_metadata
 from ee.onyx.db.license import refresh_license_cache
+from onyx.cache.interface import CACHE_TRANSIENT_ERRORS
 from onyx.configs.app_configs import ENTERPRISE_EDITION_ENABLED
 from onyx.db.engine.sql_engine import get_session_with_current_tenant
 from onyx.server.settings.models import ApplicationStatus
@@ -125,7 +126,7 @@ def apply_license_status_to_settings(settings: Settings) -> Settings:
                 # syncing) means indexed data may need protection.
                 settings.application_status = _BLOCKING_STATUS
             settings.ee_features_enabled = False
-    except RedisError as e:
+    except CACHE_TRANSIENT_ERRORS as e:
         logger.warning(f"Failed to check license metadata for settings: {e}")
         # Fail closed - disable EE features if we can't verify license
         settings.ee_features_enabled = False

backend/ee/onyx/server/tenants/billing_api.py

@@ -21,7 +21,6 @@ import asyncio
 import httpx
 from fastapi import APIRouter
 from fastapi import Depends
-from fastapi import HTTPException
 
 from ee.onyx.auth.users import current_admin_user
 from ee.onyx.server.tenants.access import control_plane_dep
@@ -43,6 +42,8 @@ from onyx.auth.users import User
 from onyx.configs.app_configs import STRIPE_PUBLISHABLE_KEY_OVERRIDE
 from onyx.configs.app_configs import STRIPE_PUBLISHABLE_KEY_URL
 from onyx.configs.app_configs import WEB_DOMAIN
+from onyx.error_handling.error_codes import OnyxErrorCode
+from onyx.error_handling.exceptions import OnyxError
 from onyx.utils.logger import setup_logger
 from shared_configs.contextvars import CURRENT_TENANT_ID_CONTEXTVAR
 from shared_configs.contextvars import get_current_tenant_id
@@ -116,9 +117,14 @@ async def create_customer_portal_session(
     try:
         portal_url = fetch_customer_portal_session(tenant_id, return_url)
         return {"stripe_customer_portal_url": portal_url}
-    except Exception as e:
+    except OnyxError:
+        raise
+    except Exception:
         logger.exception("Failed to create customer portal session")
-        raise HTTPException(status_code=500, detail=str(e))
+        raise OnyxError(
+            OnyxErrorCode.INTERNAL_ERROR,
+            "Failed to create customer portal session",
+        )
 
 
 @router.post("/create-checkout-session")
@@ -134,9 +140,14 @@ async def create_checkout_session(
     try:
         checkout_url = fetch_stripe_checkout_session(tenant_id, billing_period, seats)
         return {"stripe_checkout_url": checkout_url}
-    except Exception as e:
+    except OnyxError:
+        raise
+    except Exception:
         logger.exception("Failed to create checkout session")
-        raise HTTPException(status_code=500, detail=str(e))
+        raise OnyxError(
+            OnyxErrorCode.INTERNAL_ERROR,
+            "Failed to create checkout session",
+        )
 
 
 @router.post("/create-subscription-session")
@@ -147,15 +158,20 @@ async def create_subscription_session(
     try:
         tenant_id = CURRENT_TENANT_ID_CONTEXTVAR.get()
         if not tenant_id:
-            raise HTTPException(status_code=400, detail="Tenant ID not found")
+            raise OnyxError(OnyxErrorCode.VALIDATION_ERROR, "Tenant ID not found")
 
         billing_period = request.billing_period if request else "monthly"
         session_id = fetch_stripe_checkout_session(tenant_id, billing_period)
         return SubscriptionSessionResponse(sessionId=session_id)
 
-    except Exception as e:
+    except OnyxError:
+        raise
+    except Exception:
         logger.exception("Failed to create subscription session")
-        raise HTTPException(status_code=500, detail=str(e))
+        raise OnyxError(
+            OnyxErrorCode.INTERNAL_ERROR,
+            "Failed to create subscription session",
+        )
 
 
 @router.get("/stripe-publishable-key")
@@ -186,18 +202,18 @@ async def get_stripe_publishable_key() -> StripePublishableKeyResponse:
         if STRIPE_PUBLISHABLE_KEY_OVERRIDE:
             key = STRIPE_PUBLISHABLE_KEY_OVERRIDE.strip()
             if not key.startswith("pk_"):
-                raise HTTPException(
-                    status_code=500,
-                    detail="Invalid Stripe publishable key format",
+                raise OnyxError(
+                    OnyxErrorCode.INTERNAL_ERROR,
+                    "Invalid Stripe publishable key format",
                 )
             _stripe_publishable_key_cache = key
             return StripePublishableKeyResponse(publishable_key=key)
 
         # Fall back to S3 bucket
         if not STRIPE_PUBLISHABLE_KEY_URL:
-            raise HTTPException(
-                status_code=500,
-                detail="Stripe publishable key is not configured",
+            raise OnyxError(
+                OnyxErrorCode.INTERNAL_ERROR,
+                "Stripe publishable key is not configured",
             )
 
         try:
@@ -208,15 +224,15 @@ async def get_stripe_publishable_key() -> StripePublishableKeyResponse:
 
                 # Validate key format
                 if not key.startswith("pk_"):
-                    raise HTTPException(
-                        status_code=500,
-                        detail="Invalid Stripe publishable key format",
+                    raise OnyxError(
+                        OnyxErrorCode.INTERNAL_ERROR,
+                        "Invalid Stripe publishable key format",
                     )
 
                 _stripe_publishable_key_cache = key
                 return StripePublishableKeyResponse(publishable_key=key)
         except httpx.HTTPError:
-            raise HTTPException(
-                status_code=500,
-                detail="Failed to fetch Stripe publishable key",
+            raise OnyxError(
+                OnyxErrorCode.INTERNAL_ERROR,
+                "Failed to fetch Stripe publishable key",
             )

backend/ee/onyx/server/user_group/api.py

@@ -5,6 +5,8 @@ from sqlalchemy.exc import IntegrityError
 from sqlalchemy.orm import Session
 
 from ee.onyx.db.user_group import add_users_to_user_group
+from ee.onyx.db.user_group import delete_user_group as db_delete_user_group
+from ee.onyx.db.user_group import fetch_user_group
 from ee.onyx.db.user_group import fetch_user_groups
 from ee.onyx.db.user_group import fetch_user_groups_for_user
 from ee.onyx.db.user_group import insert_user_group
@@ -20,6 +22,7 @@ from ee.onyx.server.user_group.models import UserGroupUpdate
 from onyx.auth.users import current_admin_user
 from onyx.auth.users import current_curator_or_admin_user
 from onyx.auth.users import current_user
+from onyx.configs.app_configs import DISABLE_VECTOR_DB
 from onyx.configs.constants import PUBLIC_API_TAGS
 from onyx.db.engine.sql_engine import get_session
 from onyx.db.models import User
@@ -153,3 +156,8 @@ def delete_user_group(
         prepare_user_group_for_deletion(db_session, user_group_id)
     except ValueError as e:
         raise HTTPException(status_code=404, detail=str(e))
+
+    if DISABLE_VECTOR_DB:
+        user_group = fetch_user_group(db_session, user_group_id)
+        if user_group:
+            db_delete_user_group(db_session, user_group)

backend/onyx/auth/users.py

@@ -120,7 +120,6 @@ from onyx.db.models import User
 from onyx.db.pat import fetch_user_for_pat
 from onyx.db.users import get_user_by_email
 from onyx.redis.redis_pool import get_async_redis_connection
-from onyx.redis.redis_pool import get_redis_client
 from onyx.server.settings.store import load_settings
 from onyx.server.utils import BasicAuthenticationError
 from onyx.utils.logger import setup_logger
@@ -201,13 +200,14 @@ def user_needs_to_be_verified() -> bool:
 
 
 def anonymous_user_enabled(*, tenant_id: str | None = None) -> bool:
-    redis_client = get_redis_client(tenant_id=tenant_id)
-    value = redis_client.get(OnyxRedisLocks.ANONYMOUS_USER_ENABLED)
+    from onyx.cache.factory import get_cache_backend
+
+    cache = get_cache_backend(tenant_id=tenant_id)
+    value = cache.get(OnyxRedisLocks.ANONYMOUS_USER_ENABLED)
 
     if value is None:
         return False
 
-    assert isinstance(value, bytes)
     return int(value.decode("utf-8")) == 1
 
 
@@ -543,7 +543,7 @@ class UserManager(UUIDIDMixin, BaseUserManager[User, uuid.UUID]):
         result = await db_session.execute(
             select(Persona.id)
             .where(
-                Persona.is_default_persona.is_(True),
+                Persona.featured.is_(True),
                 Persona.is_public.is_(True),
                 Persona.is_visible.is_(True),
                 Persona.deleted.is_(False),
@@ -725,11 +725,19 @@ class UserManager(UUIDIDMixin, BaseUserManager[User, uuid.UUID]):
                         if user_by_session:
                             user = user_by_session
 
+                # If the user is inactive, check seat availability before
+                # upgrading role — otherwise they'd become an inactive BASIC
+                # user who still can't log in.
+                if not user.is_active:
+                    with get_session_with_current_tenant() as sync_db:
+                        enforce_seat_limit(sync_db)
+
                 await self.user_db.update(
                     user,
                     {
                         "is_verified": is_verified_by_default,
                         "role": UserRole.BASIC,
+                        **({"is_active": True} if not user.is_active else {}),
                     },
                 )
 

backend/onyx/background/celery/apps/background.py

@@ -1,142 +0,0 @@
-from typing import Any
-from typing import cast
-
-from celery import Celery
-from celery import signals
-from celery import Task
-from celery.apps.worker import Worker
-from celery.signals import celeryd_init
-from celery.signals import worker_init
-from celery.signals import worker_process_init
-from celery.signals import worker_ready
-from celery.signals import worker_shutdown
-
-import onyx.background.celery.apps.app_base as app_base
-from onyx.background.celery.celery_utils import httpx_init_vespa_pool
-from onyx.configs.app_configs import MANAGED_VESPA
-from onyx.configs.app_configs import VESPA_CLOUD_CERT_PATH
-from onyx.configs.app_configs import VESPA_CLOUD_KEY_PATH
-from onyx.configs.constants import POSTGRES_CELERY_WORKER_BACKGROUND_APP_NAME
-from onyx.db.engine.sql_engine import SqlEngine
-from onyx.utils.logger import setup_logger
-from shared_configs.configs import MULTI_TENANT
-
-
-logger = setup_logger()
-
-celery_app = Celery(__name__)
-celery_app.config_from_object("onyx.background.celery.configs.background")
-celery_app.Task = app_base.TenantAwareTask  # type: ignore [misc]
-
-
-@signals.task_prerun.connect
-def on_task_prerun(
-    sender: Any | None = None,
-    task_id: str | None = None,
-    task: Task | None = None,
-    args: tuple | None = None,
-    kwargs: dict | None = None,
-    **kwds: Any,
-) -> None:
-    app_base.on_task_prerun(sender, task_id, task, args, kwargs, **kwds)
-
-
-@signals.task_postrun.connect
-def on_task_postrun(
-    sender: Any | None = None,
-    task_id: str | None = None,
-    task: Task | None = None,
-    args: tuple | None = None,
-    kwargs: dict | None = None,
-    retval: Any | None = None,
-    state: str | None = None,
-    **kwds: Any,
-) -> None:
-    app_base.on_task_postrun(sender, task_id, task, args, kwargs, retval, state, **kwds)
-
-
-@celeryd_init.connect
-def on_celeryd_init(sender: str, conf: Any = None, **kwargs: Any) -> None:
-    app_base.on_celeryd_init(sender, conf, **kwargs)
-
-
-@worker_init.connect
-def on_worker_init(sender: Worker, **kwargs: Any) -> None:
-    EXTRA_CONCURRENCY = 8  # small extra fudge factor for connection limits
-
-    logger.info("worker_init signal received for consolidated background worker.")
-
-    SqlEngine.set_app_name(POSTGRES_CELERY_WORKER_BACKGROUND_APP_NAME)
-    pool_size = cast(int, sender.concurrency)  # type: ignore
-    SqlEngine.init_engine(pool_size=pool_size, max_overflow=EXTRA_CONCURRENCY)
-
-    # Initialize Vespa httpx pool (needed for light worker tasks)
-    if MANAGED_VESPA:
-        httpx_init_vespa_pool(
-            sender.concurrency + EXTRA_CONCURRENCY,  # type: ignore
-            ssl_cert=VESPA_CLOUD_CERT_PATH,
-            ssl_key=VESPA_CLOUD_KEY_PATH,
-        )
-    else:
-        httpx_init_vespa_pool(sender.concurrency + EXTRA_CONCURRENCY)  # type: ignore
-
-    app_base.wait_for_redis(sender, **kwargs)
-    app_base.wait_for_db(sender, **kwargs)
-    app_base.wait_for_vespa_or_shutdown(sender, **kwargs)
-
-    # Less startup checks in multi-tenant case
-    if MULTI_TENANT:
-        return
-
-    app_base.on_secondary_worker_init(sender, **kwargs)
-
-
-@worker_ready.connect
-def on_worker_ready(sender: Any, **kwargs: Any) -> None:
-    app_base.on_worker_ready(sender, **kwargs)
-
-
-@worker_shutdown.connect
-def on_worker_shutdown(sender: Any, **kwargs: Any) -> None:
-    app_base.on_worker_shutdown(sender, **kwargs)
-
-
-@worker_process_init.connect
-def init_worker(**kwargs: Any) -> None:  # noqa: ARG001
-    SqlEngine.reset_engine()
-
-
-@signals.setup_logging.connect
-def on_setup_logging(
-    loglevel: Any, logfile: Any, format: Any, colorize: Any, **kwargs: Any
-) -> None:
-    app_base.on_setup_logging(loglevel, logfile, format, colorize, **kwargs)
-
-
-base_bootsteps = app_base.get_bootsteps()
-for bootstep in base_bootsteps:
-    celery_app.steps["worker"].add(bootstep)
-
-celery_app.autodiscover_tasks(
-    app_base.filter_task_modules(
-        [
-            # Original background worker tasks
-            "onyx.background.celery.tasks.pruning",
-            "onyx.background.celery.tasks.monitoring",
-            "onyx.background.celery.tasks.user_file_processing",
-            "onyx.background.celery.tasks.llm_model_update",
-            # Light worker tasks
-            "onyx.background.celery.tasks.shared",
-            "onyx.background.celery.tasks.vespa",
-            "onyx.background.celery.tasks.connector_deletion",
-            "onyx.background.celery.tasks.doc_permission_syncing",
-            "onyx.background.celery.tasks.opensearch_migration",
-            # Docprocessing worker tasks
-            "onyx.background.celery.tasks.docprocessing",
-            # Docfetching worker tasks
-            "onyx.background.celery.tasks.docfetching",
-            # Sandbox cleanup tasks (isolated in build feature)
-            "onyx.server.features.build.sandbox.tasks",
-        ]
-    )
-)

backend/onyx/background/celery/configs/background.py

@@ -1,23 +0,0 @@
-import onyx.background.celery.configs.base as shared_config
-from onyx.configs.app_configs import CELERY_WORKER_BACKGROUND_CONCURRENCY
-
-broker_url = shared_config.broker_url
-broker_connection_retry_on_startup = shared_config.broker_connection_retry_on_startup
-broker_pool_limit = shared_config.broker_pool_limit
-broker_transport_options = shared_config.broker_transport_options
-
-redis_socket_keepalive = shared_config.redis_socket_keepalive
-redis_retry_on_timeout = shared_config.redis_retry_on_timeout
-redis_backend_health_check_interval = shared_config.redis_backend_health_check_interval
-
-result_backend = shared_config.result_backend
-result_expires = shared_config.result_expires  # 86400 seconds is the default
-
-task_default_priority = shared_config.task_default_priority
-task_acks_late = shared_config.task_acks_late
-
-worker_concurrency = CELERY_WORKER_BACKGROUND_CONCURRENCY
-worker_pool = "threads"
-# Increased from 1 to 4 to handle fast light worker tasks more efficiently
-# This allows the worker to prefetch multiple tasks per thread
-worker_prefetch_multiplier = 4

backend/onyx/background/celery/tasks/beat_schedule.py

@@ -241,8 +241,7 @@ _VECTOR_DB_BEAT_TASK_NAMES: set[str] = {
     "check-for-index-attempt-cleanup",
     "check-for-doc-permissions-sync",
     "check-for-external-group-sync",
-    "check-for-documents-for-opensearch-migration",
-    "migrate-documents-from-vespa-to-opensearch",
+    "migrate-chunks-from-vespa-to-opensearch",
 }
 
 if DISABLE_VECTOR_DB:

backend/onyx/background/celery/tasks/opensearch_migration/tasks.py

@@ -30,6 +30,7 @@ from onyx.background.celery.tasks.opensearch_migration.transformer import (
     transform_vespa_chunks_to_opensearch_chunks,
 )
 from onyx.configs.app_configs import ENABLE_OPENSEARCH_INDEXING_FOR_ONYX
+from onyx.configs.app_configs import VESPA_MIGRATION_REQUEST_TIMEOUT_S
 from onyx.configs.constants import OnyxCeleryTask
 from onyx.configs.constants import OnyxRedisLocks
 from onyx.db.engine.sql_engine import get_session_with_current_tenant
@@ -47,6 +48,7 @@ from onyx.document_index.interfaces_new import TenantState
 from onyx.document_index.opensearch.opensearch_document_index import (
     OpenSearchDocumentIndex,
 )
+from onyx.document_index.vespa.shared_utils.utils import get_vespa_http_client
 from onyx.document_index.vespa.vespa_document_index import VespaDocumentIndex
 from onyx.indexing.models import IndexingSetting
 from onyx.redis.redis_pool import get_redis_client
@@ -146,7 +148,12 @@ def migrate_chunks_from_vespa_to_opensearch_task(
             task_logger.error(err_str)
             return False
 
-        with get_session_with_current_tenant() as db_session:
+        with (
+            get_session_with_current_tenant() as db_session,
+            get_vespa_http_client(
+                timeout=VESPA_MIGRATION_REQUEST_TIMEOUT_S
+            ) as vespa_client,
+        ):
             try_insert_opensearch_tenant_migration_record_with_commit(db_session)
             search_settings = get_current_search_settings(db_session)
             tenant_state = TenantState(tenant_id=tenant_id, multitenant=MULTI_TENANT)
@@ -161,6 +168,7 @@ def migrate_chunks_from_vespa_to_opensearch_task(
                 index_name=search_settings.index_name,
                 tenant_state=tenant_state,
                 large_chunks_enabled=False,
+                httpx_client=vespa_client,
             )
 
             sanitized_doc_start_time = time.monotonic()

backend/onyx/background/celery/tasks/user_file_processing/tasks.py

@@ -414,34 +414,31 @@ def _process_user_file_with_indexing(
         raise RuntimeError(f"Indexing pipeline failed for user file {user_file_id}")
 
 
-@shared_task(
-    name=OnyxCeleryTask.PROCESS_SINGLE_USER_FILE,
-    bind=True,
-    ignore_result=True,
-)
-def process_single_user_file(
-    self: Task, *, user_file_id: str, tenant_id: str  # noqa: ARG001
+def process_user_file_impl(
+    *, user_file_id: str, tenant_id: str, redis_locking: bool
 ) -> None:
-    task_logger.info(f"process_single_user_file - Starting id={user_file_id}")
+    """Core implementation for processing a single user file.
+
+    When redis_locking=True, acquires a per-file Redis lock and clears the
+    queued-key guard (Celery path).  When redis_locking=False, skips all Redis
+    operations (BackgroundTask path).
+    """
+    task_logger.info(f"process_user_file_impl - Starting id={user_file_id}")
     start = time.monotonic()
 
-    redis_client = get_redis_client(tenant_id=tenant_id)
-
-    # Clear the "queued" guard set by the beat generator so that the next beat
-    # cycle can re-enqueue this file if it is still in PROCESSING state after
-    # this task completes or fails.
-    redis_client.delete(_user_file_queued_key(user_file_id))
-
-    file_lock: RedisLock = redis_client.lock(
-        _user_file_lock_key(user_file_id),
-        timeout=CELERY_USER_FILE_PROCESSING_LOCK_TIMEOUT,
-    )
-
-    if not file_lock.acquire(blocking=False):
-        task_logger.info(
-            f"process_single_user_file - Lock held, skipping user_file_id={user_file_id}"
+    file_lock: RedisLock | None = None
+    if redis_locking:
+        redis_client = get_redis_client(tenant_id=tenant_id)
+        redis_client.delete(_user_file_queued_key(user_file_id))
+        file_lock = redis_client.lock(
+            _user_file_lock_key(user_file_id),
+            timeout=CELERY_USER_FILE_PROCESSING_LOCK_TIMEOUT,
         )
-        return None
+        if file_lock is not None and not file_lock.acquire(blocking=False):
+            task_logger.info(
+                f"process_user_file_impl - Lock held, skipping user_file_id={user_file_id}"
+            )
+            return
 
     documents: list[Document] = []
     try:
@@ -449,15 +446,18 @@ def process_single_user_file(
             uf = db_session.get(UserFile, _as_uuid(user_file_id))
             if not uf:
                 task_logger.warning(
-                    f"process_single_user_file - UserFile not found id={user_file_id}"
+                    f"process_user_file_impl - UserFile not found id={user_file_id}"
                 )
-                return None
+                return
 
-            if uf.status != UserFileStatus.PROCESSING:
+            if uf.status not in (
+                UserFileStatus.PROCESSING,
+                UserFileStatus.INDEXING,
+            ):
                 task_logger.info(
-                    f"process_single_user_file - Skipping id={user_file_id} status={uf.status}"
+                    f"process_user_file_impl - Skipping id={user_file_id} status={uf.status}"
                 )
-                return None
+                return
 
             connector = LocalFileConnector(
                 file_locations=[uf.file_id],
@@ -471,7 +471,6 @@ def process_single_user_file(
                         [doc for doc in batch if not isinstance(doc, HierarchyNode)]
                     )
 
-                # update the document id to userfile id in the documents
                 for document in documents:
                     document.id = str(user_file_id)
                     document.source = DocumentSource.USER_FILE
@@ -493,9 +492,8 @@ def process_single_user_file(
 
             except Exception as e:
                 task_logger.exception(
-                    f"process_single_user_file - Error processing file id={user_file_id} - {e.__class__.__name__}"
+                    f"process_user_file_impl - Error processing file id={user_file_id} - {e.__class__.__name__}"
                 )
-                # don't update the status if the user file is being deleted
                 current_user_file = db_session.get(UserFile, _as_uuid(user_file_id))
                 if (
                     current_user_file
@@ -504,33 +502,43 @@ def process_single_user_file(
                     uf.status = UserFileStatus.FAILED
                     db_session.add(uf)
                     db_session.commit()
-                return None
+                return
 
         elapsed = time.monotonic() - start
         task_logger.info(
-            f"process_single_user_file - Finished id={user_file_id} docs={len(documents)} elapsed={elapsed:.2f}s"
+            f"process_user_file_impl - Finished id={user_file_id} docs={len(documents)} elapsed={elapsed:.2f}s"
         )
-        return None
     except Exception as e:
-        # Attempt to mark the file as failed
         with get_session_with_current_tenant() as db_session:
             uf = db_session.get(UserFile, _as_uuid(user_file_id))
             if uf:
-                # don't update the status if the user file is being deleted
                 if uf.status != UserFileStatus.DELETING:
                     uf.status = UserFileStatus.FAILED
                 db_session.add(uf)
                 db_session.commit()
 
         task_logger.exception(
-            f"process_single_user_file - Error processing file id={user_file_id} - {e.__class__.__name__}"
+            f"process_user_file_impl - Error processing file id={user_file_id} - {e.__class__.__name__}"
         )
-        return None
+        raise
     finally:
-        if file_lock.owned():
+        if file_lock is not None and file_lock.owned():
             file_lock.release()
 
 
+@shared_task(
+    name=OnyxCeleryTask.PROCESS_SINGLE_USER_FILE,
+    bind=True,
+    ignore_result=True,
+)
+def process_single_user_file(
+    self: Task, *, user_file_id: str, tenant_id: str  # noqa: ARG001
+) -> None:
+    process_user_file_impl(
+        user_file_id=user_file_id, tenant_id=tenant_id, redis_locking=True
+    )
+
+
 @shared_task(
     name=OnyxCeleryTask.CHECK_FOR_USER_FILE_DELETE,
     soft_time_limit=300,
@@ -581,36 +589,38 @@ def check_for_user_file_delete(self: Task, *, tenant_id: str) -> None:
     return None
 
 
-@shared_task(
-    name=OnyxCeleryTask.DELETE_SINGLE_USER_FILE,
-    bind=True,
-    ignore_result=True,
-)
-def process_single_user_file_delete(
-    self: Task, *, user_file_id: str, tenant_id: str  # noqa: ARG001
+def delete_user_file_impl(
+    *, user_file_id: str, tenant_id: str, redis_locking: bool
 ) -> None:
-    """Process a single user file delete."""
-    task_logger.info(f"process_single_user_file_delete - Starting id={user_file_id}")
-    redis_client = get_redis_client(tenant_id=tenant_id)
-    file_lock: RedisLock = redis_client.lock(
-        _user_file_delete_lock_key(user_file_id),
-        timeout=CELERY_GENERIC_BEAT_LOCK_TIMEOUT,
-    )
-    if not file_lock.acquire(blocking=False):
-        task_logger.info(
-            f"process_single_user_file_delete - Lock held, skipping user_file_id={user_file_id}"
+    """Core implementation for deleting a single user file.
+
+    When redis_locking=True, acquires a per-file Redis lock (Celery path).
+    When redis_locking=False, skips Redis operations (BackgroundTask path).
+    """
+    task_logger.info(f"delete_user_file_impl - Starting id={user_file_id}")
+
+    file_lock: RedisLock | None = None
+    if redis_locking:
+        redis_client = get_redis_client(tenant_id=tenant_id)
+        file_lock = redis_client.lock(
+            _user_file_delete_lock_key(user_file_id),
+            timeout=CELERY_GENERIC_BEAT_LOCK_TIMEOUT,
         )
-        return None
+        if file_lock is not None and not file_lock.acquire(blocking=False):
+            task_logger.info(
+                f"delete_user_file_impl - Lock held, skipping user_file_id={user_file_id}"
+            )
+            return
+
     try:
         with get_session_with_current_tenant() as db_session:
             user_file = db_session.get(UserFile, _as_uuid(user_file_id))
             if not user_file:
                 task_logger.info(
-                    f"process_single_user_file_delete - User file not found id={user_file_id}"
+                    f"delete_user_file_impl - User file not found id={user_file_id}"
                 )
-                return None
+                return
 
-            # 1) Delete vector DB chunks (skip when disabled)
             if not DISABLE_VECTOR_DB:
                 if MANAGED_VESPA:
                     httpx_init_vespa_pool(
@@ -648,7 +658,6 @@ def process_single_user_file_delete(
                         chunk_count=chunk_count,
                     )
 
-            # 2) Delete the user-uploaded file content from filestore (blob + metadata)
             file_store = get_default_file_store()
             try:
                 file_store.delete_file(user_file.file_id)
@@ -656,26 +665,34 @@ def process_single_user_file_delete(
                     user_file_id_to_plaintext_file_name(user_file.id)
                 )
             except Exception as e:
-                # This block executed only if the file is not found in the filestore
                 task_logger.exception(
-                    f"process_single_user_file_delete - Error deleting file id={user_file.id} - {e.__class__.__name__}"
+                    f"delete_user_file_impl - Error deleting file id={user_file.id} - {e.__class__.__name__}"
                 )
 
-            # 3) Finally, delete the UserFile row
             db_session.delete(user_file)
             db_session.commit()
-            task_logger.info(
-                f"process_single_user_file_delete - Completed id={user_file_id}"
-            )
+            task_logger.info(f"delete_user_file_impl - Completed id={user_file_id}")
     except Exception as e:
         task_logger.exception(
-            f"process_single_user_file_delete - Error processing file id={user_file_id} - {e.__class__.__name__}"
+            f"delete_user_file_impl - Error processing file id={user_file_id} - {e.__class__.__name__}"
         )
-        return None
+        raise
     finally:
-        if file_lock.owned():
+        if file_lock is not None and file_lock.owned():
             file_lock.release()
-    return None
+
+
+@shared_task(
+    name=OnyxCeleryTask.DELETE_SINGLE_USER_FILE,
+    bind=True,
+    ignore_result=True,
+)
+def process_single_user_file_delete(
+    self: Task, *, user_file_id: str, tenant_id: str  # noqa: ARG001
+) -> None:
+    delete_user_file_impl(
+        user_file_id=user_file_id, tenant_id=tenant_id, redis_locking=True
+    )
 
 
 @shared_task(
@@ -747,32 +764,30 @@ def check_for_user_file_project_sync(self: Task, *, tenant_id: str) -> None:
     return None
 
 
-@shared_task(
-    name=OnyxCeleryTask.PROCESS_SINGLE_USER_FILE_PROJECT_SYNC,
-    bind=True,
-    ignore_result=True,
-)
-def process_single_user_file_project_sync(
-    self: Task, *, user_file_id: str, tenant_id: str  # noqa: ARG001
+def project_sync_user_file_impl(
+    *, user_file_id: str, tenant_id: str, redis_locking: bool
 ) -> None:
-    """Process a single user file project sync."""
-    task_logger.info(
-        f"process_single_user_file_project_sync - Starting id={user_file_id}"
-    )
+    """Core implementation for syncing a user file's project/persona metadata.
 
-    redis_client = get_redis_client(tenant_id=tenant_id)
-    redis_client.delete(_user_file_project_sync_queued_key(user_file_id))
+    When redis_locking=True, acquires a per-file Redis lock and clears the
+    queued-key guard (Celery path).  When redis_locking=False, skips Redis
+    operations (BackgroundTask path).
+    """
+    task_logger.info(f"project_sync_user_file_impl - Starting id={user_file_id}")
 
-    file_lock: RedisLock = redis_client.lock(
-        user_file_project_sync_lock_key(user_file_id),
-        timeout=CELERY_USER_FILE_PROJECT_SYNC_LOCK_TIMEOUT,
-    )
-
-    if not file_lock.acquire(blocking=False):
-        task_logger.info(
-            f"process_single_user_file_project_sync - Lock held, skipping user_file_id={user_file_id}"
+    file_lock: RedisLock | None = None
+    if redis_locking:
+        redis_client = get_redis_client(tenant_id=tenant_id)
+        redis_client.delete(_user_file_project_sync_queued_key(user_file_id))
+        file_lock = redis_client.lock(
+            user_file_project_sync_lock_key(user_file_id),
+            timeout=CELERY_USER_FILE_PROJECT_SYNC_LOCK_TIMEOUT,
         )
-        return None
+        if file_lock is not None and not file_lock.acquire(blocking=False):
+            task_logger.info(
+                f"project_sync_user_file_impl - Lock held, skipping user_file_id={user_file_id}"
+            )
+            return
 
     try:
         with get_session_with_current_tenant() as db_session:
@@ -783,11 +798,10 @@ def process_single_user_file_project_sync(
             ).scalar_one_or_none()
             if not user_file:
                 task_logger.info(
-                    f"process_single_user_file_project_sync - User file not found id={user_file_id}"
+                    f"project_sync_user_file_impl - User file not found id={user_file_id}"
                 )
-                return None
+                return
 
-            # Sync project metadata to vector DB (skip when disabled)
             if not DISABLE_VECTOR_DB:
                 if MANAGED_VESPA:
                     httpx_init_vespa_pool(
@@ -822,7 +836,7 @@ def process_single_user_file_project_sync(
                     )
 
             task_logger.info(
-                f"process_single_user_file_project_sync - User file id={user_file_id}"
+                f"project_sync_user_file_impl - User file id={user_file_id}"
             )
 
             user_file.needs_project_sync = False
@@ -835,11 +849,22 @@ def process_single_user_file_project_sync(
 
     except Exception as e:
         task_logger.exception(
-            f"process_single_user_file_project_sync - Error syncing project for file id={user_file_id} - {e.__class__.__name__}"
+            f"project_sync_user_file_impl - Error syncing project for file id={user_file_id} - {e.__class__.__name__}"
         )
-        return None
+        raise
     finally:
-        if file_lock.owned():
+        if file_lock is not None and file_lock.owned():
             file_lock.release()
 
-    return None
+
+@shared_task(
+    name=OnyxCeleryTask.PROCESS_SINGLE_USER_FILE_PROJECT_SYNC,
+    bind=True,
+    ignore_result=True,
+)
+def process_single_user_file_project_sync(
+    self: Task, *, user_file_id: str, tenant_id: str  # noqa: ARG001
+) -> None:
+    project_sync_user_file_impl(
+        user_file_id=user_file_id, tenant_id=tenant_id, redis_locking=True
+    )

backend/onyx/background/celery/versioned_apps/background.py

@@ -1,10 +0,0 @@
-from celery import Celery
-
-from onyx.utils.variable_functionality import fetch_versioned_implementation
-from onyx.utils.variable_functionality import set_is_ee_based_on_env_variable
-
-set_is_ee_based_on_env_variable()
-app: Celery = fetch_versioned_implementation(
-    "onyx.background.celery.apps.background",
-    "celery_app",
-)

backend/onyx/background/periodic_poller.py

@@ -0,0 +1,307 @@
+"""Periodic poller for NO_VECTOR_DB deployments.
+
+Replaces Celery Beat and background workers with a lightweight daemon thread
+that runs from the API server process.  Two responsibilities:
+
+1. Recovery polling (every 30 s): re-processes user files stuck in
+   PROCESSING / DELETING / needs_sync states via the drain loops defined
+   in ``task_utils.py``.
+
+2. Periodic task execution (configurable intervals): runs LLM model updates
+   and scheduled evals at their configured cadences, with Postgres advisory
+   lock deduplication across multiple API server instances.
+"""
+
+import threading
+import time
+from collections.abc import Callable
+from dataclasses import dataclass
+from dataclasses import field
+
+from onyx.utils.logger import setup_logger
+
+logger = setup_logger()
+
+RECOVERY_INTERVAL_SECONDS = 30
+PERIODIC_TASK_LOCK_BASE = 20_000
+PERIODIC_TASK_KV_PREFIX = "periodic_poller:last_claimed:"
+
+
+# ------------------------------------------------------------------
+# Periodic task definitions
+# ------------------------------------------------------------------
+
+
+_NEVER_RAN: float = -1e18
+
+
+@dataclass
+class _PeriodicTaskDef:
+    name: str
+    interval_seconds: float
+    lock_id: int
+    run_fn: Callable[[], None]
+    last_run_at: float = field(default=_NEVER_RAN)
+
+
+def _run_auto_llm_update() -> None:
+    from onyx.configs.app_configs import AUTO_LLM_CONFIG_URL
+
+    if not AUTO_LLM_CONFIG_URL:
+        return
+
+    from onyx.db.engine.sql_engine import get_session_with_current_tenant
+    from onyx.llm.well_known_providers.auto_update_service import (
+        sync_llm_models_from_github,
+    )
+
+    with get_session_with_current_tenant() as db_session:
+        sync_llm_models_from_github(db_session)
+
+
+def _run_cache_cleanup() -> None:
+    from onyx.cache.postgres_backend import cleanup_expired_cache_entries
+
+    cleanup_expired_cache_entries()
+
+
+def _run_scheduled_eval() -> None:
+    from onyx.configs.app_configs import BRAINTRUST_API_KEY
+    from onyx.configs.app_configs import SCHEDULED_EVAL_DATASET_NAMES
+    from onyx.configs.app_configs import SCHEDULED_EVAL_PERMISSIONS_EMAIL
+    from onyx.configs.app_configs import SCHEDULED_EVAL_PROJECT
+
+    if not all(
+        [
+            BRAINTRUST_API_KEY,
+            SCHEDULED_EVAL_PROJECT,
+            SCHEDULED_EVAL_DATASET_NAMES,
+            SCHEDULED_EVAL_PERMISSIONS_EMAIL,
+        ]
+    ):
+        return
+
+    from datetime import datetime
+    from datetime import timezone
+
+    from onyx.evals.eval import run_eval
+    from onyx.evals.models import EvalConfigurationOptions
+
+    run_timestamp = datetime.now(timezone.utc).strftime("%Y-%m-%d")
+    for dataset_name in SCHEDULED_EVAL_DATASET_NAMES:
+        try:
+            run_eval(
+                configuration=EvalConfigurationOptions(
+                    search_permissions_email=SCHEDULED_EVAL_PERMISSIONS_EMAIL,
+                    dataset_name=dataset_name,
+                    no_send_logs=False,
+                    braintrust_project=SCHEDULED_EVAL_PROJECT,
+                    experiment_name=f"{dataset_name} - {run_timestamp}",
+                ),
+                remote_dataset_name=dataset_name,
+            )
+        except Exception:
+            logger.exception(
+                f"Periodic poller - Failed scheduled eval for dataset {dataset_name}"
+            )
+
+
+_CACHE_CLEANUP_INTERVAL_SECONDS = 300
+
+
+def _build_periodic_tasks() -> list[_PeriodicTaskDef]:
+    from onyx.cache.interface import CacheBackendType
+    from onyx.configs.app_configs import AUTO_LLM_CONFIG_URL
+    from onyx.configs.app_configs import AUTO_LLM_UPDATE_INTERVAL_SECONDS
+    from onyx.configs.app_configs import CACHE_BACKEND
+    from onyx.configs.app_configs import SCHEDULED_EVAL_DATASET_NAMES
+
+    tasks: list[_PeriodicTaskDef] = []
+    if CACHE_BACKEND == CacheBackendType.POSTGRES:
+        tasks.append(
+            _PeriodicTaskDef(
+                name="cache-cleanup",
+                interval_seconds=_CACHE_CLEANUP_INTERVAL_SECONDS,
+                lock_id=PERIODIC_TASK_LOCK_BASE + 2,
+                run_fn=_run_cache_cleanup,
+            )
+        )
+    if AUTO_LLM_CONFIG_URL:
+        tasks.append(
+            _PeriodicTaskDef(
+                name="auto-llm-update",
+                interval_seconds=AUTO_LLM_UPDATE_INTERVAL_SECONDS,
+                lock_id=PERIODIC_TASK_LOCK_BASE,
+                run_fn=_run_auto_llm_update,
+            )
+        )
+    if SCHEDULED_EVAL_DATASET_NAMES:
+        tasks.append(
+            _PeriodicTaskDef(
+                name="scheduled-eval",
+                interval_seconds=7 * 24 * 3600,
+                lock_id=PERIODIC_TASK_LOCK_BASE + 1,
+                run_fn=_run_scheduled_eval,
+            )
+        )
+    return tasks
+
+
+# ------------------------------------------------------------------
+# Periodic task runner with advisory-lock-guarded claim
+# ------------------------------------------------------------------
+
+
+def _try_claim_task(task_def: _PeriodicTaskDef) -> bool:
+    """Atomically check whether *task_def* should run and record a claim.
+
+    Uses a transaction-scoped advisory lock for atomicity combined with a
+    ``KVStore`` timestamp for cross-instance dedup.  The DB session is held
+    only for this brief claim transaction, not during task execution.
+    """
+    from datetime import datetime
+    from datetime import timezone
+
+    from sqlalchemy import text
+
+    from onyx.db.engine.sql_engine import get_session_with_current_tenant
+    from onyx.db.models import KVStore
+
+    kv_key = PERIODIC_TASK_KV_PREFIX + task_def.name
+
+    with get_session_with_current_tenant() as db_session:
+        acquired = db_session.execute(
+            text("SELECT pg_try_advisory_xact_lock(:id)"),
+            {"id": task_def.lock_id},
+        ).scalar()
+        if not acquired:
+            return False
+
+        row = db_session.query(KVStore).filter_by(key=kv_key).first()
+        if row and row.value is not None:
+            last_claimed = datetime.fromisoformat(str(row.value))
+            elapsed = (datetime.now(timezone.utc) - last_claimed).total_seconds()
+            if elapsed < task_def.interval_seconds:
+                return False
+
+        now_ts = datetime.now(timezone.utc).isoformat()
+        if row:
+            row.value = now_ts
+        else:
+            db_session.add(KVStore(key=kv_key, value=now_ts))
+        db_session.commit()
+
+    return True
+
+
+def _try_run_periodic_task(task_def: _PeriodicTaskDef) -> None:
+    """Run *task_def* if its interval has elapsed and no peer holds the lock."""
+    now = time.monotonic()
+    if now - task_def.last_run_at < task_def.interval_seconds:
+        return
+
+    if not _try_claim_task(task_def):
+        return
+
+    try:
+        task_def.run_fn()
+        task_def.last_run_at = now
+    except Exception:
+        logger.exception(
+            f"Periodic poller - Error running periodic task {task_def.name}"
+        )
+
+
+# ------------------------------------------------------------------
+# Recovery / drain loop runner
+# ------------------------------------------------------------------
+
+
+def _run_drain_loops(tenant_id: str) -> None:
+    from onyx.background.task_utils import drain_delete_loop
+    from onyx.background.task_utils import drain_processing_loop
+    from onyx.background.task_utils import drain_project_sync_loop
+
+    drain_processing_loop(tenant_id)
+    drain_delete_loop(tenant_id)
+    drain_project_sync_loop(tenant_id)
+
+
+# ------------------------------------------------------------------
+# Startup recovery (10g)
+# ------------------------------------------------------------------
+
+
+def recover_stuck_user_files(tenant_id: str) -> None:
+    """Run all drain loops once to re-process files left in intermediate states.
+
+    Called from ``lifespan()`` on startup when ``DISABLE_VECTOR_DB`` is set.
+    """
+    logger.info("recover_stuck_user_files - Checking for stuck user files")
+    try:
+        _run_drain_loops(tenant_id)
+    except Exception:
+        logger.exception("recover_stuck_user_files - Error during recovery")
+
+
+# ------------------------------------------------------------------
+# Daemon thread (10f)
+# ------------------------------------------------------------------
+
+_shutdown_event = threading.Event()
+_poller_thread: threading.Thread | None = None
+
+
+def _poller_loop(tenant_id: str) -> None:
+    from shared_configs.contextvars import CURRENT_TENANT_ID_CONTEXTVAR
+
+    CURRENT_TENANT_ID_CONTEXTVAR.set(tenant_id)
+
+    periodic_tasks = _build_periodic_tasks()
+    logger.info(
+        f"Periodic poller started with {len(periodic_tasks)} periodic task(s): "
+        f"{[t.name for t in periodic_tasks]}"
+    )
+
+    while not _shutdown_event.is_set():
+        try:
+            _run_drain_loops(tenant_id)
+        except Exception:
+            logger.exception("Periodic poller - Error in recovery polling")
+
+        for task_def in periodic_tasks:
+            try:
+                _try_run_periodic_task(task_def)
+            except Exception:
+                logger.exception(
+                    f"Periodic poller - Unhandled error checking task {task_def.name}"
+                )
+
+        _shutdown_event.wait(RECOVERY_INTERVAL_SECONDS)
+
+
+def start_periodic_poller(tenant_id: str) -> None:
+    """Start the periodic poller daemon thread."""
+    global _poller_thread  # noqa: PLW0603
+    _shutdown_event.clear()
+    _poller_thread = threading.Thread(
+        target=_poller_loop,
+        args=(tenant_id,),
+        daemon=True,
+        name="no-vectordb-periodic-poller",
+    )
+    _poller_thread.start()
+    logger.info("Periodic poller thread started")
+
+
+def stop_periodic_poller() -> None:
+    """Signal the periodic poller to stop and wait for it to exit."""
+    global _poller_thread  # noqa: PLW0603
+    if _poller_thread is None:
+        return
+    _shutdown_event.set()
+    _poller_thread.join(timeout=10)
+    if _poller_thread.is_alive():
+        logger.warning("Periodic poller thread did not stop within timeout")
+    _poller_thread = None
+    logger.info("Periodic poller thread stopped")

backend/onyx/background/task_utils.py

@@ -1,3 +1,33 @@
+"""Background task utilities.
+
+Contains query-history report helpers (used by all deployment modes) and
+in-process background task execution helpers for NO_VECTOR_DB mode:
+
+- Atomic claim-and-mark helpers that prevent duplicate processing
+- Drain loops that process all pending user file work
+
+Each claim function runs a short-lived transaction: SELECT ... FOR UPDATE
+SKIP LOCKED, UPDATE the row to remove it from future queries, COMMIT.
+After the commit the row lock is released, but the row is no longer
+eligible for re-claiming.  No long-lived sessions or advisory locks.
+"""
+
+from uuid import UUID
+
+import sqlalchemy as sa
+from sqlalchemy import select
+from sqlalchemy.orm import Session
+
+from onyx.db.enums import UserFileStatus
+from onyx.db.models import UserFile
+from onyx.utils.logger import setup_logger
+
+logger = setup_logger()
+
+# ------------------------------------------------------------------
+# Query-history report helpers (pre-existing, used by all modes)
+# ------------------------------------------------------------------
+
 QUERY_REPORT_NAME_PREFIX = "query-history"
 
 
@@ -9,3 +39,168 @@ def construct_query_history_report_name(
 
 def extract_task_id_from_query_history_report_name(name: str) -> str:
     return name.removeprefix(f"{QUERY_REPORT_NAME_PREFIX}-").removesuffix(".csv")
+
+
+# ------------------------------------------------------------------
+# Atomic claim-and-mark helpers
+# ------------------------------------------------------------------
+# Each function runs inside a single short-lived session/transaction:
+#   1. SELECT ... FOR UPDATE SKIP LOCKED  (locks one eligible row)
+#   2. UPDATE the row so it is no longer eligible
+#   3. COMMIT  (releases the row lock)
+# After the commit, no other drain loop can claim the same row.
+
+
+def _claim_next_processing_file(db_session: Session) -> UUID | None:
+    """Claim the next PROCESSING file by transitioning it to INDEXING.
+
+    Returns the file id, or None when no eligible files remain.
+    """
+    file_id = db_session.execute(
+        select(UserFile.id)
+        .where(UserFile.status == UserFileStatus.PROCESSING)
+        .order_by(UserFile.created_at)
+        .limit(1)
+        .with_for_update(skip_locked=True)
+    ).scalar_one_or_none()
+    if file_id is None:
+        return None
+
+    db_session.execute(
+        sa.update(UserFile)
+        .where(UserFile.id == file_id)
+        .values(status=UserFileStatus.INDEXING)
+    )
+    db_session.commit()
+    return file_id
+
+
+def _claim_next_deleting_file(
+    db_session: Session,
+    exclude_ids: set[UUID] | None = None,
+) -> UUID | None:
+    """Claim the next DELETING file.
+
+    No status transition needed — the impl deletes the row on success.
+    The short-lived FOR UPDATE lock prevents concurrent claims.
+    *exclude_ids* prevents re-processing the same file if the impl fails.
+    """
+    stmt = (
+        select(UserFile.id)
+        .where(UserFile.status == UserFileStatus.DELETING)
+        .order_by(UserFile.created_at)
+        .limit(1)
+        .with_for_update(skip_locked=True)
+    )
+    if exclude_ids:
+        stmt = stmt.where(UserFile.id.notin_(exclude_ids))
+    file_id = db_session.execute(stmt).scalar_one_or_none()
+    db_session.commit()
+    return file_id
+
+
+def _claim_next_sync_file(
+    db_session: Session,
+    exclude_ids: set[UUID] | None = None,
+) -> UUID | None:
+    """Claim the next file needing project/persona sync.
+
+    No status transition needed — the impl clears the sync flags on
+    success.  The short-lived FOR UPDATE lock prevents concurrent claims.
+    *exclude_ids* prevents re-processing the same file if the impl fails.
+    """
+    stmt = (
+        select(UserFile.id)
+        .where(
+            sa.and_(
+                sa.or_(
+                    UserFile.needs_project_sync.is_(True),
+                    UserFile.needs_persona_sync.is_(True),
+                ),
+                UserFile.status == UserFileStatus.COMPLETED,
+            )
+        )
+        .order_by(UserFile.created_at)
+        .limit(1)
+        .with_for_update(skip_locked=True)
+    )
+    if exclude_ids:
+        stmt = stmt.where(UserFile.id.notin_(exclude_ids))
+    file_id = db_session.execute(stmt).scalar_one_or_none()
+    db_session.commit()
+    return file_id
+
+
+# ------------------------------------------------------------------
+# Drain loops — process *all* pending work of each type
+# ------------------------------------------------------------------
+
+
+def drain_processing_loop(tenant_id: str) -> None:
+    """Process all pending PROCESSING user files."""
+    from onyx.background.celery.tasks.user_file_processing.tasks import (
+        process_user_file_impl,
+    )
+    from onyx.db.engine.sql_engine import get_session_with_current_tenant
+
+    while True:
+        with get_session_with_current_tenant() as session:
+            file_id = _claim_next_processing_file(session)
+        if file_id is None:
+            break
+        try:
+            process_user_file_impl(
+                user_file_id=str(file_id),
+                tenant_id=tenant_id,
+                redis_locking=False,
+            )
+        except Exception:
+            logger.exception(f"Failed to process user file {file_id}")
+
+
+def drain_delete_loop(tenant_id: str) -> None:
+    """Delete all pending DELETING user files."""
+    from onyx.background.celery.tasks.user_file_processing.tasks import (
+        delete_user_file_impl,
+    )
+    from onyx.db.engine.sql_engine import get_session_with_current_tenant
+
+    failed: set[UUID] = set()
+    while True:
+        with get_session_with_current_tenant() as session:
+            file_id = _claim_next_deleting_file(session, exclude_ids=failed)
+        if file_id is None:
+            break
+        try:
+            delete_user_file_impl(
+                user_file_id=str(file_id),
+                tenant_id=tenant_id,
+                redis_locking=False,
+            )
+        except Exception:
+            logger.exception(f"Failed to delete user file {file_id}")
+            failed.add(file_id)
+
+
+def drain_project_sync_loop(tenant_id: str) -> None:
+    """Sync all pending project/persona metadata for user files."""
+    from onyx.background.celery.tasks.user_file_processing.tasks import (
+        project_sync_user_file_impl,
+    )
+    from onyx.db.engine.sql_engine import get_session_with_current_tenant
+
+    failed: set[UUID] = set()
+    while True:
+        with get_session_with_current_tenant() as session:
+            file_id = _claim_next_sync_file(session, exclude_ids=failed)
+        if file_id is None:
+            break
+        try:
+            project_sync_user_file_impl(
+                user_file_id=str(file_id),
+                tenant_id=tenant_id,
+                redis_locking=False,
+            )
+        except Exception:
+            logger.exception(f"Failed to sync user file {file_id}")
+            failed.add(file_id)

backend/onyx/cache/factory.py

@@ -0,0 +1,51 @@
+from collections.abc import Callable
+
+from onyx.cache.interface import CacheBackend
+from onyx.cache.interface import CacheBackendType
+from onyx.configs.app_configs import CACHE_BACKEND
+
+
+def _build_redis_backend(tenant_id: str) -> CacheBackend:
+    from onyx.cache.redis_backend import RedisCacheBackend
+    from onyx.redis.redis_pool import redis_pool
+
+    return RedisCacheBackend(redis_pool.get_client(tenant_id))
+
+
+def _build_postgres_backend(tenant_id: str) -> CacheBackend:
+    from onyx.cache.postgres_backend import PostgresCacheBackend
+
+    return PostgresCacheBackend(tenant_id)
+
+
+_BACKEND_BUILDERS: dict[CacheBackendType, Callable[[str], CacheBackend]] = {
+    CacheBackendType.REDIS: _build_redis_backend,
+    CacheBackendType.POSTGRES: _build_postgres_backend,
+}
+
+
+def get_cache_backend(*, tenant_id: str | None = None) -> CacheBackend:
+    """Return a tenant-aware ``CacheBackend``.
+
+    If *tenant_id* is ``None``, the current tenant is read from the
+    thread-local context variable (same behaviour as ``get_redis_client``).
+    """
+    if tenant_id is None:
+        from shared_configs.contextvars import get_current_tenant_id
+
+        tenant_id = get_current_tenant_id()
+
+    builder = _BACKEND_BUILDERS.get(CACHE_BACKEND)
+    if builder is None:
+        raise ValueError(
+            f"Unsupported CACHE_BACKEND={CACHE_BACKEND!r}. "
+            f"Supported values: {[t.value for t in CacheBackendType]}"
+        )
+    return builder(tenant_id)
+
+
+def get_shared_cache_backend() -> CacheBackend:
+    """Return a ``CacheBackend`` in the shared (cross-tenant) namespace."""
+    from shared_configs.configs import DEFAULT_REDIS_PREFIX
+
+    return get_cache_backend(tenant_id=DEFAULT_REDIS_PREFIX)

backend/onyx/cache/interface.py

@@ -0,0 +1,115 @@
+import abc
+from enum import Enum
+
+from redis.exceptions import RedisError
+from sqlalchemy.exc import SQLAlchemyError
+
+TTL_KEY_NOT_FOUND = -2
+TTL_NO_EXPIRY = -1
+
+CACHE_TRANSIENT_ERRORS: tuple[type[Exception], ...] = (RedisError, SQLAlchemyError)
+"""Exception types that represent transient cache connectivity / operational
+failures.  Callers that want to fail-open (or fail-closed) on cache errors
+should catch this tuple instead of bare ``Exception``.
+
+When adding a new ``CacheBackend`` implementation, add its transient error
+base class(es) here so all call-sites pick it up automatically."""
+
+
+class CacheBackendType(str, Enum):
+    REDIS = "redis"
+    POSTGRES = "postgres"
+
+
+class CacheLock(abc.ABC):
+    """Abstract distributed lock returned by CacheBackend.lock()."""
+
+    @abc.abstractmethod
+    def acquire(
+        self,
+        blocking: bool = True,
+        blocking_timeout: float | None = None,
+    ) -> bool:
+        raise NotImplementedError
+
+    @abc.abstractmethod
+    def release(self) -> None:
+        raise NotImplementedError
+
+    @abc.abstractmethod
+    def owned(self) -> bool:
+        raise NotImplementedError
+
+    def __enter__(self) -> "CacheLock":
+        if not self.acquire():
+            raise RuntimeError("Failed to acquire lock")
+        return self
+
+    def __exit__(self, *args: object) -> None:
+        self.release()
+
+
+class CacheBackend(abc.ABC):
+    """Thin abstraction over a key-value cache with TTL, locks, and blocking lists.
+
+    Covers the subset of Redis operations used outside of Celery. When
+    CACHE_BACKEND=postgres, a PostgreSQL-backed implementation is used instead.
+    """
+
+    # -- basic key/value ---------------------------------------------------
+
+    @abc.abstractmethod
+    def get(self, key: str) -> bytes | None:
+        raise NotImplementedError
+
+    @abc.abstractmethod
+    def set(
+        self,
+        key: str,
+        value: str | bytes | int | float,
+        ex: int | None = None,
+    ) -> None:
+        raise NotImplementedError
+
+    @abc.abstractmethod
+    def delete(self, key: str) -> None:
+        raise NotImplementedError
+
+    @abc.abstractmethod
+    def exists(self, key: str) -> bool:
+        raise NotImplementedError
+
+    # -- TTL ---------------------------------------------------------------
+
+    @abc.abstractmethod
+    def expire(self, key: str, seconds: int) -> None:
+        raise NotImplementedError
+
+    @abc.abstractmethod
+    def ttl(self, key: str) -> int:
+        """Return remaining TTL in seconds.
+
+        Returns ``TTL_NO_EXPIRY`` (-1) if key exists without expiry,
+        ``TTL_KEY_NOT_FOUND`` (-2) if key is missing or expired.
+        """
+        raise NotImplementedError
+
+    # -- distributed lock --------------------------------------------------
+
+    @abc.abstractmethod
+    def lock(self, name: str, timeout: float | None = None) -> CacheLock:
+        raise NotImplementedError
+
+    # -- blocking list (used by MCP OAuth BLPOP pattern) -------------------
+
+    @abc.abstractmethod
+    def rpush(self, key: str, value: str | bytes) -> None:
+        raise NotImplementedError
+
+    @abc.abstractmethod
+    def blpop(self, keys: list[str], timeout: int = 0) -> tuple[bytes, bytes] | None:
+        """Block until a value is available on one of *keys*, or *timeout* expires.
+
+        Returns ``(key, value)`` or ``None`` on timeout.
+        """
+        raise NotImplementedError

backend/onyx/cache/postgres_backend.py

@@ -0,0 +1,323 @@
+"""PostgreSQL-backed ``CacheBackend`` for NO_VECTOR_DB deployments.
+
+Uses the ``cache_store`` table for key-value storage, PostgreSQL advisory locks
+for distributed locking, and a polling loop for the BLPOP pattern.
+"""
+
+import hashlib
+import struct
+import time
+import uuid
+from contextlib import AbstractContextManager
+from datetime import datetime
+from datetime import timedelta
+from datetime import timezone
+
+from sqlalchemy import delete
+from sqlalchemy import func
+from sqlalchemy import or_
+from sqlalchemy import select
+from sqlalchemy import update
+from sqlalchemy.dialects.postgresql import insert as pg_insert
+from sqlalchemy.orm import Session
+
+from onyx.cache.interface import CacheBackend
+from onyx.cache.interface import CacheLock
+from onyx.cache.interface import TTL_KEY_NOT_FOUND
+from onyx.cache.interface import TTL_NO_EXPIRY
+from onyx.db.models import CacheStore
+
+_LIST_KEY_PREFIX = "_q:"
+# ASCII: ':' (0x3A) < ';' (0x3B). Upper bound for range queries so [prefix+, prefix;)
+# captures all list-item keys (e.g. _q:mylist:123:uuid) without including other
+# lists whose names share a prefix (e.g. _q:mylist2:...).
+_LIST_KEY_RANGE_TERMINATOR = ";"
+_LIST_ITEM_TTL_SECONDS = 3600
+_LOCK_POLL_INTERVAL = 0.1
+_BLPOP_POLL_INTERVAL = 0.25
+
+
+def _list_item_key(key: str) -> str:
+    """Unique key for a list item. Timestamp for FIFO ordering; UUID prevents
+    collision when concurrent rpush calls occur within the same nanosecond.
+    """
+    return f"{_LIST_KEY_PREFIX}{key}:{time.time_ns()}:{uuid.uuid4().hex}"
+
+
+def _to_bytes(value: str | bytes | int | float) -> bytes:
+    if isinstance(value, bytes):
+        return value
+    return str(value).encode()
+
+
+# ------------------------------------------------------------------
+# Lock
+# ------------------------------------------------------------------
+
+
+class PostgresCacheLock(CacheLock):
+    """Advisory-lock-based distributed lock.
+
+    Uses ``get_session_with_tenant`` for connection lifecycle.  The lock is tied
+    to the session's connection; releasing or closing the session frees it.
+
+    NOTE: Unlike Redis locks, advisory locks do not auto-expire after
+    ``timeout`` seconds.  They are released when ``release()`` is
+    called or when the session is closed.
+    """
+
+    def __init__(self, lock_id: int, timeout: float | None, tenant_id: str) -> None:
+        self._lock_id = lock_id
+        self._timeout = timeout
+        self._tenant_id = tenant_id
+        self._session_cm: AbstractContextManager[Session] | None = None
+        self._session: Session | None = None
+        self._acquired = False
+
+    def acquire(
+        self,
+        blocking: bool = True,
+        blocking_timeout: float | None = None,
+    ) -> bool:
+        from onyx.db.engine.sql_engine import get_session_with_tenant
+
+        self._session_cm = get_session_with_tenant(tenant_id=self._tenant_id)
+        self._session = self._session_cm.__enter__()
+        try:
+            if not blocking:
+                return self._try_lock()
+
+            effective_timeout = blocking_timeout or self._timeout
+            deadline = (
+                (time.monotonic() + effective_timeout) if effective_timeout else None
+            )
+            while True:
+                if self._try_lock():
+                    return True
+                if deadline is not None and time.monotonic() >= deadline:
+                    return False
+                time.sleep(_LOCK_POLL_INTERVAL)
+        finally:
+            if not self._acquired:
+                self._close_session()
+
+    def release(self) -> None:
+        if not self._acquired or self._session is None:
+            return
+        try:
+            self._session.execute(select(func.pg_advisory_unlock(self._lock_id)))
+        finally:
+            self._acquired = False
+            self._close_session()
+
+    def owned(self) -> bool:
+        return self._acquired
+
+    def _close_session(self) -> None:
+        if self._session_cm is not None:
+            try:
+                self._session_cm.__exit__(None, None, None)
+            finally:
+                self._session_cm = None
+                self._session = None
+
+    def _try_lock(self) -> bool:
+        assert self._session is not None
+        result = self._session.execute(
+            select(func.pg_try_advisory_lock(self._lock_id))
+        ).scalar()
+        if result:
+            self._acquired = True
+            return True
+        return False
+
+
+# ------------------------------------------------------------------
+# Backend
+# ------------------------------------------------------------------
+
+
+class PostgresCacheBackend(CacheBackend):
+    """``CacheBackend`` backed by the ``cache_store`` table in PostgreSQL.
+
+    Each operation opens and closes its own database session so the backend
+    is safe to share across threads.  Tenant isolation is handled by
+    SQLAlchemy's ``schema_translate_map`` (set by ``get_session_with_tenant``).
+    """
+
+    def __init__(self, tenant_id: str) -> None:
+        self._tenant_id = tenant_id
+
+    # -- basic key/value ---------------------------------------------------
+
+    def get(self, key: str) -> bytes | None:
+        from onyx.db.engine.sql_engine import get_session_with_tenant
+
+        stmt = select(CacheStore.value).where(
+            CacheStore.key == key,
+            or_(CacheStore.expires_at.is_(None), CacheStore.expires_at > func.now()),
+        )
+        with get_session_with_tenant(tenant_id=self._tenant_id) as session:
+            value = session.execute(stmt).scalar_one_or_none()
+        if value is None:
+            return None
+        return bytes(value)
+
+    def set(
+        self,
+        key: str,
+        value: str | bytes | int | float,
+        ex: int | None = None,
+    ) -> None:
+        from onyx.db.engine.sql_engine import get_session_with_tenant
+
+        value_bytes = _to_bytes(value)
+        expires_at = (
+            datetime.now(timezone.utc) + timedelta(seconds=ex)
+            if ex is not None
+            else None
+        )
+        stmt = (
+            pg_insert(CacheStore)
+            .values(key=key, value=value_bytes, expires_at=expires_at)
+            .on_conflict_do_update(
+                index_elements=[CacheStore.key],
+                set_={"value": value_bytes, "expires_at": expires_at},
+            )
+        )
+        with get_session_with_tenant(tenant_id=self._tenant_id) as session:
+            session.execute(stmt)
+            session.commit()
+
+    def delete(self, key: str) -> None:
+        from onyx.db.engine.sql_engine import get_session_with_tenant
+
+        with get_session_with_tenant(tenant_id=self._tenant_id) as session:
+            session.execute(delete(CacheStore).where(CacheStore.key == key))
+            session.commit()
+
+    def exists(self, key: str) -> bool:
+        from onyx.db.engine.sql_engine import get_session_with_tenant
+
+        stmt = (
+            select(CacheStore.key)
+            .where(
+                CacheStore.key == key,
+                or_(
+                    CacheStore.expires_at.is_(None),
+                    CacheStore.expires_at > func.now(),
+                ),
+            )
+            .limit(1)
+        )
+        with get_session_with_tenant(tenant_id=self._tenant_id) as session:
+            return session.execute(stmt).first() is not None
+
+    # -- TTL ---------------------------------------------------------------
+
+    def expire(self, key: str, seconds: int) -> None:
+        from onyx.db.engine.sql_engine import get_session_with_tenant
+
+        new_exp = datetime.now(timezone.utc) + timedelta(seconds=seconds)
+        stmt = (
+            update(CacheStore).where(CacheStore.key == key).values(expires_at=new_exp)
+        )
+        with get_session_with_tenant(tenant_id=self._tenant_id) as session:
+            session.execute(stmt)
+            session.commit()
+
+    def ttl(self, key: str) -> int:
+        from onyx.db.engine.sql_engine import get_session_with_tenant
+
+        stmt = select(CacheStore.expires_at).where(CacheStore.key == key)
+        with get_session_with_tenant(tenant_id=self._tenant_id) as session:
+            result = session.execute(stmt).first()
+        if result is None:
+            return TTL_KEY_NOT_FOUND
+        expires_at: datetime | None = result[0]
+        if expires_at is None:
+            return TTL_NO_EXPIRY
+        remaining = (expires_at - datetime.now(timezone.utc)).total_seconds()
+        if remaining <= 0:
+            return TTL_KEY_NOT_FOUND
+        return int(remaining)
+
+    # -- distributed lock --------------------------------------------------
+
+    def lock(self, name: str, timeout: float | None = None) -> CacheLock:
+        return PostgresCacheLock(
+            self._lock_id_for(name), timeout, tenant_id=self._tenant_id
+        )
+
+    # -- blocking list (MCP OAuth BLPOP pattern) ---------------------------
+
+    def rpush(self, key: str, value: str | bytes) -> None:
+        self.set(_list_item_key(key), value, ex=_LIST_ITEM_TTL_SECONDS)
+
+    def blpop(self, keys: list[str], timeout: int = 0) -> tuple[bytes, bytes] | None:
+        if timeout <= 0:
+            raise ValueError(
+                "PostgresCacheBackend.blpop requires timeout > 0. "
+                "timeout=0 would block the calling thread indefinitely "
+                "with no way to interrupt short of process termination."
+            )
+        from onyx.db.engine.sql_engine import get_session_with_tenant
+
+        deadline = time.monotonic() + timeout
+        while True:
+            for key in keys:
+                lower = f"{_LIST_KEY_PREFIX}{key}:"
+                upper = f"{_LIST_KEY_PREFIX}{key}{_LIST_KEY_RANGE_TERMINATOR}"
+                stmt = (
+                    select(CacheStore)
+                    .where(
+                        CacheStore.key >= lower,
+                        CacheStore.key < upper,
+                        or_(
+                            CacheStore.expires_at.is_(None),
+                            CacheStore.expires_at > func.now(),
+                        ),
+                    )
+                    .order_by(CacheStore.key)
+                    .limit(1)
+                    .with_for_update(skip_locked=True)
+                )
+                with get_session_with_tenant(tenant_id=self._tenant_id) as session:
+                    row = session.execute(stmt).scalars().first()
+                    if row is not None:
+                        value = bytes(row.value) if row.value else b""
+                        session.delete(row)
+                        session.commit()
+                        return (key.encode(), value)
+            if time.monotonic() >= deadline:
+                return None
+            time.sleep(_BLPOP_POLL_INTERVAL)
+
+    # -- helpers -----------------------------------------------------------
+
+    def _lock_id_for(self, name: str) -> int:
+        """Map *name* to a 64-bit signed int for ``pg_advisory_lock``."""
+        h = hashlib.md5(f"{self._tenant_id}:{name}".encode()).digest()
+        return struct.unpack("q", h[:8])[0]
+
+
+# ------------------------------------------------------------------
+# Periodic cleanup
+# ------------------------------------------------------------------
+
+
+def cleanup_expired_cache_entries() -> None:
+    """Delete rows whose ``expires_at`` is in the past.
+
+    Called by the periodic poller every 5 minutes.
+    """
+    from onyx.db.engine.sql_engine import get_session_with_current_tenant
+
+    with get_session_with_current_tenant() as session:
+        session.execute(
+            delete(CacheStore).where(
+                CacheStore.expires_at.is_not(None),
+                CacheStore.expires_at < func.now(),
+            )
+        )
+        session.commit()

backend/onyx/cache/redis_backend.py

@@ -0,0 +1,92 @@
+from typing import cast
+
+from redis.client import Redis
+from redis.lock import Lock as RedisLock
+
+from onyx.cache.interface import CacheBackend
+from onyx.cache.interface import CacheLock
+
+
+class RedisCacheLock(CacheLock):
+    """Wraps ``redis.lock.Lock`` behind the ``CacheLock`` interface."""
+
+    def __init__(self, lock: RedisLock) -> None:
+        self._lock = lock
+
+    def acquire(
+        self,
+        blocking: bool = True,
+        blocking_timeout: float | None = None,
+    ) -> bool:
+        return bool(
+            self._lock.acquire(
+                blocking=blocking,
+                blocking_timeout=blocking_timeout,
+            )
+        )
+
+    def release(self) -> None:
+        self._lock.release()
+
+    def owned(self) -> bool:
+        return bool(self._lock.owned())
+
+
+class RedisCacheBackend(CacheBackend):
+    """``CacheBackend`` implementation that delegates to a ``redis.Redis`` client.
+
+    This is a thin pass-through — every method maps 1-to-1 to the underlying
+    Redis command.  ``TenantRedis`` key-prefixing is handled by the client
+    itself (provided by ``get_redis_client``).
+    """
+
+    def __init__(self, redis_client: Redis) -> None:
+        self._r = redis_client
+
+    # -- basic key/value ---------------------------------------------------
+
+    def get(self, key: str) -> bytes | None:
+        val = self._r.get(key)
+        if val is None:
+            return None
+        if isinstance(val, bytes):
+            return val
+        return str(val).encode()
+
+    def set(
+        self,
+        key: str,
+        value: str | bytes | int | float,
+        ex: int | None = None,
+    ) -> None:
+        self._r.set(key, value, ex=ex)
+
+    def delete(self, key: str) -> None:
+        self._r.delete(key)
+
+    def exists(self, key: str) -> bool:
+        return bool(self._r.exists(key))
+
+    # -- TTL ---------------------------------------------------------------
+
+    def expire(self, key: str, seconds: int) -> None:
+        self._r.expire(key, seconds)
+
+    def ttl(self, key: str) -> int:
+        return cast(int, self._r.ttl(key))
+
+    # -- distributed lock --------------------------------------------------
+
+    def lock(self, name: str, timeout: float | None = None) -> CacheLock:
+        return RedisCacheLock(self._r.lock(name, timeout=timeout))
+
+    # -- blocking list (MCP OAuth BLPOP pattern) ---------------------------
+
+    def rpush(self, key: str, value: str | bytes) -> None:
+        self._r.rpush(key, value)
+
+    def blpop(self, keys: list[str], timeout: int = 0) -> tuple[bytes, bytes] | None:
+        result = cast(list[bytes] | None, self._r.blpop(keys, timeout=timeout))
+        if result is None:
+            return None
+        return (result[0], result[1])

backend/onyx/chat/chat_processing_checker.py

@@ -1,57 +1,52 @@
 from uuid import UUID
 
-from redis.client import Redis
+from onyx.cache.interface import CacheBackend
 
-# Redis key prefixes for chat message processing
 PREFIX = "chatprocessing"
 FENCE_PREFIX = f"{PREFIX}_fence"
 FENCE_TTL = 30 * 60  # 30 minutes
 
 
 def _get_fence_key(chat_session_id: UUID) -> str:
-    """
-    Generate the Redis key for a chat session processing a message.
+    """Generate the cache key for a chat session processing fence.
 
     Args:
         chat_session_id: The UUID of the chat session
 
     Returns:
-        The fence key string (tenant_id is automatically added by the Redis client)
+        The fence key string. Tenant isolation is handled automatically
+        by the cache backend (Redis key-prefixing or Postgres schema routing).
     """
     return f"{FENCE_PREFIX}_{chat_session_id}"
 
 
 def set_processing_status(
-    chat_session_id: UUID, redis_client: Redis, value: bool
+    chat_session_id: UUID, cache: CacheBackend, value: bool
 ) -> None:
-    """
-    Set or clear the fence for a chat session processing a message.
+    """Set or clear the fence for a chat session processing a message.
 
-    If the key exists, we are processing a message. If the key does not exist, we are not processing a message.
+    If the key exists, a message is being processed.
 
     Args:
         chat_session_id: The UUID of the chat session
-        redis_client: The Redis client to use
+        cache: Tenant-aware cache backend
         value: True to set the fence, False to clear it
     """
     fence_key = _get_fence_key(chat_session_id)
-
     if value:
-        redis_client.set(fence_key, 0, ex=FENCE_TTL)
+        cache.set(fence_key, 0, ex=FENCE_TTL)
     else:
-        redis_client.delete(fence_key)
+        cache.delete(fence_key)
 
 
-def is_chat_session_processing(chat_session_id: UUID, redis_client: Redis) -> bool:
-    """
-    Check if the chat session is processing a message.
+def is_chat_session_processing(chat_session_id: UUID, cache: CacheBackend) -> bool:
+    """Check if the chat session is processing a message.
 
     Args:
         chat_session_id: The UUID of the chat session
-        redis_client: The Redis client to use
+        cache: Tenant-aware cache backend
 
     Returns:
         True if the chat session is processing a message, False otherwise
     """
-    fence_key = _get_fence_key(chat_session_id)
-    return bool(redis_client.exists(fence_key))
+    return cache.exists(_get_fence_key(chat_session_id))

backend/onyx/chat/llm_loop.py

@@ -1,6 +1,7 @@
 import json
 import time
 from collections.abc import Callable
+from typing import Any
 from typing import Literal
 
 from sqlalchemy.orm import Session
@@ -51,6 +52,7 @@ from onyx.tools.built_in_tools import STOPPING_TOOLS_NAMES
 from onyx.tools.interface import Tool
 from onyx.tools.models import ChatFile
 from onyx.tools.models import MemoryToolResponseSnapshot
+from onyx.tools.models import PythonToolRichResponse
 from onyx.tools.models import ToolCallInfo
 from onyx.tools.models import ToolCallKickoff
 from onyx.tools.models import ToolResponse
@@ -530,11 +532,13 @@ def _create_file_tool_metadata_message(
     """
     lines = [
         "You have access to the following files. Use the read_file tool to "
-        "read sections of any file:"
+        "read sections of any file. You MUST pass the file_id UUID (not the "
+        "filename) to read_file:"
     ]
     for meta in file_metadata:
         lines.append(
-            f'- {meta.file_id}: "{meta.filename}" (~{meta.approx_char_count:,} chars)'
+            f'- file_id="{meta.file_id}" filename="{meta.filename}" '
+            f"(~{meta.approx_char_count:,} chars)"
         )
 
     message_content = "\n".join(lines)
@@ -558,12 +562,16 @@ def _create_context_files_message(
     # Format as documents JSON as described in README
     documents_list = []
     for idx, file_text in enumerate(context_files.file_texts, start=1):
-        documents_list.append(
-            {
-                "document": idx,
-                "contents": file_text,
-            }
+        title = (
+            context_files.file_metadata[idx - 1].filename
+            if idx - 1 < len(context_files.file_metadata)
+            else None
         )
+        entry: dict[str, Any] = {"document": idx}
+        if title:
+            entry["title"] = title
+        entry["contents"] = file_text
+        documents_list.append(entry)
 
     documents_json = json.dumps({"documents": documents_list}, indent=2)
     message_content = f"Here are some documents provided for context, they may not all be relevant:\n{documents_json}"
@@ -959,6 +967,13 @@ def run_llm_loop(
                 ):
                     generated_images = tool_response.rich_response.generated_images
 
+                # Extract generated_files if this is a code interpreter response
+                generated_files = None
+                if isinstance(tool_response.rich_response, PythonToolRichResponse):
+                    generated_files = (
+                        tool_response.rich_response.generated_files or None
+                    )
+
                 # Persist memory if this is a memory tool response
                 memory_snapshot: MemoryToolResponseSnapshot | None = None
                 if isinstance(tool_response.rich_response, MemoryToolResponse):
@@ -1010,6 +1025,7 @@ def run_llm_loop(
                     tool_call_response=saved_response,
                     search_docs=displayed_docs or search_docs,
                     generated_images=generated_images,
+                    generated_files=generated_files,
                 )
                 # Add to state container for partial save support
                 state_container.add_tool_call(tool_call_info)

backend/onyx/chat/process_message.py

@@ -11,9 +11,10 @@ from contextvars import Token
 from uuid import UUID
 
 from pydantic import BaseModel
-from redis.client import Redis
 from sqlalchemy.orm import Session
 
+from onyx.cache.factory import get_cache_backend
+from onyx.cache.interface import CacheBackend
 from onyx.chat.chat_processing_checker import set_processing_status
 from onyx.chat.chat_state import ChatStateContainer
 from onyx.chat.chat_state import run_chat_loop_with_state_containers
@@ -79,7 +80,6 @@ from onyx.llm.request_context import reset_llm_mock_response
 from onyx.llm.request_context import set_llm_mock_response
 from onyx.llm.utils import litellm_exception_to_error_msg
 from onyx.onyxbot.slack.models import SlackContext
-from onyx.redis.redis_pool import get_redis_client
 from onyx.server.query_and_chat.models import AUTO_PLACE_AFTER_LATEST_MESSAGE
 from onyx.server.query_and_chat.models import MessageResponseIDInfo
 from onyx.server.query_and_chat.models import SendMessageRequest
@@ -448,7 +448,7 @@ def handle_stream_message_objects(
 
     llm: LLM | None = None
     chat_session: ChatSession | None = None
-    redis_client: Redis | None = None
+    cache: CacheBackend | None = None
 
     user_id = user.id
     if user.is_anonymous:
@@ -809,19 +809,19 @@ def handle_stream_message_objects(
             )
             simple_chat_history.insert(0, summary_simple)
 
-        redis_client = get_redis_client()
+        cache = get_cache_backend()
 
         reset_cancel_status(
             chat_session.id,
-            redis_client,
+            cache,
         )
 
         def check_is_connected() -> bool:
-            return check_stop_signal(chat_session.id, redis_client)
+            return check_stop_signal(chat_session.id, cache)
 
         set_processing_status(
             chat_session_id=chat_session.id,
-            redis_client=redis_client,
+            cache=cache,
             value=True,
         )
 
@@ -968,10 +968,10 @@ def handle_stream_message_objects(
             reset_llm_mock_response(mock_response_token)
 
         try:
-            if redis_client is not None and chat_session is not None:
+            if cache is not None and chat_session is not None:
                 set_processing_status(
                     chat_session_id=chat_session.id,
-                    redis_client=redis_client,
+                    cache=cache,
                     value=False,
                 )
         except Exception:

backend/onyx/chat/save_chat.py

@@ -1,4 +1,5 @@
 import json
+import mimetypes
 
 from sqlalchemy.orm import Session
 
@@ -12,14 +13,41 @@ from onyx.db.chat import create_db_search_doc
 from onyx.db.models import ChatMessage
 from onyx.db.models import ToolCall
 from onyx.db.tools import create_tool_call_no_commit
+from onyx.file_store.models import FileDescriptor
 from onyx.natural_language_processing.utils import BaseTokenizer
 from onyx.natural_language_processing.utils import get_tokenizer
+from onyx.server.query_and_chat.chat_utils import mime_type_to_chat_file_type
 from onyx.tools.models import ToolCallInfo
 from onyx.utils.logger import setup_logger
 
 logger = setup_logger()
 
 
+def _extract_referenced_file_descriptors(
+    tool_calls: list[ToolCallInfo],
+    message_text: str,
+) -> list[FileDescriptor]:
+    """Extract FileDescriptors for code interpreter files referenced in the message text."""
+    descriptors: list[FileDescriptor] = []
+    for tool_call_info in tool_calls:
+        if not tool_call_info.generated_files:
+            continue
+        for gen_file in tool_call_info.generated_files:
+            file_id = (
+                gen_file.file_link.rsplit("/", 1)[-1] if gen_file.file_link else ""
+            )
+            if file_id and file_id in message_text:
+                mime_type, _ = mimetypes.guess_type(gen_file.filename)
+                descriptors.append(
+                    FileDescriptor(
+                        id=file_id,
+                        type=mime_type_to_chat_file_type(mime_type),
+                        name=gen_file.filename,
+                    )
+                )
+    return descriptors
+
+
 def _create_and_link_tool_calls(
     tool_calls: list[ToolCallInfo],
     assistant_message: ChatMessage,
@@ -297,5 +325,14 @@ def save_chat_turn(
         citation_number_to_search_doc_id if citation_number_to_search_doc_id else None
     )
 
+    # 8. Attach code interpreter generated files that the assistant actually
+    # referenced in its response, so they are available via load_all_chat_files
+    # on subsequent turns. Files not mentioned are intermediate artifacts.
+    if message_text:
+        referenced = _extract_referenced_file_descriptors(tool_calls, message_text)
+        if referenced:
+            existing_files = assistant_message.files or []
+            assistant_message.files = existing_files + referenced
+
     # Finally save the messages, tool calls, and docs
     db_session.commit()

backend/onyx/chat/stop_signal_checker.py

@@ -1,65 +1,58 @@
 from uuid import UUID
 
-from redis.client import Redis
+from onyx.cache.interface import CacheBackend
 
-# Redis key prefixes for chat session stop signals
 PREFIX = "chatsessionstop"
 FENCE_PREFIX = f"{PREFIX}_fence"
-FENCE_TTL = 10 * 60  # 10 minutes - defensive TTL to prevent memory leaks
+FENCE_TTL = 10 * 60  # 10 minutes
 
 
 def _get_fence_key(chat_session_id: UUID) -> str:
-    """
-    Generate the Redis key for a chat session stop signal fence.
+    """Generate the cache key for a chat session stop signal fence.
 
     Args:
         chat_session_id: The UUID of the chat session
 
     Returns:
-        The fence key string (tenant_id is automatically added by the Redis client)
+        The fence key string. Tenant isolation is handled automatically
+        by the cache backend (Redis key-prefixing or Postgres schema routing).
     """
     return f"{FENCE_PREFIX}_{chat_session_id}"
 
 
-def set_fence(chat_session_id: UUID, redis_client: Redis, value: bool) -> None:
-    """
-    Set or clear the stop signal fence for a chat session.
+def set_fence(chat_session_id: UUID, cache: CacheBackend, value: bool) -> None:
+    """Set or clear the stop signal fence for a chat session.
 
     Args:
         chat_session_id: The UUID of the chat session
-        redis_client: Redis client to use (tenant-aware client that auto-prefixes keys)
+        cache: Tenant-aware cache backend
         value: True to set the fence (stop signal), False to clear it
     """
     fence_key = _get_fence_key(chat_session_id)
     if not value:
-        redis_client.delete(fence_key)
+        cache.delete(fence_key)
         return
-
-    redis_client.set(fence_key, 0, ex=FENCE_TTL)
+    cache.set(fence_key, 0, ex=FENCE_TTL)
 
 
-def is_connected(chat_session_id: UUID, redis_client: Redis) -> bool:
-    """
-    Check if the chat session should continue (not stopped).
+def is_connected(chat_session_id: UUID, cache: CacheBackend) -> bool:
+    """Check if the chat session should continue (not stopped).
 
     Args:
         chat_session_id: The UUID of the chat session to check
-        redis_client: Redis client to use for checking the stop signal (tenant-aware client that auto-prefixes keys)
+        cache: Tenant-aware cache backend
 
     Returns:
         True if the session should continue, False if it should stop
     """
-    fence_key = _get_fence_key(chat_session_id)
-    return not bool(redis_client.exists(fence_key))
+    return not cache.exists(_get_fence_key(chat_session_id))
 
 
-def reset_cancel_status(chat_session_id: UUID, redis_client: Redis) -> None:
-    """
-    Clear the stop signal for a chat session.
+def reset_cancel_status(chat_session_id: UUID, cache: CacheBackend) -> None:
+    """Clear the stop signal for a chat session.
 
     Args:
         chat_session_id: The UUID of the chat session
-        redis_client: Redis client to use (tenant-aware client that auto-prefixes keys)
+        cache: Tenant-aware cache backend
     """
-    fence_key = _get_fence_key(chat_session_id)
-    redis_client.delete(fence_key)
+    cache.delete(_get_fence_key(chat_session_id))

backend/onyx/configs/app_configs.py

@@ -6,6 +6,7 @@ from datetime import timezone
 from typing import cast
 
 from onyx.auth.schemas import AuthBackend
+from onyx.cache.interface import CacheBackendType
 from onyx.configs.constants import AuthType
 from onyx.configs.constants import QueryHistoryType
 from onyx.file_processing.enums import HtmlBasedConnectorTransformLinksStrategy
@@ -54,6 +55,12 @@ DISABLE_USER_KNOWLEDGE = os.environ.get("DISABLE_USER_KNOWLEDGE", "").lower() ==
 # are disabled but core chat, tools, user file uploads, and Projects still work.
 DISABLE_VECTOR_DB = os.environ.get("DISABLE_VECTOR_DB", "").lower() == "true"
 
+# Which backend to use for caching, locks, and ephemeral state.
+# "redis" (default) or "postgres" (only valid when DISABLE_VECTOR_DB=true).
+CACHE_BACKEND = CacheBackendType(
+    os.environ.get("CACHE_BACKEND", CacheBackendType.REDIS)
+)
+
 # Maximum token count for a single uploaded file. Files exceeding this are rejected.
 # Defaults to 100k tokens (or 10M when vector DB is disabled).
 _DEFAULT_FILE_TOKEN_LIMIT = 10_000_000 if DISABLE_VECTOR_DB else 100_000
@@ -488,14 +495,7 @@ CELERY_WORKER_PRIMARY_POOL_OVERFLOW = int(
     os.environ.get("CELERY_WORKER_PRIMARY_POOL_OVERFLOW") or 4
 )
 
-# Consolidated background worker (light, docprocessing, docfetching, heavy, monitoring, user_file_processing)
-# separate workers' defaults: light=24, docprocessing=6, docfetching=1, heavy=4, kg=2, monitoring=1, user_file=2
-# Total would be 40, but we use a more conservative default of 20 for the consolidated worker
-CELERY_WORKER_BACKGROUND_CONCURRENCY = int(
-    os.environ.get("CELERY_WORKER_BACKGROUND_CONCURRENCY") or 20
-)
-
-# Individual worker concurrency settings (used when USE_LIGHTWEIGHT_BACKGROUND_WORKER is False or on Kuberenetes deployments)
+# Individual worker concurrency settings
 CELERY_WORKER_HEAVY_CONCURRENCY = int(
     os.environ.get("CELERY_WORKER_HEAVY_CONCURRENCY") or 4
 )
@@ -812,7 +812,9 @@ RERANK_COUNT = int(os.environ.get("RERANK_COUNT") or 1000)
 # Tool Configs
 #####
 # Code Interpreter Service Configuration
-CODE_INTERPRETER_BASE_URL = os.environ.get("CODE_INTERPRETER_BASE_URL")
+CODE_INTERPRETER_BASE_URL = os.environ.get(
+    "CODE_INTERPRETER_BASE_URL", "http://localhost:8000"
+)
 
 CODE_INTERPRETER_DEFAULT_TIMEOUT_MS = int(
     os.environ.get("CODE_INTERPRETER_DEFAULT_TIMEOUT_MS") or 60_000
@@ -893,6 +895,9 @@ CUSTOM_ANSWER_VALIDITY_CONDITIONS = json.loads(
 )
 
 VESPA_REQUEST_TIMEOUT = int(os.environ.get("VESPA_REQUEST_TIMEOUT") or "15")
+VESPA_MIGRATION_REQUEST_TIMEOUT_S = int(
+    os.environ.get("VESPA_MIGRATION_REQUEST_TIMEOUT_S") or "120"
+)
 
 SYSTEM_RECURSION_LIMIT = int(os.environ.get("SYSTEM_RECURSION_LIMIT") or "1000")
 

backend/onyx/configs/constants.py

@@ -84,7 +84,6 @@ POSTGRES_CELERY_WORKER_LIGHT_APP_NAME = "celery_worker_light"
 POSTGRES_CELERY_WORKER_DOCPROCESSING_APP_NAME = "celery_worker_docprocessing"
 POSTGRES_CELERY_WORKER_DOCFETCHING_APP_NAME = "celery_worker_docfetching"
 POSTGRES_CELERY_WORKER_INDEXING_CHILD_APP_NAME = "celery_worker_indexing_child"
-POSTGRES_CELERY_WORKER_BACKGROUND_APP_NAME = "celery_worker_background"
 POSTGRES_CELERY_WORKER_HEAVY_APP_NAME = "celery_worker_heavy"
 POSTGRES_CELERY_WORKER_MONITORING_APP_NAME = "celery_worker_monitoring"
 POSTGRES_CELERY_WORKER_USER_FILE_PROCESSING_APP_NAME = (

backend/onyx/db/chat.py

@@ -5,7 +5,6 @@ from datetime import timezone
 from typing import Tuple
 from uuid import UUID
 
-from fastapi import HTTPException
 from sqlalchemy import delete
 from sqlalchemy import desc
 from sqlalchemy import exists
@@ -32,6 +31,8 @@ from onyx.db.models import SearchDoc as DBSearchDoc
 from onyx.db.models import ToolCall
 from onyx.db.models import User
 from onyx.db.persona import get_best_persona_id_for_user
+from onyx.error_handling.error_codes import OnyxErrorCode
+from onyx.error_handling.exceptions import OnyxError
 from onyx.file_store.file_store import get_default_file_store
 from onyx.file_store.models import FileDescriptor
 from onyx.llm.override_models import LLMOverride
@@ -98,6 +99,7 @@ def get_chat_sessions_by_user(
     db_session: Session,
     include_onyxbot_flows: bool = False,
     limit: int = 50,
+    before: datetime | None = None,
     project_id: int | None = None,
     only_non_project_chats: bool = False,
     include_failed_chats: bool = False,
@@ -112,6 +114,9 @@ def get_chat_sessions_by_user(
     if deleted is not None:
         stmt = stmt.where(ChatSession.deleted == deleted)
 
+    if before is not None:
+        stmt = stmt.where(ChatSession.time_updated < before)
+
     if limit:
         stmt = stmt.limit(limit)
 
@@ -223,7 +228,9 @@ def duplicate_chat_session_for_user_from_slack(
         db_session=db_session,
     )
     if not chat_session:
-        raise HTTPException(status_code=400, detail="Invalid Chat Session ID provided")
+        raise OnyxError(
+            OnyxErrorCode.SESSION_NOT_FOUND, "Invalid Chat Session ID provided"
+        )
 
     # This enforces permissions and sets a default
     new_persona_id = get_best_persona_id_for_user(

backend/onyx/db/connector_credential_pair.py

@@ -2,7 +2,6 @@ from datetime import datetime
 from enum import Enum
 from typing import TypeVarTuple
 
-from fastapi import HTTPException
 from sqlalchemy import delete
 from sqlalchemy import desc
 from sqlalchemy import exists
@@ -32,6 +31,8 @@ from onyx.db.models import User
 from onyx.db.models import User__UserGroup
 from onyx.db.models import UserGroup__ConnectorCredentialPair
 from onyx.db.models import UserRole
+from onyx.error_handling.error_codes import OnyxErrorCode
+from onyx.error_handling.exceptions import OnyxError
 from onyx.server.models import StatusResponse
 from onyx.utils.logger import setup_logger
 from onyx.utils.variable_functionality import fetch_ee_implementation_or_noop
@@ -539,7 +540,7 @@ def add_credential_to_connector(
         )
 
     if connector is None:
-        raise HTTPException(status_code=404, detail="Connector does not exist")
+        raise OnyxError(OnyxErrorCode.CONNECTOR_NOT_FOUND, "Connector does not exist")
 
     if access_type == AccessType.SYNC:
         if not fetch_ee_implementation_or_noop(
@@ -547,9 +548,9 @@ def add_credential_to_connector(
             "check_if_valid_sync_source",
             noop_return_value=True,
         )(connector.source):
-            raise HTTPException(
-                status_code=400,
-                detail=f"Connector of type {connector.source} does not support SYNC access type",
+            raise OnyxError(
+                OnyxErrorCode.VALIDATION_ERROR,
+                f"Connector of type {connector.source} does not support SYNC access type",
             )
 
     if credential is None:
@@ -557,9 +558,9 @@ def add_credential_to_connector(
             f"Credential {credential_id} does not exist or does not belong to user"
         )
         logger.error(error_msg)
-        raise HTTPException(
-            status_code=401,
-            detail=error_msg,
+        raise OnyxError(
+            OnyxErrorCode.CREDENTIAL_NOT_FOUND,
+            error_msg,
         )
 
     existing_association = (
@@ -622,12 +623,12 @@ def remove_credential_from_connector(
     )
 
     if connector is None:
-        raise HTTPException(status_code=404, detail="Connector does not exist")
+        raise OnyxError(OnyxErrorCode.CONNECTOR_NOT_FOUND, "Connector does not exist")
 
     if credential is None:
-        raise HTTPException(
-            status_code=404,
-            detail="Credential does not exist or does not belong to user",
+        raise OnyxError(
+            OnyxErrorCode.CREDENTIAL_NOT_FOUND,
+            "Credential does not exist or does not belong to user",
         )
 
     association = get_connector_credential_pair_for_user(

backend/onyx/db/document_set.py

@@ -13,6 +13,7 @@ from sqlalchemy.orm import aliased
 from sqlalchemy.orm import selectinload
 from sqlalchemy.orm import Session
 
+from onyx.configs.app_configs import DISABLE_VECTOR_DB
 from onyx.db.connector_credential_pair import get_cc_pair_groups_for_ids
 from onyx.db.connector_credential_pair import get_connector_credential_pairs
 from onyx.db.enums import AccessType
@@ -246,6 +247,7 @@ def insert_document_set(
             description=document_set_creation_request.description,
             user_id=user_id,
             is_public=document_set_creation_request.is_public,
+            is_up_to_date=DISABLE_VECTOR_DB,
             time_last_modified_by_user=func.now(),
         )
         db_session.add(new_document_set_row)
@@ -336,7 +338,8 @@ def update_document_set(
             )
 
         document_set_row.description = document_set_update_request.description
-        document_set_row.is_up_to_date = False
+        if not DISABLE_VECTOR_DB:
+            document_set_row.is_up_to_date = False
         document_set_row.is_public = document_set_update_request.is_public
         document_set_row.time_last_modified_by_user = func.now()
         versioned_private_doc_set_fn = fetch_versioned_implementation(

backend/onyx/db/engine/async_sql_engine.py

@@ -4,7 +4,6 @@ from typing import Any
 from typing import AsyncContextManager
 
 import asyncpg  # type: ignore
-from fastapi import HTTPException
 from sqlalchemy import event
 from sqlalchemy import pool
 from sqlalchemy.ext.asyncio import AsyncEngine
@@ -28,6 +27,8 @@ from onyx.db.engine.sql_engine import build_connection_string
 from onyx.db.engine.sql_engine import is_valid_schema_name
 from onyx.db.engine.sql_engine import SqlEngine
 from onyx.db.engine.sql_engine import USE_IAM_AUTH
+from onyx.error_handling.error_codes import OnyxErrorCode
+from onyx.error_handling.exceptions import OnyxError
 from shared_configs.configs import MULTI_TENANT
 from shared_configs.configs import POSTGRES_DEFAULT_SCHEMA_STANDARD_VALUE
 from shared_configs.contextvars import get_current_tenant_id
@@ -114,7 +115,7 @@ async def get_async_session(
         tenant_id = get_current_tenant_id()
 
     if not is_valid_schema_name(tenant_id):
-        raise HTTPException(status_code=400, detail="Invalid tenant ID")
+        raise OnyxError(OnyxErrorCode.VALIDATION_ERROR, "Invalid tenant ID")
 
     engine = get_sqlalchemy_async_engine()
 

backend/onyx/db/engine/sql_engine.py

@@ -6,7 +6,6 @@ from collections.abc import Generator
 from contextlib import contextmanager
 from typing import Any
 
-from fastapi import HTTPException
 from sqlalchemy import event
 from sqlalchemy import pool
 from sqlalchemy.engine import create_engine
@@ -27,6 +26,8 @@ from onyx.configs.app_configs import POSTGRES_USE_NULL_POOL
 from onyx.configs.app_configs import POSTGRES_USER
 from onyx.configs.constants import POSTGRES_UNKNOWN_APP_NAME
 from onyx.db.engine.iam_auth import provide_iam_token
+from onyx.error_handling.error_codes import OnyxErrorCode
+from onyx.error_handling.exceptions import OnyxError
 from onyx.server.utils import BasicAuthenticationError
 from onyx.utils.logger import setup_logger
 from shared_configs.configs import MULTI_TENANT
@@ -344,7 +345,7 @@ def get_session_with_tenant(*, tenant_id: str) -> Generator[Session, None, None]
     engine = get_sqlalchemy_engine()
 
     if not is_valid_schema_name(tenant_id):
-        raise HTTPException(status_code=400, detail="Invalid tenant ID")
+        raise OnyxError(OnyxErrorCode.VALIDATION_ERROR, "Invalid tenant ID")
 
     # no need to use the schema translation map for self-hosted + default schema
     if not MULTI_TENANT and tenant_id == POSTGRES_DEFAULT_SCHEMA_STANDARD_VALUE:
@@ -371,7 +372,7 @@ def get_session() -> Generator[Session, None, None]:
         raise BasicAuthenticationError(detail="User must authenticate")
 
     if not is_valid_schema_name(tenant_id):
-        raise HTTPException(status_code=400, detail="Invalid tenant ID")
+        raise OnyxError(OnyxErrorCode.VALIDATION_ERROR, "Invalid tenant ID")
 
     with get_session_with_current_tenant() as db_session:
         yield db_session
@@ -390,7 +391,7 @@ def get_db_readonly_user_session_with_current_tenant() -> (
     readonly_engine = get_readonly_sqlalchemy_engine()
 
     if not is_valid_schema_name(tenant_id):
-        raise HTTPException(status_code=400, detail="Invalid tenant ID")
+        raise OnyxError(OnyxErrorCode.VALIDATION_ERROR, "Invalid tenant ID")
 
     # no need to use the schema translation map for self-hosted + default schema
     if not MULTI_TENANT and tenant_id == POSTGRES_DEFAULT_SCHEMA_STANDARD_VALUE:

backend/onyx/db/enums.py

@@ -186,6 +186,7 @@ class EmbeddingPrecision(str, PyEnum):
 
 class UserFileStatus(str, PyEnum):
     PROCESSING = "PROCESSING"
+    INDEXING = "INDEXING"
     COMPLETED = "COMPLETED"
     FAILED = "FAILED"
     CANCELED = "CANCELED"

backend/onyx/db/feedback.py

@@ -2,7 +2,6 @@ from datetime import datetime
 from datetime import timezone
 from uuid import UUID
 
-from fastapi import HTTPException
 from sqlalchemy import and_
 from sqlalchemy import asc
 from sqlalchemy import delete
@@ -26,6 +25,8 @@ from onyx.db.models import User
 from onyx.db.models import User__UserGroup
 from onyx.db.models import UserGroup__ConnectorCredentialPair
 from onyx.db.models import UserRole
+from onyx.error_handling.error_codes import OnyxErrorCode
+from onyx.error_handling.exceptions import OnyxError
 from onyx.utils.logger import setup_logger
 
 logger = setup_logger()
@@ -134,8 +135,9 @@ def update_document_boost_for_user(
     stmt = _add_user_filters(stmt, user, get_editable=True)
     result: DbDocument | None = db_session.execute(stmt).scalar_one_or_none()
     if result is None:
-        raise HTTPException(
-            status_code=400, detail="Document is not editable by this user"
+        raise OnyxError(
+            OnyxErrorCode.UNAUTHORIZED,
+            "Document is not editable by this user",
         )
 
     result.boost = boost
@@ -156,8 +158,9 @@ def update_document_hidden_for_user(
     stmt = _add_user_filters(stmt, user, get_editable=True)
     result = db_session.execute(stmt).scalar_one_or_none()
     if result is None:
-        raise HTTPException(
-            status_code=400, detail="Document is not editable by this user"
+        raise OnyxError(
+            OnyxErrorCode.UNAUTHORIZED,
+            "Document is not editable by this user",
         )
 
     result.hidden = hidden

backend/onyx/db/input_prompt.py

@@ -1,6 +1,5 @@
 from uuid import UUID
 
-from fastapi import HTTPException
 from sqlalchemy import or_
 from sqlalchemy import select
 from sqlalchemy.dialects.postgresql import insert as pg_insert
@@ -11,6 +10,8 @@ from sqlalchemy.orm import Session
 from onyx.db.models import InputPrompt
 from onyx.db.models import InputPrompt__User
 from onyx.db.models import User
+from onyx.error_handling.error_codes import OnyxErrorCode
+from onyx.error_handling.exceptions import OnyxError
 from onyx.server.features.input_prompt.models import InputPromptSnapshot
 from onyx.server.manage.models import UserInfo
 from onyx.utils.logger import setup_logger
@@ -54,9 +55,9 @@ def insert_input_prompt(
     input_prompt = result.scalar_one_or_none()
 
     if input_prompt is None:
-        raise HTTPException(
-            status_code=409,
-            detail=f"A prompt shortcut with the name '{prompt}' already exists",
+        raise OnyxError(
+            OnyxErrorCode.DUPLICATE_RESOURCE,
+            f"A prompt shortcut with the name '{prompt}' already exists",
         )
 
     db_session.commit()
@@ -78,7 +79,7 @@ def update_input_prompt(
         raise ValueError(f"No input prompt with id {input_prompt_id}")
 
     if not validate_user_prompt_authorization(user, input_prompt):
-        raise HTTPException(status_code=401, detail="You don't own this prompt")
+        raise OnyxError(OnyxErrorCode.UNAUTHORIZED, "You don't own this prompt")
 
     input_prompt.prompt = prompt
     input_prompt.content = content
@@ -88,9 +89,9 @@ def update_input_prompt(
         db_session.commit()
     except IntegrityError:
         db_session.rollback()
-        raise HTTPException(
-            status_code=409,
-            detail=f"A prompt shortcut with the name '{prompt}' already exists",
+        raise OnyxError(
+            OnyxErrorCode.DUPLICATE_RESOURCE,
+            f"A prompt shortcut with the name '{prompt}' already exists",
         )
 
     return input_prompt
@@ -121,7 +122,7 @@ def remove_public_input_prompt(input_prompt_id: int, db_session: Session) -> Non
         raise ValueError(f"No input prompt with id {input_prompt_id}")
 
     if not input_prompt.is_public:
-        raise HTTPException(status_code=400, detail="This prompt is not public")
+        raise OnyxError(OnyxErrorCode.VALIDATION_ERROR, "This prompt is not public")
 
     db_session.delete(input_prompt)
     db_session.commit()
@@ -140,12 +141,13 @@ def remove_input_prompt(
         raise ValueError(f"No input prompt with id {input_prompt_id}")
 
     if input_prompt.is_public and not delete_public:
-        raise HTTPException(
-            status_code=400, detail="Cannot delete public prompts with this method"
+        raise OnyxError(
+            OnyxErrorCode.VALIDATION_ERROR,
+            "Cannot delete public prompts with this method",
         )
 
     if not validate_user_prompt_authorization(user, input_prompt):
-        raise HTTPException(status_code=401, detail="You do not own this prompt")
+        raise OnyxError(OnyxErrorCode.UNAUTHORIZED, "You do not own this prompt")
 
     db_session.delete(input_prompt)
     db_session.commit()
@@ -167,7 +169,7 @@ def fetch_input_prompt_by_id(
     result = db_session.scalar(query)
 
     if result is None:
-        raise HTTPException(422, "No input prompt found")
+        raise OnyxError(OnyxErrorCode.NOT_FOUND, "No input prompt found")
 
     return result
 

backend/onyx/db/llm.py

@@ -109,45 +109,38 @@ def can_user_access_llm_provider(
         is_admin: If True, bypass user group restrictions but still respect persona restrictions
 
     Access logic:
-    1. If is_public=True → everyone has access (public override)
-    2. If is_public=False:
-       - Both groups AND personas set → must satisfy BOTH (AND logic, admins bypass group check)
-       - Only groups set → must be in one of the groups (OR across groups, admins bypass)
-       - Only personas set → must use one of the personas (OR across personas, applies to admins)
-       - Neither set → NOBODY has access unless admin (locked, admin-only)
+    - is_public controls USER access (group bypass): when True, all users can access
+      regardless of group membership. When False, user must be in a whitelisted group
+      (or be admin).
+    - Persona restrictions are ALWAYS enforced when set, regardless of is_public.
+      This allows admins to make a provider available to all users while still
+      restricting which personas (assistants) can use it.
+
+    Decision matrix:
+    1. is_public=True, no personas set → everyone has access
+    2. is_public=True, personas set → all users, but only whitelisted personas
+    3. is_public=False, groups+personas set → must satisfy BOTH (admins bypass groups)
+    4. is_public=False, only groups set → must be in group (admins bypass)
+    5. is_public=False, only personas set → must use whitelisted persona
+    6. is_public=False, neither set → admin-only (locked)
     """
-    # Public override - everyone has access
-    if provider.is_public:
-        return True
-
-    # Extract IDs once to avoid multiple iterations
-    provider_group_ids = (
-        {group.id for group in provider.groups} if provider.groups else set()
-    )
-    provider_persona_ids = (
-        {p.id for p in provider.personas} if provider.personas else set()
-    )
-
+    provider_group_ids = {g.id for g in (provider.groups or [])}
+    provider_persona_ids = {p.id for p in (provider.personas or [])}
     has_groups = bool(provider_group_ids)
     has_personas = bool(provider_persona_ids)
 
-    # Both groups AND personas set → AND logic (must satisfy both)
-    if has_groups and has_personas:
-        # Admins bypass group check but still must satisfy persona restrictions
-        user_in_group = is_admin or bool(user_group_ids & provider_group_ids)
-        persona_allowed = persona.id in provider_persona_ids if persona else False
-        return user_in_group and persona_allowed
+    # Persona restrictions are always enforced when set, regardless of is_public
+    if has_personas and not (persona and persona.id in provider_persona_ids):
+        return False
+
+    if provider.is_public:
+        return True
 
-    # Only groups set → user must be in one of the groups (admins bypass)
     if has_groups:
         return is_admin or bool(user_group_ids & provider_group_ids)
 
-    # Only personas set → persona must be in allowed list (applies to admins too)
-    if has_personas:
-        return persona.id in provider_persona_ids if persona else False
-
-    # Neither groups nor personas set, and not public → admins can access
-    return is_admin
+    # No groups: either persona-whitelisted (already passed) or admin-only if locked
+    return has_personas or is_admin
 
 
 def validate_persona_ids_exist(
@@ -539,6 +532,7 @@ def fetch_default_model(
 ) -> ModelConfiguration | None:
     model_config = db_session.scalar(
         select(ModelConfiguration)
+        .options(selectinload(ModelConfiguration.llm_provider))
         .join(LLMModelFlow)
         .where(
             ModelConfiguration.is_visible == True,  # noqa: E712

backend/onyx/db/models.py

@@ -103,7 +103,6 @@ from onyx.utils.encryption import encrypt_string_to_bytes
 from onyx.utils.sensitive import SensitiveValue
 from onyx.utils.headers import HeaderItemDict
 from shared_configs.enums import EmbeddingProvider
-from onyx.context.search.enums import RecencyBiasSetting
 
 # TODO: After anonymous user migration has been deployed, make user_id columns NOT NULL
 # and update Mapped[User | None] relationships to Mapped[User] where needed.
@@ -3265,19 +3264,6 @@ class Persona(Base):
     )
     name: Mapped[str] = mapped_column(String)
     description: Mapped[str] = mapped_column(String)
-    # Number of chunks to pass to the LLM for generation.
-    num_chunks: Mapped[float | None] = mapped_column(Float, nullable=True)
-    chunks_above: Mapped[int] = mapped_column(Integer)
-    chunks_below: Mapped[int] = mapped_column(Integer)
-    # Pass every chunk through LLM for evaluation, fairly expensive
-    # Can be turned off globally by admin, in which case, this setting is ignored
-    llm_relevance_filter: Mapped[bool] = mapped_column(Boolean)
-    # Enables using LLM to extract time and source type filters
-    # Can also be admin disabled globally
-    llm_filter_extraction: Mapped[bool] = mapped_column(Boolean)
-    recency_bias: Mapped[RecencyBiasSetting] = mapped_column(
-        Enum(RecencyBiasSetting, native_enum=False)
-    )
 
     # Allows the persona to specify a specific default LLM model
     # NOTE: only is applied on the actual response generation - is not used for things like
@@ -3304,11 +3290,8 @@ class Persona(Base):
     # Treated specially (cannot be user edited etc.)
     builtin_persona: Mapped[bool] = mapped_column(Boolean, default=False)
 
-    # Default personas are personas created by admins and are automatically added
-    # to all users' assistants list.
-    is_default_persona: Mapped[bool] = mapped_column(
-        Boolean, default=False, nullable=False
-    )
+    # Featured personas are highlighted in the UI
+    featured: Mapped[bool] = mapped_column(Boolean, default=False)
     # controls whether the persona is available to be selected by users
     is_visible: Mapped[bool] = mapped_column(Boolean, default=True)
     # controls the ordering of personas in the UI
@@ -4943,7 +4926,9 @@ class ScimUserMapping(Base):
     __tablename__ = "scim_user_mapping"
 
     id: Mapped[int] = mapped_column(Integer, primary_key=True)
-    external_id: Mapped[str] = mapped_column(String, unique=True, index=True)
+    external_id: Mapped[str | None] = mapped_column(
+        String, unique=True, index=True, nullable=True
+    )
     user_id: Mapped[UUID] = mapped_column(
         ForeignKey("user.id", ondelete="CASCADE"), unique=True, nullable=False
     )
@@ -5000,3 +4985,25 @@ class CodeInterpreterServer(Base):
 
     id: Mapped[int] = mapped_column(Integer, primary_key=True)
     server_enabled: Mapped[bool] = mapped_column(Boolean, nullable=False, default=True)
+
+
+class CacheStore(Base):
+    """Key-value cache table used by ``PostgresCacheBackend``.
+
+    Replaces Redis for simple KV caching, locks, and list operations
+    when ``CACHE_BACKEND=postgres`` (NO_VECTOR_DB deployments).
+
+    Intentionally separate from ``KVStore``:
+    - Stores raw bytes (LargeBinary) vs JSONB, matching Redis semantics.
+    - Has ``expires_at`` for TTL; rows are periodically garbage-collected.
+    - Holds ephemeral data (tokens, stop signals, lock state) not
+      persistent application config, so cleanup can be aggressive.
+    """
+
+    __tablename__ = "cache_store"
+
+    key: Mapped[str] = mapped_column(String, primary_key=True)
+    value: Mapped[bytes | None] = mapped_column(LargeBinary, nullable=True)
+    expires_at: Mapped[datetime.datetime | None] = mapped_column(
+        DateTime(timezone=True), nullable=True
+    )

backend/onyx/db/persona.py

@@ -3,7 +3,6 @@ from datetime import datetime
 from enum import Enum
 from uuid import UUID
 
-from fastapi import HTTPException
 from sqlalchemy import exists
 from sqlalchemy import func
 from sqlalchemy import not_
@@ -18,11 +17,8 @@ from sqlalchemy.orm import Session
 from onyx.access.hierarchy_access import get_user_external_group_ids
 from onyx.auth.schemas import UserRole
 from onyx.configs.app_configs import CURATORS_CANNOT_VIEW_OR_EDIT_NON_OWNED_ASSISTANTS
-from onyx.configs.chat_configs import CONTEXT_CHUNKS_ABOVE
-from onyx.configs.chat_configs import CONTEXT_CHUNKS_BELOW
 from onyx.configs.constants import DEFAULT_PERSONA_ID
 from onyx.configs.constants import NotificationType
-from onyx.context.search.enums import RecencyBiasSetting
 from onyx.db.constants import SLACK_BOT_PERSONA_PREFIX
 from onyx.db.document_access import get_accessible_documents_by_ids
 from onyx.db.models import ConnectorCredentialPair
@@ -41,6 +37,8 @@ from onyx.db.models import User__UserGroup
 from onyx.db.models import UserFile
 from onyx.db.models import UserGroup
 from onyx.db.notification import create_notification
+from onyx.error_handling.error_codes import OnyxErrorCode
+from onyx.error_handling.exceptions import OnyxError
 from onyx.server.features.persona.models import FullPersonaSnapshot
 from onyx.server.features.persona.models import MinimalPersonaSnapshot
 from onyx.server.features.persona.models import PersonaSharedNotificationData
@@ -147,9 +145,9 @@ def fetch_persona_by_id_for_user(
     stmt = _add_user_filters(stmt=stmt, user=user, get_editable=get_editable)
     persona = db_session.scalars(stmt).one_or_none()
     if not persona:
-        raise HTTPException(
-            status_code=403,
-            detail=f"Persona with ID {persona_id} does not exist or user is not authorized to access it",
+        raise OnyxError(
+            OnyxErrorCode.INSUFFICIENT_PERMISSIONS,
+            f"Persona with ID {persona_id} does not exist or user is not authorized to access it",
         )
     return persona
 
@@ -254,13 +252,15 @@ def create_update_persona(
     # Permission to actually use these is checked later
 
     try:
-        # Default persona validation
-        if create_persona_request.is_default_persona:
-            # Curators can edit default personas, but not make them
+        # Featured persona validation
+        if create_persona_request.featured:
+
+            # Curators can edit featured personas, but not make them
+            # TODO this will be reworked soon with RBAC permissions feature
             if user.role == UserRole.CURATOR or user.role == UserRole.GLOBAL_CURATOR:
                 pass
             elif user.role != UserRole.ADMIN:
-                raise ValueError("Only admins can make a default persona")
+                raise ValueError("Only admins can make a featured persona")
 
         # Convert incoming string UUIDs to UUID objects for DB operations
         converted_user_file_ids = None
@@ -281,7 +281,6 @@ def create_update_persona(
             document_set_ids=create_persona_request.document_set_ids,
             tool_ids=create_persona_request.tool_ids,
             is_public=create_persona_request.is_public,
-            recency_bias=create_persona_request.recency_bias,
             llm_model_provider_override=create_persona_request.llm_model_provider_override,
             llm_model_version_override=create_persona_request.llm_model_version_override,
             starter_messages=create_persona_request.starter_messages,
@@ -295,10 +294,7 @@ def create_update_persona(
             remove_image=create_persona_request.remove_image,
             search_start_date=create_persona_request.search_start_date,
             label_ids=create_persona_request.label_ids,
-            num_chunks=create_persona_request.num_chunks,
-            llm_relevance_filter=create_persona_request.llm_relevance_filter,
-            llm_filter_extraction=create_persona_request.llm_filter_extraction,
-            is_default_persona=create_persona_request.is_default_persona,
+            featured=create_persona_request.featured,
             user_file_ids=converted_user_file_ids,
             commit=False,
             hierarchy_node_ids=create_persona_request.hierarchy_node_ids,
@@ -320,7 +316,7 @@ def create_update_persona(
 
     except ValueError as e:
         logger.exception("Failed to create persona")
-        raise HTTPException(status_code=400, detail=str(e))
+        raise OnyxError(OnyxErrorCode.VALIDATION_ERROR, str(e))
 
     return FullPersonaSnapshot.from_model(persona)
 
@@ -874,10 +870,6 @@ def upsert_persona(
     user: User | None,
     name: str,
     description: str,
-    num_chunks: float,
-    llm_relevance_filter: bool,
-    llm_filter_extraction: bool,
-    recency_bias: RecencyBiasSetting,
     llm_model_provider_override: str | None,
     llm_model_version_override: str | None,
     starter_messages: list[StarterMessage] | None,
@@ -898,13 +890,11 @@ def upsert_persona(
     remove_image: bool | None = None,
     search_start_date: datetime | None = None,
     builtin_persona: bool = False,
-    is_default_persona: bool | None = None,
+    featured: bool | None = None,
     label_ids: list[int] | None = None,
     user_file_ids: list[UUID] | None = None,
     hierarchy_node_ids: list[int] | None = None,
     document_ids: list[str] | None = None,
-    chunks_above: int = CONTEXT_CHUNKS_ABOVE,
-    chunks_below: int = CONTEXT_CHUNKS_BELOW,
     replace_base_system_prompt: bool = False,
 ) -> Persona:
     """
@@ -1015,12 +1005,6 @@ def upsert_persona(
         # `default` and `built-in` properties can only be set when creating a persona.
         existing_persona.name = name
         existing_persona.description = description
-        existing_persona.num_chunks = num_chunks
-        existing_persona.chunks_above = chunks_above
-        existing_persona.chunks_below = chunks_below
-        existing_persona.llm_relevance_filter = llm_relevance_filter
-        existing_persona.llm_filter_extraction = llm_filter_extraction
-        existing_persona.recency_bias = recency_bias
         existing_persona.llm_model_provider_override = llm_model_provider_override
         existing_persona.llm_model_version_override = llm_model_version_override
         existing_persona.starter_messages = starter_messages
@@ -1034,10 +1018,8 @@ def upsert_persona(
         if label_ids is not None:
             existing_persona.labels.clear()
             existing_persona.labels = labels or []
-        existing_persona.is_default_persona = (
-            is_default_persona
-            if is_default_persona is not None
-            else existing_persona.is_default_persona
+        existing_persona.featured = (
+            featured if featured is not None else existing_persona.featured
         )
         # Update embedded prompt fields if provided
         if system_prompt is not None:
@@ -1090,12 +1072,6 @@ def upsert_persona(
             is_public=is_public,
             name=name,
             description=description,
-            num_chunks=num_chunks,
-            chunks_above=chunks_above,
-            chunks_below=chunks_below,
-            llm_relevance_filter=llm_relevance_filter,
-            llm_filter_extraction=llm_filter_extraction,
-            recency_bias=recency_bias,
             builtin_persona=builtin_persona,
             system_prompt=system_prompt or "",
             task_prompt=task_prompt or "",
@@ -1111,9 +1087,7 @@ def upsert_persona(
             display_priority=display_priority,
             is_visible=is_visible,
             search_start_date=search_start_date,
-            is_default_persona=(
-                is_default_persona if is_default_persona is not None else False
-            ),
+            featured=(featured if featured is not None else False),
             user_files=user_files or [],
             labels=labels or [],
             hierarchy_nodes=hierarchy_nodes or [],
@@ -1158,9 +1132,9 @@ def delete_old_default_personas(
     db_session.commit()
 
 
-def update_persona_is_default(
+def update_persona_featured(
     persona_id: int,
-    is_default: bool,
+    featured: bool,
     db_session: Session,
     user: User,
 ) -> None:
@@ -1168,7 +1142,7 @@ def update_persona_is_default(
         db_session=db_session, persona_id=persona_id, user=user, get_editable=True
     )
 
-    persona.is_default_persona = is_default
+    persona.featured = featured
     db_session.commit()
 
 

backend/onyx/db/projects.py

@@ -3,14 +3,14 @@ import uuid
 from typing import List
 from uuid import UUID
 
-from fastapi import HTTPException
 from fastapi import UploadFile
 from pydantic import BaseModel
 from pydantic import ConfigDict
 from sqlalchemy import func
 from sqlalchemy.orm import Session
+from starlette.background import BackgroundTasks
 
-from onyx.background.celery.versioned_apps.client import app as client_app
+from onyx.configs.app_configs import DISABLE_VECTOR_DB
 from onyx.configs.constants import FileOrigin
 from onyx.configs.constants import OnyxCeleryPriority
 from onyx.configs.constants import OnyxCeleryQueues
@@ -19,6 +19,8 @@ from onyx.db.models import Project__UserFile
 from onyx.db.models import User
 from onyx.db.models import UserFile
 from onyx.db.models import UserProject
+from onyx.error_handling.error_codes import OnyxErrorCode
+from onyx.error_handling.exceptions import OnyxError
 from onyx.server.documents.connector import upload_files
 from onyx.server.features.projects.projects_file_utils import categorize_uploaded_files
 from onyx.server.features.projects.projects_file_utils import RejectedFile
@@ -51,7 +53,7 @@ def create_user_files(
 ) -> CategorizedFilesResult:
 
     # Categorize the files
-    categorized_files = categorize_uploaded_files(files)
+    categorized_files = categorize_uploaded_files(files, db_session)
     # NOTE: At the moment, zip metadata is not used for user files.
     # Should revisit to decide whether this should be a feature.
     upload_response = upload_files(categorized_files.acceptable, FileOrigin.USER_FILE)
@@ -105,11 +107,11 @@ def upload_files_to_user_files_with_indexing(
     user: User,
     temp_id_map: dict[str, str] | None,
     db_session: Session,
+    background_tasks: BackgroundTasks | None = None,
 ) -> CategorizedFilesResult:
-    # Validate project ownership if a project_id is provided
     if project_id is not None and user is not None:
         if not check_project_ownership(project_id, user.id, db_session):
-            raise HTTPException(status_code=404, detail="Project not found")
+            raise OnyxError(OnyxErrorCode.NOT_FOUND, "Project not found")
 
     categorized_files_result = create_user_files(
         files,
@@ -127,16 +129,27 @@ def upload_files_to_user_files_with_indexing(
         logger.warning(
             f"File {rejected_file.filename} rejected for {rejected_file.reason}"
         )
-    for user_file in user_files:
-        task = client_app.send_task(
-            OnyxCeleryTask.PROCESS_SINGLE_USER_FILE,
-            kwargs={"user_file_id": user_file.id, "tenant_id": tenant_id},
-            queue=OnyxCeleryQueues.USER_FILE_PROCESSING,
-            priority=OnyxCeleryPriority.HIGH,
-        )
-        logger.info(
-            f"Triggered indexing for user_file_id={user_file.id} with task_id={task.id}"
-        )
+
+    if DISABLE_VECTOR_DB and background_tasks is not None:
+        from onyx.background.task_utils import drain_processing_loop
+
+        background_tasks.add_task(drain_processing_loop, tenant_id)
+        for user_file in user_files:
+            logger.info(f"Queued in-process processing for user_file_id={user_file.id}")
+    else:
+        from onyx.background.celery.versioned_apps.client import app as client_app
+
+        for user_file in user_files:
+            task = client_app.send_task(
+                OnyxCeleryTask.PROCESS_SINGLE_USER_FILE,
+                kwargs={"user_file_id": user_file.id, "tenant_id": tenant_id},
+                queue=OnyxCeleryQueues.USER_FILE_PROCESSING,
+                priority=OnyxCeleryPriority.HIGH,
+            )
+            logger.info(
+                f"Triggered indexing for user_file_id={user_file.id} "
+                f"with task_id={task.id}"
+            )
 
     return CategorizedFilesResult(
         user_files=user_files,

backend/onyx/db/slack_channel_config.py

@@ -5,8 +5,6 @@ from sqlalchemy import select
 from sqlalchemy.orm import joinedload
 from sqlalchemy.orm import Session
 
-from onyx.configs.chat_configs import MAX_CHUNKS_FED_TO_CHAT
-from onyx.context.search.enums import RecencyBiasSetting
 from onyx.db.constants import DEFAULT_PERSONA_SLACK_CHANNEL_NAME
 from onyx.db.constants import SLACK_BOT_PERSONA_PREFIX
 from onyx.db.models import ChannelConfig
@@ -45,8 +43,6 @@ def create_slack_channel_persona(
     channel_name: str | None,
     document_set_ids: list[int],
     existing_persona_id: int | None = None,
-    num_chunks: float = MAX_CHUNKS_FED_TO_CHAT,
-    enable_auto_filters: bool = False,
 ) -> Persona:
     """NOTE: does not commit changes"""
 
@@ -73,17 +69,13 @@ def create_slack_channel_persona(
         system_prompt="",
         task_prompt="",
         datetime_aware=True,
-        num_chunks=num_chunks,
-        llm_relevance_filter=True,
-        llm_filter_extraction=enable_auto_filters,
-        recency_bias=RecencyBiasSetting.AUTO,
         tool_ids=[search_tool.id],
         document_set_ids=document_set_ids,
         llm_model_provider_override=None,
         llm_model_version_override=None,
         starter_messages=None,
         is_public=True,
-        is_default_persona=False,
+        featured=False,
         db_session=db_session,
         commit=False,
     )

backend/onyx/db/users.py

@@ -2,7 +2,6 @@ from collections.abc import Sequence
 from typing import Any
 from uuid import UUID
 
-from fastapi import HTTPException
 from fastapi_users.password import PasswordHelper
 from sqlalchemy import func
 from sqlalchemy import select
@@ -24,6 +23,8 @@ from onyx.db.models import Persona__User
 from onyx.db.models import SamlAccount
 from onyx.db.models import User
 from onyx.db.models import User__UserGroup
+from onyx.error_handling.error_codes import OnyxErrorCode
+from onyx.error_handling.exceptions import OnyxError
 from onyx.utils.variable_functionality import fetch_ee_implementation_or_noop
 
 
@@ -44,22 +45,22 @@ def validate_user_role_update(
     """
 
     if current_role == UserRole.SLACK_USER:
-        raise HTTPException(
-            status_code=400,
-            detail="To change a Slack User's role, they must first login to Onyx via the web app.",
+        raise OnyxError(
+            OnyxErrorCode.VALIDATION_ERROR,
+            "To change a Slack User's role, they must first login to Onyx via the web app.",
         )
 
     if current_role == UserRole.EXT_PERM_USER:
         # This shouldn't happen, but just in case
-        raise HTTPException(
-            status_code=400,
-            detail="To change an External Permissioned User's role, they must first login to Onyx via the web app.",
+        raise OnyxError(
+            OnyxErrorCode.VALIDATION_ERROR,
+            "To change an External Permissioned User's role, they must first login to Onyx via the web app.",
         )
 
     if current_role == UserRole.LIMITED:
-        raise HTTPException(
-            status_code=400,
-            detail="To change a Limited User's role, they must first login to Onyx via the web app.",
+        raise OnyxError(
+            OnyxErrorCode.VALIDATION_ERROR,
+            "To change a Limited User's role, they must first login to Onyx via the web app.",
         )
 
     if explicit_override:
@@ -67,40 +68,34 @@ def validate_user_role_update(
 
     if requested_role == UserRole.CURATOR:
         # This shouldn't happen, but just in case
-        raise HTTPException(
-            status_code=400,
-            detail="Curator role must be set via the User Group Menu",
+        raise OnyxError(
+            OnyxErrorCode.VALIDATION_ERROR,
+            "Curator role must be set via the User Group Menu",
         )
 
     if requested_role == UserRole.LIMITED:
         # This shouldn't happen, but just in case
-        raise HTTPException(
-            status_code=400,
-            detail=(
-                "A user cannot be set to a Limited User role. "
-                "This role is automatically assigned to users through certain endpoints in the API."
-            ),
+        raise OnyxError(
+            OnyxErrorCode.VALIDATION_ERROR,
+            "A user cannot be set to a Limited User role. "
+            "This role is automatically assigned to users through certain endpoints in the API.",
         )
 
     if requested_role == UserRole.SLACK_USER:
         # This shouldn't happen, but just in case
-        raise HTTPException(
-            status_code=400,
-            detail=(
-                "A user cannot be set to a Slack User role. "
-                "This role is automatically assigned to users who only use Onyx via Slack."
-            ),
+        raise OnyxError(
+            OnyxErrorCode.VALIDATION_ERROR,
+            "A user cannot be set to a Slack User role. "
+            "This role is automatically assigned to users who only use Onyx via Slack.",
         )
 
     if requested_role == UserRole.EXT_PERM_USER:
         # This shouldn't happen, but just in case
-        raise HTTPException(
-            status_code=400,
-            detail=(
-                "A user cannot be set to an External Permissioned User role. "
-                "This role is automatically assigned to users who have been "
-                "pulled in to the system via an external permissions system."
-            ),
+        raise OnyxError(
+            OnyxErrorCode.VALIDATION_ERROR,
+            "A user cannot be set to an External Permissioned User role. "
+            "This role is automatically assigned to users who have been "
+            "pulled in to the system via an external permissions system.",
         )
 
 

backend/onyx/document_index/opensearch/opensearch_document_index.py

@@ -6,7 +6,6 @@ import httpx
 from opensearchpy import NotFoundError
 
 from onyx.access.models import DocumentAccess
-from onyx.configs.app_configs import USING_AWS_MANAGED_OPENSEARCH
 from onyx.configs.app_configs import VERIFY_CREATE_OPENSEARCH_INDEX_ON_INIT_MT
 from onyx.configs.chat_configs import NUM_RETURNED_HITS
 from onyx.configs.chat_configs import TITLE_CONTENT_RATIO
@@ -563,12 +562,7 @@ class OpenSearchDocumentIndex(DocumentIndex):
         )
 
         if not self._client.index_exists():
-            if USING_AWS_MANAGED_OPENSEARCH:
-                index_settings = (
-                    DocumentSchema.get_index_settings_for_aws_managed_opensearch()
-                )
-            else:
-                index_settings = DocumentSchema.get_index_settings()
+            index_settings = DocumentSchema.get_index_settings_based_on_environment()
             self._client.create_index(
                 mappings=expected_mappings,
                 settings=index_settings,

backend/onyx/document_index/opensearch/schema.py

@@ -12,6 +12,7 @@ from pydantic import model_validator
 from pydantic import SerializerFunctionWrapHandler
 
 from onyx.configs.app_configs import OPENSEARCH_TEXT_ANALYZER
+from onyx.configs.app_configs import USING_AWS_MANAGED_OPENSEARCH
 from onyx.document_index.interfaces_new import TenantState
 from onyx.document_index.opensearch.constants import DEFAULT_MAX_CHUNK_SIZE
 from onyx.document_index.opensearch.constants import EF_CONSTRUCTION
@@ -525,7 +526,7 @@ class DocumentSchema:
         }
 
     @staticmethod
-    def get_index_settings_for_aws_managed_opensearch() -> dict[str, Any]:
+    def get_index_settings_for_aws_managed_opensearch_st_dev() -> dict[str, Any]:
         """
         Settings for AWS-managed OpenSearch.
 
@@ -546,3 +547,41 @@ class DocumentSchema:
                 "knn.algo_param.ef_search": EF_SEARCH,
             }
         }
+
+    @staticmethod
+    def get_index_settings_for_aws_managed_opensearch_mt_cloud() -> dict[str, Any]:
+        """
+        Settings for AWS-managed OpenSearch in multi-tenant cloud.
+
+        324 shards very roughly targets a storage load of ~30Gb per shard, which
+        according to AWS OpenSearch documentation is within a good target range.
+
+        As documented above we need 2 replicas for a total of 3 copies of the
+        data because the cluster is configured with 3-AZ awareness.
+        """
+        return {
+            "index": {
+                "number_of_shards": 324,
+                "number_of_replicas": 2,
+                # Required for vector search.
+                "knn": True,
+                "knn.algo_param.ef_search": EF_SEARCH,
+            }
+        }
+
+    @staticmethod
+    def get_index_settings_based_on_environment() -> dict[str, Any]:
+        """
+        Returns the index settings based on the environment.
+        """
+        if USING_AWS_MANAGED_OPENSEARCH:
+            if MULTI_TENANT:
+                return (
+                    DocumentSchema.get_index_settings_for_aws_managed_opensearch_mt_cloud()
+                )
+            else:
+                return (
+                    DocumentSchema.get_index_settings_for_aws_managed_opensearch_st_dev()
+                )
+        else:
+            return DocumentSchema.get_index_settings()

backend/onyx/document_index/vespa/chunk_retrieval.py

@@ -1,5 +1,6 @@
 import json
 import string
+import time
 from collections.abc import Callable
 from collections.abc import Mapping
 from datetime import datetime
@@ -18,6 +19,7 @@ from onyx.background.celery.tasks.opensearch_migration.transformer import (
 )
 from onyx.configs.app_configs import LOG_VESPA_TIMING_INFORMATION
 from onyx.configs.app_configs import VESPA_LANGUAGE_OVERRIDE
+from onyx.configs.app_configs import VESPA_MIGRATION_REQUEST_TIMEOUT_S
 from onyx.context.search.models import IndexFilters
 from onyx.context.search.models import InferenceChunkUncleaned
 from onyx.document_index.interfaces import VespaChunkRequest
@@ -338,12 +340,18 @@ def get_all_chunks_paginated(
             params["continuation"] = continuation_token
 
         response: httpx.Response | None = None
+        start_time = time.monotonic()
         try:
-            with get_vespa_http_client() as http_client:
+            with get_vespa_http_client(
+                timeout=VESPA_MIGRATION_REQUEST_TIMEOUT_S
+            ) as http_client:
                 response = http_client.get(url, params=params)
                 response.raise_for_status()
         except httpx.HTTPError as e:
-            error_base = f"Failed to get chunks from Vespa slice {slice_id} with continuation token {continuation_token}."
+            error_base = (
+                f"Failed to get chunks from Vespa slice {slice_id} with continuation token "
+                f"{continuation_token} in {time.monotonic() - start_time:.3f} seconds."
+            )
             logger.exception(
                 f"Request URL: {e.request.url}\n"
                 f"Request Headers: {e.request.headers}\n"

backend/onyx/document_index/vespa/shared_utils/utils.py

@@ -52,7 +52,9 @@ def replace_invalid_doc_id_characters(text: str) -> str:
     return text.replace("'", "_")
 
 
-def get_vespa_http_client(no_timeout: bool = False, http2: bool = True) -> httpx.Client:
+def get_vespa_http_client(
+    no_timeout: bool = False, http2: bool = True, timeout: int | None = None
+) -> httpx.Client:
     """
     Configures and returns an HTTP client for communicating with Vespa,
     including authentication if needed.
@@ -64,7 +66,7 @@ def get_vespa_http_client(no_timeout: bool = False, http2: bool = True) -> httpx
             else None
         ),
         verify=False if not MANAGED_VESPA else True,
-        timeout=None if no_timeout else VESPA_REQUEST_TIMEOUT,
+        timeout=None if no_timeout else (timeout or VESPA_REQUEST_TIMEOUT),
         http2=http2,
     )
 

backend/onyx/error_handling/error_codes.py

@@ -0,0 +1,101 @@
+"""
+Standardized error codes for the Onyx backend.
+
+Usage:
+    from onyx.error_handling.error_codes import OnyxErrorCode
+    from onyx.error_handling.exceptions import OnyxError
+
+    raise OnyxError(OnyxErrorCode.UNAUTHENTICATED, "Token expired")
+"""
+
+from enum import Enum
+
+
+class OnyxErrorCode(Enum):
+    """
+    Each member is a tuple of (error_code_string, http_status_code).
+
+    The error_code_string is a stable, machine-readable identifier that
+    API consumers can match on. The http_status_code is the default HTTP
+    status to return.
+    """
+
+    # ------------------------------------------------------------------
+    # Authentication (401)
+    # ------------------------------------------------------------------
+    UNAUTHENTICATED = ("UNAUTHENTICATED", 401)
+    INVALID_TOKEN = ("INVALID_TOKEN", 401)
+    TOKEN_EXPIRED = ("TOKEN_EXPIRED", 401)
+    CSRF_FAILURE = ("CSRF_FAILURE", 403)
+
+    # ------------------------------------------------------------------
+    # Authorization (403)
+    # ------------------------------------------------------------------
+    UNAUTHORIZED = ("UNAUTHORIZED", 403)
+    INSUFFICIENT_PERMISSIONS = ("INSUFFICIENT_PERMISSIONS", 403)
+    ADMIN_ONLY = ("ADMIN_ONLY", 403)
+    EE_REQUIRED = ("EE_REQUIRED", 403)
+
+    # ------------------------------------------------------------------
+    # Validation / Bad Request (400)
+    # ------------------------------------------------------------------
+    VALIDATION_ERROR = ("VALIDATION_ERROR", 400)
+    INVALID_INPUT = ("INVALID_INPUT", 400)
+    MISSING_REQUIRED_FIELD = ("MISSING_REQUIRED_FIELD", 400)
+
+    # ------------------------------------------------------------------
+    # Not Found (404)
+    # ------------------------------------------------------------------
+    NOT_FOUND = ("NOT_FOUND", 404)
+    CONNECTOR_NOT_FOUND = ("CONNECTOR_NOT_FOUND", 404)
+    CREDENTIAL_NOT_FOUND = ("CREDENTIAL_NOT_FOUND", 404)
+    PERSONA_NOT_FOUND = ("PERSONA_NOT_FOUND", 404)
+    DOCUMENT_NOT_FOUND = ("DOCUMENT_NOT_FOUND", 404)
+    SESSION_NOT_FOUND = ("SESSION_NOT_FOUND", 404)
+    USER_NOT_FOUND = ("USER_NOT_FOUND", 404)
+
+    # ------------------------------------------------------------------
+    # Conflict (409)
+    # ------------------------------------------------------------------
+    CONFLICT = ("CONFLICT", 409)
+    DUPLICATE_RESOURCE = ("DUPLICATE_RESOURCE", 409)
+
+    # ------------------------------------------------------------------
+    # Rate Limiting / Quotas (429 / 402)
+    # ------------------------------------------------------------------
+    RATE_LIMITED = ("RATE_LIMITED", 429)
+    SEAT_LIMIT_EXCEEDED = ("SEAT_LIMIT_EXCEEDED", 402)
+
+    # ------------------------------------------------------------------
+    # Connector / Credential Errors (400-range)
+    # ------------------------------------------------------------------
+    CONNECTOR_VALIDATION_FAILED = ("CONNECTOR_VALIDATION_FAILED", 400)
+    CREDENTIAL_INVALID = ("CREDENTIAL_INVALID", 400)
+    CREDENTIAL_EXPIRED = ("CREDENTIAL_EXPIRED", 401)
+
+    # ------------------------------------------------------------------
+    # Server Errors (5xx)
+    # ------------------------------------------------------------------
+    INTERNAL_ERROR = ("INTERNAL_ERROR", 500)
+    NOT_IMPLEMENTED = ("NOT_IMPLEMENTED", 501)
+    SERVICE_UNAVAILABLE = ("SERVICE_UNAVAILABLE", 503)
+    BAD_GATEWAY = ("BAD_GATEWAY", 502)
+    LLM_PROVIDER_ERROR = ("LLM_PROVIDER_ERROR", 502)
+    GATEWAY_TIMEOUT = ("GATEWAY_TIMEOUT", 504)
+
+    def __init__(self, code: str, status_code: int) -> None:
+        self.code = code
+        self.status_code = status_code
+
+    def detail(self, message: str | None = None) -> dict[str, str]:
+        """Build a structured error detail dict.
+
+        Returns a dict like:
+            {"error_code": "UNAUTHENTICATED", "message": "Token expired"}
+
+        If no message is supplied, the error code itself is used as the message.
+        """
+        return {
+            "error_code": self.code,
+            "message": message or self.code,
+        }

backend/onyx/error_handling/exceptions.py

@@ -0,0 +1,82 @@
+"""OnyxError — the single exception type for all Onyx business errors.
+
+Raise ``OnyxError`` instead of ``HTTPException`` in business code.  A global
+FastAPI exception handler (registered via ``register_onyx_exception_handlers``)
+converts it into a JSON response with the standard
+``{"error_code": "...", "message": "..."}`` shape.
+
+Usage::
+
+    from onyx.error_handling.error_codes import OnyxErrorCode
+    from onyx.error_handling.exceptions import OnyxError
+
+    raise OnyxError(OnyxErrorCode.NOT_FOUND, "Session not found")
+
+For upstream errors with a dynamic HTTP status (e.g. billing service),
+use ``status_code_override``::
+
+    raise OnyxError(
+        OnyxErrorCode.BAD_GATEWAY,
+        detail,
+        status_code_override=upstream_status,
+    )
+"""
+
+from fastapi import FastAPI
+from fastapi import Request
+from fastapi.responses import JSONResponse
+
+from onyx.error_handling.error_codes import OnyxErrorCode
+from onyx.utils.logger import setup_logger
+
+logger = setup_logger()
+
+
+class OnyxError(Exception):
+    """Structured error that maps to a specific ``OnyxErrorCode``.
+
+    Attributes:
+        error_code: The ``OnyxErrorCode`` enum member.
+        message: Human-readable message (defaults to the error code string).
+        status_code: HTTP status — either overridden or from the error code.
+    """
+
+    def __init__(
+        self,
+        error_code: OnyxErrorCode,
+        message: str | None = None,
+        *,
+        status_code_override: int | None = None,
+    ) -> None:
+        self.error_code = error_code
+        self.message = message or error_code.code
+        self._status_code_override = status_code_override
+        super().__init__(self.message)
+
+    @property
+    def status_code(self) -> int:
+        return self._status_code_override or self.error_code.status_code
+
+
+def register_onyx_exception_handlers(app: FastAPI) -> None:
+    """Register a global handler that converts ``OnyxError`` to JSON responses.
+
+    Must be called *after* the app is created but *before* it starts serving.
+    The handler logs at WARNING for 4xx and ERROR for 5xx.
+    """
+
+    @app.exception_handler(OnyxError)
+    async def _handle_onyx_error(
+        request: Request,  # noqa: ARG001
+        exc: OnyxError,
+    ) -> JSONResponse:
+        status_code = exc.status_code
+        if status_code >= 500:
+            logger.error(f"OnyxError {exc.error_code.code}: {exc.message}")
+        elif status_code >= 400:
+            logger.warning(f"OnyxError {exc.error_code.code}: {exc.message}")
+
+        return JSONResponse(
+            status_code=status_code,
+            content=exc.error_code.detail(exc.message),
+        )

backend/onyx/federated_connectors/oauth_utils.py

@@ -4,39 +4,33 @@ import base64
 import json
 import uuid
 from typing import Any
-from typing import cast
-from typing import Dict
-from typing import Optional
 
+from onyx.cache.factory import get_cache_backend
 from onyx.configs.app_configs import WEB_DOMAIN
-from onyx.redis.redis_pool import get_redis_client
 from onyx.utils.logger import setup_logger
 
 logger = setup_logger()
 
-# Redis key prefix for OAuth state
 OAUTH_STATE_PREFIX = "federated_oauth"
-# Default TTL for OAuth state (5 minutes)
-OAUTH_STATE_TTL = 300
+OAUTH_STATE_TTL = 300  # 5 minutes
 
 
 class OAuthSession:
-    """Represents an OAuth session stored in Redis."""
+    """Represents an OAuth session stored in the cache backend."""
 
     def __init__(
         self,
         federated_connector_id: int,
         user_id: str,
-        redirect_uri: Optional[str] = None,
-        additional_data: Optional[Dict[str, Any]] = None,
+        redirect_uri: str | None = None,
+        additional_data: dict[str, Any] | None = None,
     ):
         self.federated_connector_id = federated_connector_id
         self.user_id = user_id
         self.redirect_uri = redirect_uri
         self.additional_data = additional_data or {}
 
-    def to_dict(self) -> Dict[str, Any]:
-        """Convert to dictionary for Redis storage."""
+    def to_dict(self) -> dict[str, Any]:
         return {
             "federated_connector_id": self.federated_connector_id,
             "user_id": self.user_id,
@@ -45,8 +39,7 @@ class OAuthSession:
         }
 
     @classmethod
-    def from_dict(cls, data: Dict[str, Any]) -> "OAuthSession":
-        """Create from dictionary retrieved from Redis."""
+    def from_dict(cls, data: dict[str, Any]) -> "OAuthSession":
         return cls(
             federated_connector_id=data["federated_connector_id"],
             user_id=data["user_id"],
@@ -58,31 +51,27 @@ class OAuthSession:
 def generate_oauth_state(
     federated_connector_id: int,
     user_id: str,
-    redirect_uri: Optional[str] = None,
-    additional_data: Optional[Dict[str, Any]] = None,
+    redirect_uri: str | None = None,
+    additional_data: dict[str, Any] | None = None,
     ttl: int = OAUTH_STATE_TTL,
 ) -> str:
     """
-    Generate a secure state parameter and store session data in Redis.
+    Generate a secure state parameter and store session data in the cache backend.
 
     Args:
         federated_connector_id: ID of the federated connector
         user_id: ID of the user initiating OAuth
         redirect_uri: Optional redirect URI after OAuth completion
         additional_data: Any additional data to store with the session
-        ttl: Time-to-live in seconds for the Redis key
+        ttl: Time-to-live in seconds for the cache key
 
     Returns:
         Base64-encoded state parameter
     """
     # Generate a random UUID for the state
     state_uuid = uuid.uuid4()
+    state_b64 = base64.urlsafe_b64encode(state_uuid.bytes).decode("utf-8").rstrip("=")
 
-    # Convert UUID to base64 for URL-safe state parameter
-    state_bytes = state_uuid.bytes
-    state_b64 = base64.urlsafe_b64encode(state_bytes).decode("utf-8").rstrip("=")
-
-    # Create session object
     session = OAuthSession(
         federated_connector_id=federated_connector_id,
         user_id=user_id,
@@ -90,15 +79,9 @@ def generate_oauth_state(
         additional_data=additional_data,
     )
 
-    # Store in Redis with TTL
-    redis_client = get_redis_client()
-    redis_key = f"{OAUTH_STATE_PREFIX}:{state_uuid}"
-
-    redis_client.set(
-        redis_key,
-        json.dumps(session.to_dict()),
-        ex=ttl,
-    )
+    cache = get_cache_backend()
+    cache_key = f"{OAUTH_STATE_PREFIX}:{state_uuid}"
+    cache.set(cache_key, json.dumps(session.to_dict()), ex=ttl)
 
     logger.info(
         f"Generated OAuth state for federated_connector_id={federated_connector_id}, "
@@ -125,18 +108,15 @@ def verify_oauth_state(state: str) -> OAuthSession:
     state_bytes = base64.urlsafe_b64decode(padded_state)
     state_uuid = uuid.UUID(bytes=state_bytes)
 
-    # Look up in Redis
-    redis_client = get_redis_client()
-    redis_key = f"{OAUTH_STATE_PREFIX}:{state_uuid}"
+    cache = get_cache_backend()
+    cache_key = f"{OAUTH_STATE_PREFIX}:{state_uuid}"
 
-    session_data = cast(bytes, redis_client.get(redis_key))
+    session_data = cache.get(cache_key)
     if not session_data:
-        raise ValueError(f"OAuth state not found in Redis: {state}")
+        raise ValueError(f"OAuth state not found: {state}")
 
-    # Delete the key after retrieval (one-time use)
-    redis_client.delete(redis_key)
+    cache.delete(cache_key)
 
-    # Parse and return session
     session_dict = json.loads(session_data)
     return OAuthSession.from_dict(session_dict)
 

backend/onyx/key_value_store/store.py

@@ -1,13 +1,11 @@
 import json
 from typing import cast
 
-from redis.client import Redis
-
+from onyx.cache.interface import CacheBackend
 from onyx.db.engine.sql_engine import get_session_with_current_tenant
 from onyx.db.models import KVStore
 from onyx.key_value_store.interface import KeyValueStore
 from onyx.key_value_store.interface import KvKeyNotFoundError
-from onyx.redis.redis_pool import get_redis_client
 from onyx.utils.logger import setup_logger
 from onyx.utils.special_types import JSON_ro
 
@@ -20,22 +18,27 @@ KV_REDIS_KEY_EXPIRATION = 60 * 60 * 24  # 1 Day
 
 
 class PgRedisKVStore(KeyValueStore):
-    def __init__(self, redis_client: Redis | None = None) -> None:
-        # If no redis_client is provided, fall back to the context var
-        if redis_client is not None:
-            self.redis_client = redis_client
-        else:
-            self.redis_client = get_redis_client()
+    def __init__(self, cache: CacheBackend | None = None) -> None:
+        self._cache = cache
+
+    def _get_cache(self) -> CacheBackend:
+        if self._cache is None:
+            from onyx.cache.factory import get_cache_backend
+
+            self._cache = get_cache_backend()
+        return self._cache
 
     def store(self, key: str, val: JSON_ro, encrypt: bool = False) -> None:
-        # Not encrypted in Redis, but encrypted in Postgres
+        # Not encrypted in Cache backend (typically Redis), but encrypted in Postgres
         try:
-            self.redis_client.set(
+            self._get_cache().set(
                 REDIS_KEY_PREFIX + key, json.dumps(val), ex=KV_REDIS_KEY_EXPIRATION
             )
         except Exception as e:
-            # Fallback gracefully to Postgres if Redis fails
-            logger.error(f"Failed to set value in Redis for key '{key}': {str(e)}")
+            # Fallback gracefully to Postgres if Cache backend fails
+            logger.error(
+                f"Failed to set value in Cache backend for key '{key}': {str(e)}"
+            )
 
         encrypted_val = val if encrypt else None
         plain_val = val if not encrypt else None
@@ -53,16 +56,12 @@ class PgRedisKVStore(KeyValueStore):
     def load(self, key: str, refresh_cache: bool = False) -> JSON_ro:
         if not refresh_cache:
             try:
-                redis_value = self.redis_client.get(REDIS_KEY_PREFIX + key)
-                if redis_value:
-                    if not isinstance(redis_value, bytes):
-                        raise ValueError(
-                            f"Redis value for key '{key}' is not a bytes object"
-                        )
-                    return json.loads(redis_value.decode("utf-8"))
+                cached = self._get_cache().get(REDIS_KEY_PREFIX + key)
+                if cached is not None:
+                    return json.loads(cached.decode("utf-8"))
             except Exception as e:
                 logger.error(
-                    f"Failed to get value from Redis for key '{key}': {str(e)}"
+                    f"Failed to get value from cache for key '{key}': {str(e)}"
                 )
 
         with get_session_with_current_tenant() as db_session:
@@ -79,21 +78,21 @@ class PgRedisKVStore(KeyValueStore):
                 value = None
 
             try:
-                self.redis_client.set(
+                self._get_cache().set(
                     REDIS_KEY_PREFIX + key,
                     json.dumps(value),
                     ex=KV_REDIS_KEY_EXPIRATION,
                 )
             except Exception as e:
-                logger.error(f"Failed to set value in Redis for key '{key}': {str(e)}")
+                logger.error(f"Failed to set value in cache for key '{key}': {str(e)}")
 
             return cast(JSON_ro, value)
 
     def delete(self, key: str) -> None:
         try:
-            self.redis_client.delete(REDIS_KEY_PREFIX + key)
+            self._get_cache().delete(REDIS_KEY_PREFIX + key)
         except Exception as e:
-            logger.error(f"Failed to delete value from Redis for key '{key}': {str(e)}")
+            logger.error(f"Failed to delete value from cache for key '{key}': {str(e)}")
 
         with get_session_with_current_tenant() as db_session:
             result = db_session.query(KVStore).filter_by(key=key).delete()

backend/onyx/llm/litellm_singleton/monkey_patches.py

@@ -67,6 +67,18 @@ Status checked against LiteLLM v1.81.6-nightly (2026-02-02):
    STATUS: STILL NEEDED - litellm_core_utils/litellm_logging.py lines 3185-3199 set
            usage as a dict with chat completion format instead of keeping it as
            ResponseAPIUsage. Our patch creates a deep copy before modification.
+
+7. Responses API metadata=None TypeError (_patch_responses_metadata_none):
+   - LiteLLM's @client decorator wrapper in utils.py uses kwargs.get("metadata", {})
+     to check for router calls, but when metadata is explicitly None (key exists with
+     value None), the default {} is not used
+   - This causes "argument of type 'NoneType' is not iterable" TypeError which swallows
+     the real exception (e.g. AuthenticationError for wrong API key)
+   - Surfaces as: APIConnectionError: OpenAIException - argument of type 'NoneType' is
+     not iterable
+   STATUS: STILL NEEDED - litellm/utils.py wrapper function (line 1721) does not guard
+           against metadata being explicitly None. Triggered when Responses API bridge
+           passes **litellm_params containing metadata=None.
 """
 
 import time
@@ -725,6 +737,44 @@ def _patch_logging_assembled_streaming_response() -> None:
     LiteLLMLoggingObj._get_assembled_streaming_response = _patched_get_assembled_streaming_response  # type: ignore[method-assign]
 
 
+def _patch_responses_metadata_none() -> None:
+    """
+    Patches litellm.responses to normalize metadata=None to metadata={} in kwargs.
+
+    LiteLLM's @client decorator wrapper in utils.py (line 1721) does:
+        _is_litellm_router_call = "model_group" in kwargs.get("metadata", {})
+    When metadata is explicitly None in kwargs, kwargs.get("metadata", {}) returns
+    None (the key exists, so the default is not used), causing:
+        TypeError: argument of type 'NoneType' is not iterable
+
+    This swallows the real exception (e.g. AuthenticationError) and surfaces as:
+        APIConnectionError: OpenAIException - argument of type 'NoneType' is not iterable
+
+    This happens when the Responses API bridge calls litellm.responses() with
+    **litellm_params which may contain metadata=None.
+
+    STATUS: STILL NEEDED - litellm/utils.py wrapper function uses kwargs.get("metadata", {})
+            which does not guard against metadata being explicitly None. Same pattern exists
+            on line 1407 for async path.
+    """
+    import litellm as _litellm
+    from functools import wraps
+
+    original_responses = _litellm.responses
+
+    if getattr(original_responses, "_metadata_patched", False):
+        return
+
+    @wraps(original_responses)
+    def _patched_responses(*args: Any, **kwargs: Any) -> Any:
+        if kwargs.get("metadata") is None:
+            kwargs["metadata"] = {}
+        return original_responses(*args, **kwargs)
+
+    _patched_responses._metadata_patched = True  # type: ignore[attr-defined]
+    _litellm.responses = _patched_responses
+
+
 def apply_monkey_patches() -> None:
     """
     Apply all necessary monkey patches to LiteLLM for compatibility.
@@ -736,6 +786,7 @@ def apply_monkey_patches() -> None:
     - Patching AzureOpenAIResponsesAPIConfig.should_fake_stream to enable native streaming
     - Patching ResponsesAPIResponse.model_construct to fix usage format in all code paths
     - Patching LiteLLMLoggingObj._get_assembled_streaming_response to avoid mutating original response
+    - Patching litellm.responses to fix metadata=None causing TypeError in error handling
     """
     _patch_ollama_chunk_parser()
     _patch_openai_responses_parallel_tool_calls()
@@ -743,3 +794,4 @@ def apply_monkey_patches() -> None:
     _patch_azure_responses_should_fake_stream()
     _patch_responses_api_usage_format()
     _patch_logging_assembled_streaming_response()
+    _patch_responses_metadata_none()

backend/onyx/llm/well_known_providers/auto_update_service.py

@@ -13,44 +13,38 @@ from datetime import datetime
 import httpx
 from sqlalchemy.orm import Session
 
+from onyx.cache.factory import get_cache_backend
 from onyx.configs.app_configs import AUTO_LLM_CONFIG_URL
 from onyx.db.llm import fetch_auto_mode_providers
 from onyx.db.llm import sync_auto_mode_models
 from onyx.llm.well_known_providers.auto_update_models import LLMRecommendations
-from onyx.redis.redis_pool import get_redis_client
 from onyx.utils.logger import setup_logger
 
 logger = setup_logger()
 
-# Redis key for caching the last updated timestamp (per-tenant)
-_REDIS_KEY_LAST_UPDATED_AT = "auto_llm_update:last_updated_at"
+_CACHE_KEY_LAST_UPDATED_AT = "auto_llm_update:last_updated_at"
+_CACHE_TTL_SECONDS = 60 * 60 * 24  # 24 hours
 
 
 def _get_cached_last_updated_at() -> datetime | None:
-    """Get the cached last_updated_at timestamp from Redis."""
     try:
-        redis_client = get_redis_client()
-        value = redis_client.get(_REDIS_KEY_LAST_UPDATED_AT)
-        if value and isinstance(value, bytes):
-            # Value is bytes, decode to string then parse as ISO format
+        value = get_cache_backend().get(_CACHE_KEY_LAST_UPDATED_AT)
+        if value is not None:
             return datetime.fromisoformat(value.decode("utf-8"))
     except Exception as e:
-        logger.warning(f"Failed to get cached last_updated_at from Redis: {e}")
+        logger.warning(f"Failed to get cached last_updated_at: {e}")
     return None
 
 
 def _set_cached_last_updated_at(updated_at: datetime) -> None:
-    """Set the cached last_updated_at timestamp in Redis."""
     try:
-        redis_client = get_redis_client()
-        # Store as ISO format string, with 24 hour expiration
-        redis_client.set(
-            _REDIS_KEY_LAST_UPDATED_AT,
+        get_cache_backend().set(
+            _CACHE_KEY_LAST_UPDATED_AT,
             updated_at.isoformat(),
-            ex=60 * 60 * 24,  # 24 hours
+            ex=_CACHE_TTL_SECONDS,
         )
     except Exception as e:
-        logger.warning(f"Failed to set cached last_updated_at in Redis: {e}")
+        logger.warning(f"Failed to set cached last_updated_at: {e}")
 
 
 def fetch_llm_recommendations_from_github(
@@ -148,9 +142,8 @@ def sync_llm_models_from_github(
 
 
 def reset_cache() -> None:
-    """Reset the cache timestamp in Redis. Useful for testing."""
+    """Reset the cache timestamp. Useful for testing."""
     try:
-        redis_client = get_redis_client()
-        redis_client.delete(_REDIS_KEY_LAST_UPDATED_AT)
+        get_cache_backend().delete(_CACHE_KEY_LAST_UPDATED_AT)
     except Exception as e:
-        logger.warning(f"Failed to reset cache in Redis: {e}")
+        logger.warning(f"Failed to reset cache: {e}")

backend/onyx/main.py

@@ -32,11 +32,14 @@ from onyx.auth.schemas import UserUpdate
 from onyx.auth.users import auth_backend
 from onyx.auth.users import create_onyx_oauth_router
 from onyx.auth.users import fastapi_users
+from onyx.cache.interface import CacheBackendType
 from onyx.configs.app_configs import APP_API_PREFIX
 from onyx.configs.app_configs import APP_HOST
 from onyx.configs.app_configs import APP_PORT
 from onyx.configs.app_configs import AUTH_RATE_LIMITING_ENABLED
 from onyx.configs.app_configs import AUTH_TYPE
+from onyx.configs.app_configs import CACHE_BACKEND
+from onyx.configs.app_configs import DISABLE_VECTOR_DB
 from onyx.configs.app_configs import LOG_ENDPOINT_LATENCY
 from onyx.configs.app_configs import OAUTH_CLIENT_ID
 from onyx.configs.app_configs import OAUTH_CLIENT_SECRET
@@ -56,6 +59,7 @@ from onyx.db.engine.async_sql_engine import get_sqlalchemy_async_engine
 from onyx.db.engine.connection_warmup import warm_up_connections
 from onyx.db.engine.sql_engine import get_session_with_current_tenant
 from onyx.db.engine.sql_engine import SqlEngine
+from onyx.error_handling.exceptions import register_onyx_exception_handlers
 from onyx.file_store.file_store import get_default_file_store
 from onyx.server.api_key.api import router as api_key_router
 from onyx.server.auth_check import check_router_auth
@@ -254,8 +258,53 @@ def include_auth_router_with_prefix(
     )
 
 
+def validate_cache_backend_settings() -> None:
+    """Validate that CACHE_BACKEND=postgres is only used with DISABLE_VECTOR_DB.
+
+    The Postgres cache backend eliminates the Redis dependency, but only works
+    when Celery is not running (which requires DISABLE_VECTOR_DB=true).
+    """
+    if CACHE_BACKEND == CacheBackendType.POSTGRES and not DISABLE_VECTOR_DB:
+        raise RuntimeError(
+            "CACHE_BACKEND=postgres requires DISABLE_VECTOR_DB=true. "
+            "The Postgres cache backend is only supported in no-vector-DB "
+            "deployments where Celery is replaced by the in-process task runner."
+        )
+
+
+def validate_no_vector_db_settings() -> None:
+    """Validate that DISABLE_VECTOR_DB is not combined with incompatible settings.
+
+    Raises RuntimeError if DISABLE_VECTOR_DB is set alongside MULTI_TENANT or ENABLE_CRAFT,
+    since these modes require infrastructure that is removed in no-vector-DB deployments.
+    """
+    if not DISABLE_VECTOR_DB:
+        return
+
+    if MULTI_TENANT:
+        raise RuntimeError(
+            "DISABLE_VECTOR_DB cannot be used with MULTI_TENANT. "
+            "Multi-tenant deployments require the vector database for "
+            "per-tenant document indexing and search. Run in single-tenant "
+            "mode when disabling the vector database."
+        )
+
+    from onyx.server.features.build.configs import ENABLE_CRAFT
+
+    if ENABLE_CRAFT:
+        raise RuntimeError(
+            "DISABLE_VECTOR_DB cannot be used with ENABLE_CRAFT. "
+            "Onyx Craft requires background workers for sandbox lifecycle "
+            "management, which are removed in no-vector-DB deployments. "
+            "Disable Craft (ENABLE_CRAFT=false) when disabling the vector database."
+        )
+
+
 @asynccontextmanager
 async def lifespan(app: FastAPI) -> AsyncGenerator[None, None]:  # noqa: ARG001
+    validate_no_vector_db_settings()
+    validate_cache_backend_settings()
+
     # Set recursion limit
     if SYSTEM_RECURSION_LIMIT is not None:
         sys.setrecursionlimit(SYSTEM_RECURSION_LIMIT)
@@ -324,8 +373,20 @@ async def lifespan(app: FastAPI) -> AsyncGenerator[None, None]:  # noqa: ARG001
     if AUTH_RATE_LIMITING_ENABLED:
         await setup_auth_limiter()
 
+    if DISABLE_VECTOR_DB:
+        from onyx.background.periodic_poller import recover_stuck_user_files
+        from onyx.background.periodic_poller import start_periodic_poller
+
+        recover_stuck_user_files(POSTGRES_DEFAULT_SCHEMA)
+        start_periodic_poller(POSTGRES_DEFAULT_SCHEMA)
+
     yield
 
+    if DISABLE_VECTOR_DB:
+        from onyx.background.periodic_poller import stop_periodic_poller
+
+        stop_periodic_poller()
+
     SqlEngine.reset_engine()
 
     if AUTH_RATE_LIMITING_ENABLED:
@@ -384,6 +445,8 @@ def get_application(lifespan_override: Lifespan | None = None) -> FastAPI:
         status.HTTP_500_INTERNAL_SERVER_ERROR, log_http_error
     )
 
+    register_onyx_exception_handlers(application)
+
     include_router_with_global_prefix_prepended(application, password_router)
     include_router_with_global_prefix_prepended(application, chat_router)
     include_router_with_global_prefix_prepended(application, query_router)

backend/onyx/onyxbot/slack/formatting.py

@@ -130,7 +130,7 @@ def format_slack_message(message: str | None) -> str:
     message = _transform_outside_code_blocks(message, _sanitize_html)
     message = _convert_slack_links_to_markdown(message)
     normalized_message = _normalize_link_destinations(message)
-    md = create_markdown(renderer=SlackRenderer(), plugins=["strikethrough"])
+    md = create_markdown(renderer=SlackRenderer(), plugins=["strikethrough", "table"])
     result = md(normalized_message)
     # With HTMLRenderer, result is always str (not AST list)
     assert isinstance(result, str)
@@ -146,6 +146,11 @@ class SlackRenderer(HTMLRenderer):
 
     SPECIALS: dict[str, str] = {"&": "&amp;", "<": "&lt;", ">": "&gt;"}
 
+    def __init__(self) -> None:
+        super().__init__()
+        self._table_headers: list[str] = []
+        self._current_row_cells: list[str] = []
+
     def escape_special(self, text: str) -> str:
         for special, replacement in self.SPECIALS.items():
             text = text.replace(special, replacement)
@@ -218,5 +223,48 @@ class SlackRenderer(HTMLRenderer):
         # as literal &quot; text since Slack doesn't recognize that entity.
         return self.escape_special(text)
 
+    # -- Table rendering (converts markdown tables to vertical cards) --
+
+    def table_cell(
+        self, text: str, align: str | None = None, head: bool = False  # noqa: ARG002
+    ) -> str:
+        if head:
+            self._table_headers.append(text.strip())
+        else:
+            self._current_row_cells.append(text.strip())
+        return ""
+
+    def table_head(self, text: str) -> str:  # noqa: ARG002
+        self._current_row_cells = []
+        return ""
+
+    def table_row(self, text: str) -> str:  # noqa: ARG002
+        cells = self._current_row_cells
+        self._current_row_cells = []
+        # First column becomes the bold title, remaining columns are bulleted fields
+        lines: list[str] = []
+        if cells:
+            title = cells[0]
+            if title:
+                # Avoid double-wrapping if cell already contains bold markup
+                if title.startswith("*") and title.endswith("*") and len(title) > 1:
+                    lines.append(title)
+                else:
+                    lines.append(f"*{title}*")
+            for i, cell in enumerate(cells[1:], start=1):
+                if i < len(self._table_headers):
+                    lines.append(f"  • {self._table_headers[i]}: {cell}")
+                else:
+                    lines.append(f"  • {cell}")
+        return "\n".join(lines) + "\n\n"
+
+    def table_body(self, text: str) -> str:
+        return text
+
+    def table(self, text: str) -> str:
+        self._table_headers = []
+        self._current_row_cells = []
+        return text + "\n"
+
     def paragraph(self, text: str) -> str:
         return f"{text}\n\n"

backend/onyx/onyxbot/slack/handlers/handle_message.py

@@ -3,10 +3,12 @@ import datetime
 from slack_sdk import WebClient
 from slack_sdk.errors import SlackApiError
 
+from onyx.auth.schemas import UserRole
 from onyx.configs.onyxbot_configs import ONYX_BOT_FEEDBACK_REMINDER
 from onyx.configs.onyxbot_configs import ONYX_BOT_REACT_EMOJI
 from onyx.db.engine.sql_engine import get_session_with_current_tenant
 from onyx.db.models import SlackChannelConfig
+from onyx.db.user_preferences import activate_user
 from onyx.db.users import add_slack_user_if_not_exists
 from onyx.db.users import get_user_by_email
 from onyx.onyxbot.slack.blocks import get_feedback_reminder_blocks
@@ -243,6 +245,44 @@ def handle_message(
                     )
                     return False
 
+            elif (
+                not existing_user.is_active
+                and existing_user.role == UserRole.SLACK_USER
+            ):
+                check_seat_fn = fetch_ee_implementation_or_noop(
+                    "onyx.db.license",
+                    "check_seat_availability",
+                    None,
+                )
+                seat_result = check_seat_fn(db_session=db_session)
+                if seat_result is not None and not seat_result.available:
+                    logger.info(
+                        f"Blocked inactive Slack user {message_info.email}: "
+                        f"{seat_result.error_message}"
+                    )
+                    respond_in_thread_or_channel(
+                        client=client,
+                        channel=channel,
+                        thread_ts=message_info.msg_to_respond,
+                        text=(
+                            "We weren't able to respond because your organization "
+                            "has reached its user seat limit. Your account is "
+                            "currently deactivated and cannot be reactivated "
+                            "until more seats are available. Please contact "
+                            "your Onyx administrator."
+                        ),
+                    )
+                    return False
+
+                activate_user(existing_user, db_session)
+                invalidate_license_cache_fn = fetch_ee_implementation_or_noop(
+                    "onyx.db.license",
+                    "invalidate_license_cache",
+                    None,
+                )
+                invalidate_license_cache_fn()
+                logger.info(f"Reactivated inactive Slack user {message_info.email}")
+
             add_slack_user_if_not_exists(db_session, message_info.email)
 
         # first check if we need to respond with a standard answer

backend/onyx/server/documents/connector.py

@@ -92,6 +92,7 @@ from onyx.db.connector_credential_pair import get_connector_credential_pairs_for
 from onyx.db.connector_credential_pair import (
     get_connector_credential_pairs_for_user_parallel,
 )
+from onyx.db.connector_credential_pair import verify_user_has_access_to_cc_pair
 from onyx.db.credentials import cleanup_gmail_credentials
 from onyx.db.credentials import cleanup_google_drive_credentials
 from onyx.db.credentials import create_credential
@@ -572,6 +573,43 @@ def _normalize_file_names_for_backwards_compatibility(
     return file_names + file_locations[len(file_names) :]
 
 
+def _fetch_and_check_file_connector_cc_pair_permissions(
+    connector_id: int,
+    user: User,
+    db_session: Session,
+    require_editable: bool,
+) -> ConnectorCredentialPair:
+    cc_pair = fetch_connector_credential_pair_for_connector(db_session, connector_id)
+    if cc_pair is None:
+        raise HTTPException(
+            status_code=404,
+            detail="No Connector-Credential Pair found for this connector",
+        )
+
+    has_requested_access = verify_user_has_access_to_cc_pair(
+        cc_pair_id=cc_pair.id,
+        db_session=db_session,
+        user=user,
+        get_editable=require_editable,
+    )
+    if has_requested_access:
+        return cc_pair
+
+    # Special case: global curators should be able to manage files
+    # for public file connectors even when they are not the creator.
+    if (
+        require_editable
+        and user.role == UserRole.GLOBAL_CURATOR
+        and cc_pair.access_type == AccessType.PUBLIC
+    ):
+        return cc_pair
+
+    raise HTTPException(
+        status_code=403,
+        detail="Access denied. User cannot manage files for this connector.",
+    )
+
+
 @router.post("/admin/connector/file/upload", tags=PUBLIC_API_TAGS)
 def upload_files_api(
     files: list[UploadFile],
@@ -583,7 +621,7 @@ def upload_files_api(
 @router.get("/admin/connector/{connector_id}/files", tags=PUBLIC_API_TAGS)
 def list_connector_files(
     connector_id: int,
-    user: User = Depends(current_curator_or_admin_user),  # noqa: ARG001
+    user: User = Depends(current_curator_or_admin_user),
     db_session: Session = Depends(get_session),
 ) -> ConnectorFilesResponse:
     """List all files in a file connector."""
@@ -596,6 +634,13 @@ def list_connector_files(
             status_code=400, detail="This endpoint only works with file connectors"
         )
 
+    _ = _fetch_and_check_file_connector_cc_pair_permissions(
+        connector_id=connector_id,
+        user=user,
+        db_session=db_session,
+        require_editable=False,
+    )
+
     file_locations = connector.connector_specific_config.get("file_locations", [])
     file_names = connector.connector_specific_config.get("file_names", [])
 
@@ -645,7 +690,7 @@ def update_connector_files(
     connector_id: int,
     files: list[UploadFile] | None = File(None),
     file_ids_to_remove: str = Form("[]"),
-    user: User = Depends(current_curator_or_admin_user),  # noqa: ARG001
+    user: User = Depends(current_curator_or_admin_user),
     db_session: Session = Depends(get_session),
 ) -> FileUploadResponse:
     """
@@ -663,12 +708,13 @@ def update_connector_files(
         )
 
     # Get the connector-credential pair for indexing/pruning triggers
-    cc_pair = fetch_connector_credential_pair_for_connector(db_session, connector_id)
-    if cc_pair is None:
-        raise HTTPException(
-            status_code=404,
-            detail="No Connector-Credential Pair found for this connector",
-        )
+    # and validate user permissions for file management.
+    cc_pair = _fetch_and_check_file_connector_cc_pair_permissions(
+        connector_id=connector_id,
+        user=user,
+        db_session=db_session,
+        require_editable=True,
+    )
 
     # Parse file IDs to remove
     try:

backend/onyx/server/features/build/sandbox/kubernetes/docker/templates/outputs/web/package-lock.json

@@ -7424,9 +7424,9 @@
       }
     },
     "node_modules/hono": {
-      "version": "4.11.7",
-      "resolved": "https://registry.npmjs.org/hono/-/hono-4.11.7.tgz",
-      "integrity": "sha512-l7qMiNee7t82bH3SeyUCt9UF15EVmaBvsppY2zQtrbIhl/yzBTny+YUxsVjSjQ6gaqaeVtZmGocom8TzBlA4Yw==",
+      "version": "4.12.5",
+      "resolved": "https://registry.npmjs.org/hono/-/hono-4.12.5.tgz",
+      "integrity": "sha512-3qq+FUBtlTHhtYxbxheZgY8NIFnkkC/MR8u5TTsr7YZ3wixryQ3cCwn3iZbg8p8B88iDBBAYSfZDS75t8MN7Vg==",
       "license": "MIT",
       "engines": {
         "node": ">=16.9.0"

backend/onyx/server/features/build/sandbox/kubernetes/kubernetes_sandbox_manager.py

@@ -1133,7 +1133,8 @@ done
                 # Already deleted
                 service_deleted = True
             else:
-                logger.warning(f"Error deleting Service {service_name}: {e}")
+                logger.error(f"Error deleting Service {service_name}: {e}")
+                raise
 
         pod_deleted = False
         try:
@@ -1148,7 +1149,8 @@ done
                 # Already deleted
                 pod_deleted = True
             else:
-                logger.warning(f"Error deleting Pod {pod_name}: {e}")
+                logger.error(f"Error deleting Pod {pod_name}: {e}")
+                raise
 
         # Wait for resources to be fully deleted to prevent 409 conflicts
         # on immediate re-provisioning

backend/onyx/server/features/build/sandbox/tasks/tasks.py

@@ -80,7 +80,7 @@ def cleanup_idle_sandboxes_task(self: Task, *, tenant_id: str) -> None:  # noqa:
 
     # Prevent overlapping runs of this task
     if not lock.acquire(blocking=False):
-        task_logger.debug("cleanup_idle_sandboxes_task - lock not acquired, skipping")
+        task_logger.info("cleanup_idle_sandboxes_task - lock not acquired, skipping")
         return
 
     try:

backend/onyx/server/features/document_set/api.py

@@ -11,6 +11,7 @@ from onyx.configs.app_configs import DISABLE_VECTOR_DB
 from onyx.configs.constants import OnyxCeleryPriority
 from onyx.configs.constants import OnyxCeleryTask
 from onyx.db.document_set import check_document_sets_are_public
+from onyx.db.document_set import delete_document_set as db_delete_document_set
 from onyx.db.document_set import fetch_all_document_sets_for_user
 from onyx.db.document_set import get_document_set_by_id
 from onyx.db.document_set import insert_document_set
@@ -142,7 +143,10 @@ def delete_document_set(
     except Exception as e:
         raise HTTPException(status_code=400, detail=str(e))
 
-    if not DISABLE_VECTOR_DB:
+    if DISABLE_VECTOR_DB:
+        db_session.refresh(document_set)
+        db_delete_document_set(document_set, db_session)
+    else:
         client_app.send_task(
             OnyxCeleryTask.CHECK_FOR_VESPA_SYNC_TASK,
             kwargs={"tenant_id": tenant_id},

backend/onyx/server/features/persona/api.py

@@ -32,7 +32,7 @@ from onyx.db.persona import get_persona_snapshots_for_user
 from onyx.db.persona import get_persona_snapshots_paginated
 from onyx.db.persona import mark_persona_as_deleted
 from onyx.db.persona import mark_persona_as_not_deleted
-from onyx.db.persona import update_persona_is_default
+from onyx.db.persona import update_persona_featured
 from onyx.db.persona import update_persona_label
 from onyx.db.persona import update_persona_public_status
 from onyx.db.persona import update_persona_shared
@@ -130,8 +130,8 @@ class IsPublicRequest(BaseModel):
     is_public: bool
 
 
-class IsDefaultRequest(BaseModel):
-    is_default_persona: bool
+class IsFeaturedRequest(BaseModel):
+    featured: bool
 
 
 @admin_router.patch("/{persona_id}/visible")
@@ -168,22 +168,22 @@ def patch_user_persona_public_status(
         raise HTTPException(status_code=403, detail=str(e))
 
 
-@admin_router.patch("/{persona_id}/default")
-def patch_persona_default_status(
+@admin_router.patch("/{persona_id}/featured")
+def patch_persona_featured_status(
     persona_id: int,
-    is_default_request: IsDefaultRequest,
+    is_featured_request: IsFeaturedRequest,
     user: User = Depends(current_curator_or_admin_user),
     db_session: Session = Depends(get_session),
 ) -> None:
     try:
-        update_persona_is_default(
+        update_persona_featured(
             persona_id=persona_id,
-            is_default=is_default_request.is_default_persona,
+            featured=is_featured_request.featured,
             db_session=db_session,
             user=user,
         )
     except ValueError as e:
-        logger.exception("Failed to update persona default status")
+        logger.exception("Failed to update persona featured status")
         raise HTTPException(status_code=403, detail=str(e))
 
 

backend/onyx/server/features/persona/models.py

@@ -5,7 +5,6 @@ from pydantic import BaseModel
 from pydantic import Field
 
 from onyx.configs.constants import DocumentSource
-from onyx.context.search.enums import RecencyBiasSetting
 from onyx.db.enums import HierarchyNodeType
 from onyx.db.models import Document
 from onyx.db.models import HierarchyNode
@@ -108,11 +107,7 @@ class PersonaUpsertRequest(BaseModel):
     name: str
     description: str
     document_set_ids: list[int]
-    num_chunks: float
     is_public: bool
-    recency_bias: RecencyBiasSetting
-    llm_filter_extraction: bool
-    llm_relevance_filter: bool
     llm_model_provider_override: str | None = None
     llm_model_version_override: str | None = None
     starter_messages: list[StarterMessage] | None = None
@@ -128,7 +123,7 @@ class PersonaUpsertRequest(BaseModel):
     )
     search_start_date: datetime | None = None
     label_ids: list[int] | None = None
-    is_default_persona: bool = False
+    featured: bool = False
     display_priority: int | None = None
     # Accept string UUIDs from frontend
     user_file_ids: list[str] | None = None
@@ -155,9 +150,6 @@ class MinimalPersonaSnapshot(BaseModel):
     tools: list[ToolSnapshot]
     starter_messages: list[StarterMessage] | None
 
-    llm_relevance_filter: bool
-    llm_filter_extraction: bool
-
     # only show document sets in the UI that the assistant has access to
     document_sets: list[DocumentSetSummary]
     # Counts for knowledge sources (used to determine if search tool should be enabled)
@@ -175,7 +167,7 @@ class MinimalPersonaSnapshot(BaseModel):
     is_public: bool
     is_visible: bool
     display_priority: int | None
-    is_default_persona: bool
+    featured: bool
     builtin_persona: bool
 
     # Used for filtering
@@ -214,8 +206,6 @@ class MinimalPersonaSnapshot(BaseModel):
                 if should_expose_tool_to_fe(tool)
             ],
             starter_messages=persona.starter_messages,
-            llm_relevance_filter=persona.llm_relevance_filter,
-            llm_filter_extraction=persona.llm_filter_extraction,
             document_sets=[
                 DocumentSetSummary.from_model(document_set)
                 for document_set in persona.document_sets
@@ -230,7 +220,7 @@ class MinimalPersonaSnapshot(BaseModel):
             is_public=persona.is_public,
             is_visible=persona.is_visible,
             display_priority=persona.display_priority,
-            is_default_persona=persona.is_default_persona,
+            featured=persona.featured,
             builtin_persona=persona.builtin_persona,
             labels=[PersonaLabelSnapshot.from_model(label) for label in persona.labels],
             owner=(
@@ -252,11 +242,9 @@ class PersonaSnapshot(BaseModel):
     # Return string UUIDs to frontend for consistency
     user_file_ids: list[str]
     display_priority: int | None
-    is_default_persona: bool
+    featured: bool
     builtin_persona: bool
     starter_messages: list[StarterMessage] | None
-    llm_relevance_filter: bool
-    llm_filter_extraction: bool
     tools: list[ToolSnapshot]
     labels: list["PersonaLabelSnapshot"]
     owner: MinimalUserSnapshot | None
@@ -265,7 +253,6 @@ class PersonaSnapshot(BaseModel):
     document_sets: list[DocumentSetSummary]
     llm_model_provider_override: str | None
     llm_model_version_override: str | None
-    num_chunks: float | None
     # Hierarchy nodes attached for scoped search
     hierarchy_nodes: list[HierarchyNodeSnapshot] = Field(default_factory=list)
     # Individual documents attached for scoped search
@@ -289,11 +276,9 @@ class PersonaSnapshot(BaseModel):
             icon_name=persona.icon_name,
             user_file_ids=[str(file.id) for file in persona.user_files],
             display_priority=persona.display_priority,
-            is_default_persona=persona.is_default_persona,
+            featured=persona.featured,
             builtin_persona=persona.builtin_persona,
             starter_messages=persona.starter_messages,
-            llm_relevance_filter=persona.llm_relevance_filter,
-            llm_filter_extraction=persona.llm_filter_extraction,
             tools=[
                 ToolSnapshot.from_model(tool)
                 for tool in persona.tools
@@ -324,7 +309,6 @@ class PersonaSnapshot(BaseModel):
             ],
             llm_model_provider_override=persona.llm_model_provider_override,
             llm_model_version_override=persona.llm_model_version_override,
-            num_chunks=persona.num_chunks,
             system_prompt=persona.system_prompt,
             replace_base_system_prompt=persona.replace_base_system_prompt,
             task_prompt=persona.task_prompt,
@@ -332,12 +316,10 @@ class PersonaSnapshot(BaseModel):
         )
 
 
-# Model with full context on perona's internal settings
+# Model with full context on persona's internal settings
 # This is used for flows which need to know all settings
 class FullPersonaSnapshot(PersonaSnapshot):
     search_start_date: datetime | None = None
-    llm_relevance_filter: bool = False
-    llm_filter_extraction: bool = False
 
     @classmethod
     def from_model(
@@ -360,7 +342,7 @@ class FullPersonaSnapshot(PersonaSnapshot):
             icon_name=persona.icon_name,
             user_file_ids=[str(file.id) for file in persona.user_files],
             display_priority=persona.display_priority,
-            is_default_persona=persona.is_default_persona,
+            featured=persona.featured,
             builtin_persona=persona.builtin_persona,
             starter_messages=persona.starter_messages,
             users=[
@@ -391,10 +373,7 @@ class FullPersonaSnapshot(PersonaSnapshot):
                 DocumentSetSummary.from_model(document_set_model)
                 for document_set_model in persona.document_sets
             ],
-            num_chunks=persona.num_chunks,
             search_start_date=persona.search_start_date,
-            llm_relevance_filter=persona.llm_relevance_filter,
-            llm_filter_extraction=persona.llm_filter_extraction,
             llm_model_provider_override=persona.llm_model_provider_override,
             llm_model_version_override=persona.llm_model_version_override,
             system_prompt=persona.system_prompt,

backend/onyx/server/features/projects/api.py

@@ -2,6 +2,7 @@ import json
 from uuid import UUID
 
 from fastapi import APIRouter
+from fastapi import BackgroundTasks
 from fastapi import Depends
 from fastapi import File
 from fastapi import Form
@@ -12,13 +13,7 @@ from pydantic import BaseModel
 from sqlalchemy.orm import Session
 
 from onyx.auth.users import current_user
-from onyx.background.celery.tasks.user_file_processing.tasks import (
-    enqueue_user_file_project_sync_task,
-)
-from onyx.background.celery.tasks.user_file_processing.tasks import (
-    get_user_file_project_sync_queue_depth,
-)
-from onyx.background.celery.versioned_apps.client import app as client_app
+from onyx.configs.app_configs import DISABLE_VECTOR_DB
 from onyx.configs.constants import OnyxCeleryPriority
 from onyx.configs.constants import OnyxCeleryQueues
 from onyx.configs.constants import OnyxCeleryTask
@@ -34,7 +29,6 @@ from onyx.db.models import UserProject
 from onyx.db.persona import get_personas_by_ids
 from onyx.db.projects import get_project_token_count
 from onyx.db.projects import upload_files_to_user_files_with_indexing
-from onyx.redis.redis_pool import get_redis_client
 from onyx.server.features.projects.models import CategorizedFilesSnapshot
 from onyx.server.features.projects.models import ChatSessionRequest
 from onyx.server.features.projects.models import TokenCountResponse
@@ -55,7 +49,27 @@ class UserFileDeleteResult(BaseModel):
     assistant_names: list[str] = []
 
 
-def _trigger_user_file_project_sync(user_file_id: UUID, tenant_id: str) -> None:
+def _trigger_user_file_project_sync(
+    user_file_id: UUID,
+    tenant_id: str,
+    background_tasks: BackgroundTasks | None = None,
+) -> None:
+    if DISABLE_VECTOR_DB and background_tasks is not None:
+        from onyx.background.task_utils import drain_project_sync_loop
+
+        background_tasks.add_task(drain_project_sync_loop, tenant_id)
+        logger.info(f"Queued in-process project sync for user_file_id={user_file_id}")
+        return
+
+    from onyx.background.celery.tasks.user_file_processing.tasks import (
+        enqueue_user_file_project_sync_task,
+    )
+    from onyx.background.celery.tasks.user_file_processing.tasks import (
+        get_user_file_project_sync_queue_depth,
+    )
+    from onyx.background.celery.versioned_apps.client import app as client_app
+    from onyx.redis.redis_pool import get_redis_client
+
     queue_depth = get_user_file_project_sync_queue_depth(client_app)
     if queue_depth > USER_FILE_PROJECT_SYNC_MAX_QUEUE_DEPTH:
         logger.warning(
@@ -111,6 +125,7 @@ def create_project(
 
 @router.post("/file/upload", tags=PUBLIC_API_TAGS)
 def upload_user_files(
+    bg_tasks: BackgroundTasks,
     files: list[UploadFile] = File(...),
     project_id: int | None = Form(None),
     temp_id_map: str | None = Form(None),  # JSON string mapping hashed key -> temp_id
@@ -137,12 +152,12 @@ def upload_user_files(
             user=user,
             temp_id_map=parsed_temp_id_map,
             db_session=db_session,
+            background_tasks=bg_tasks if DISABLE_VECTOR_DB else None,
         )
 
         return CategorizedFilesSnapshot.from_result(categorized_files_result)
 
     except Exception as e:
-        # Log error with type, message, and stack for easier debugging
         logger.exception(f"Error uploading files - {type(e).__name__}: {str(e)}")
         raise HTTPException(
             status_code=500,
@@ -192,6 +207,7 @@ def get_files_in_project(
 def unlink_user_file_from_project(
     project_id: int,
     file_id: UUID,
+    bg_tasks: BackgroundTasks,
     user: User = Depends(current_user),
     db_session: Session = Depends(get_session),
 ) -> Response:
@@ -208,7 +224,6 @@ def unlink_user_file_from_project(
     if project is None:
         raise HTTPException(status_code=404, detail="Project not found")
 
-    user_id = user.id
     user_file = (
         db_session.query(UserFile)
         .filter(UserFile.id == file_id, UserFile.user_id == user_id)
@@ -224,7 +239,7 @@ def unlink_user_file_from_project(
         db_session.commit()
 
     tenant_id = get_current_tenant_id()
-    _trigger_user_file_project_sync(user_file.id, tenant_id)
+    _trigger_user_file_project_sync(user_file.id, tenant_id, bg_tasks)
 
     return Response(status_code=204)
 
@@ -237,6 +252,7 @@ def unlink_user_file_from_project(
 def link_user_file_to_project(
     project_id: int,
     file_id: UUID,
+    bg_tasks: BackgroundTasks,
     user: User = Depends(current_user),
     db_session: Session = Depends(get_session),
 ) -> UserFileSnapshot:
@@ -268,7 +284,7 @@ def link_user_file_to_project(
         db_session.commit()
 
     tenant_id = get_current_tenant_id()
-    _trigger_user_file_project_sync(user_file.id, tenant_id)
+    _trigger_user_file_project_sync(user_file.id, tenant_id, bg_tasks)
 
     return UserFileSnapshot.from_model(user_file)
 
@@ -335,7 +351,7 @@ def upsert_project_instructions(
 class ProjectPayload(BaseModel):
     project: UserProjectSnapshot
     files: list[UserFileSnapshot] | None = None
-    persona_id_to_is_default: dict[int, bool] | None = None
+    persona_id_to_featured: dict[int, bool] | None = None
 
 
 @router.get(
@@ -354,13 +370,11 @@ def get_project_details(
         if session.persona_id is not None
     ]
     personas = get_personas_by_ids(persona_ids, db_session)
-    persona_id_to_is_default = {
-        persona.id: persona.is_default_persona for persona in personas
-    }
+    persona_id_to_featured = {persona.id: persona.featured for persona in personas}
     return ProjectPayload(
         project=project,
         files=files,
-        persona_id_to_is_default=persona_id_to_is_default,
+        persona_id_to_featured=persona_id_to_featured,
     )
 
 
@@ -426,6 +440,7 @@ def delete_project(
 @router.delete("/file/{file_id}", tags=PUBLIC_API_TAGS)
 def delete_user_file(
     file_id: UUID,
+    bg_tasks: BackgroundTasks,
     user: User = Depends(current_user),
     db_session: Session = Depends(get_session),
 ) -> UserFileDeleteResult:
@@ -458,15 +473,25 @@ def delete_user_file(
     db_session.commit()
 
     tenant_id = get_current_tenant_id()
-    task = client_app.send_task(
-        OnyxCeleryTask.DELETE_SINGLE_USER_FILE,
-        kwargs={"user_file_id": str(user_file.id), "tenant_id": tenant_id},
-        queue=OnyxCeleryQueues.USER_FILE_DELETE,
-        priority=OnyxCeleryPriority.HIGH,
-    )
-    logger.info(
-        f"Triggered delete for user_file_id={user_file.id} with task_id={task.id}"
-    )
+    if DISABLE_VECTOR_DB:
+        from onyx.background.task_utils import drain_delete_loop
+
+        bg_tasks.add_task(drain_delete_loop, tenant_id)
+        logger.info(f"Queued in-process delete for user_file_id={user_file.id}")
+    else:
+        from onyx.background.celery.versioned_apps.client import app as client_app
+
+        task = client_app.send_task(
+            OnyxCeleryTask.DELETE_SINGLE_USER_FILE,
+            kwargs={"user_file_id": str(user_file.id), "tenant_id": tenant_id},
+            queue=OnyxCeleryQueues.USER_FILE_DELETE,
+            priority=OnyxCeleryPriority.HIGH,
+        )
+        logger.info(
+            f"Triggered delete for user_file_id={user_file.id} "
+            f"with task_id={task.id}"
+        )
+
     return UserFileDeleteResult(
         has_associations=False, project_names=[], assistant_names=[]
     )

backend/onyx/server/features/projects/projects_file_utils.py

@@ -7,13 +7,14 @@ from PIL import UnidentifiedImageError
 from pydantic import BaseModel
 from pydantic import ConfigDict
 from pydantic import Field
+from sqlalchemy.orm import Session
 
 from onyx.configs.app_configs import FILE_TOKEN_COUNT_THRESHOLD
+from onyx.db.llm import fetch_default_llm_model
 from onyx.file_processing.extract_file_text import extract_file_text
 from onyx.file_processing.extract_file_text import get_file_ext
 from onyx.file_processing.file_types import OnyxFileExtensions
 from onyx.file_processing.password_validation import is_file_password_protected
-from onyx.llm.factory import get_default_llm
 from onyx.natural_language_processing.utils import get_tokenizer
 from onyx.utils.logger import setup_logger
 from shared_configs.configs import MULTI_TENANT
@@ -116,7 +117,9 @@ def estimate_image_tokens_for_upload(
             pass
 
 
-def categorize_uploaded_files(files: list[UploadFile]) -> CategorizedFiles:
+def categorize_uploaded_files(
+    files: list[UploadFile], db_session: Session
+) -> CategorizedFiles:
     """
     Categorize uploaded files based on text extractability and tokenized length.
 
@@ -128,11 +131,11 @@ def categorize_uploaded_files(files: list[UploadFile]) -> CategorizedFiles:
     """
 
     results = CategorizedFiles()
-    llm = get_default_llm()
+    default_model = fetch_default_llm_model(db_session)
 
-    tokenizer = get_tokenizer(
-        model_name=llm.config.model_name, provider_type=llm.config.model_provider
-    )
+    model_name = default_model.name if default_model else None
+    provider_type = default_model.llm_provider.provider if default_model else None
+    tokenizer = get_tokenizer(model_name=model_name, provider_type=provider_type)
 
     # Check if threshold checks should be skipped
     skip_threshold = False

backend/onyx/server/features/release_notes/utils.py

@@ -8,10 +8,10 @@ import httpx
 from sqlalchemy.orm import Session
 
 from onyx import __version__
+from onyx.cache.factory import get_shared_cache_backend
 from onyx.configs.app_configs import INSTANCE_TYPE
 from onyx.configs.constants import OnyxRedisLocks
 from onyx.db.release_notes import create_release_notifications_for_versions
-from onyx.redis.redis_pool import get_shared_redis_client
 from onyx.server.features.release_notes.constants import AUTO_REFRESH_THRESHOLD_SECONDS
 from onyx.server.features.release_notes.constants import FETCH_TIMEOUT
 from onyx.server.features.release_notes.constants import GITHUB_CHANGELOG_RAW_URL
@@ -113,60 +113,46 @@ def parse_mdx_to_release_note_entries(mdx_content: str) -> list[ReleaseNoteEntry
 
 
 def get_cached_etag() -> str | None:
-    """Get the cached GitHub ETag from Redis."""
-    redis_client = get_shared_redis_client()
+    cache = get_shared_cache_backend()
     try:
-        etag = redis_client.get(REDIS_KEY_ETAG)
+        etag = cache.get(REDIS_KEY_ETAG)
         if etag:
-            return etag.decode("utf-8") if isinstance(etag, bytes) else str(etag)
+            return etag.decode("utf-8")
         return None
     except Exception as e:
-        logger.error(f"Failed to get cached etag from Redis: {e}")
+        logger.error(f"Failed to get cached etag: {e}")
         return None
 
 
 def get_last_fetch_time() -> datetime | None:
-    """Get the last fetch timestamp from Redis."""
-    redis_client = get_shared_redis_client()
+    cache = get_shared_cache_backend()
     try:
-        fetched_at_str = redis_client.get(REDIS_KEY_FETCHED_AT)
-        if not fetched_at_str:
+        raw = cache.get(REDIS_KEY_FETCHED_AT)
+        if not raw:
             return None
 
-        decoded = (
-            fetched_at_str.decode("utf-8")
-            if isinstance(fetched_at_str, bytes)
-            else str(fetched_at_str)
-        )
-
-        last_fetch = datetime.fromisoformat(decoded)
-
-        # Defensively ensure timezone awareness
-        # fromisoformat() returns naive datetime if input lacks timezone
+        last_fetch = datetime.fromisoformat(raw.decode("utf-8"))
         if last_fetch.tzinfo is None:
-            # Assume UTC for naive datetimes
             last_fetch = last_fetch.replace(tzinfo=timezone.utc)
         else:
-            # Convert to UTC if timezone-aware
             last_fetch = last_fetch.astimezone(timezone.utc)
 
         return last_fetch
     except Exception as e:
-        logger.error(f"Failed to get last fetch time from Redis: {e}")
+        logger.error(f"Failed to get last fetch time from cache: {e}")
         return None
 
 
 def save_fetch_metadata(etag: str | None) -> None:
-    """Save ETag and fetch timestamp to Redis."""
-    redis_client = get_shared_redis_client()
+    cache = get_shared_cache_backend()
     now = datetime.now(timezone.utc)
 
     try:
-        redis_client.set(REDIS_KEY_FETCHED_AT, now.isoformat(), ex=REDIS_CACHE_TTL)
+        cache.set(REDIS_KEY_FETCHED_AT, now.isoformat(), ex=REDIS_CACHE_TTL)
         if etag:
-            redis_client.set(REDIS_KEY_ETAG, etag, ex=REDIS_CACHE_TTL)
+            cache.set(REDIS_KEY_ETAG, etag, ex=REDIS_CACHE_TTL)
     except Exception as e:
-        logger.error(f"Failed to save fetch metadata to Redis: {e}")
+        logger.error(f"Failed to save fetch metadata to cache: {e}")
 
 
 def is_cache_stale() -> bool:
@@ -196,11 +182,10 @@ def ensure_release_notes_fresh_and_notify(db_session: Session) -> None:
     if not is_cache_stale():
         return
 
-    # Acquire lock to prevent concurrent fetches
-    redis_client = get_shared_redis_client()
-    lock = redis_client.lock(
+    cache = get_shared_cache_backend()
+    lock = cache.lock(
         OnyxRedisLocks.RELEASE_NOTES_FETCH_LOCK,
-        timeout=90,  # 90 second timeout for the lock
+        timeout=90,
     )
 
     # Non-blocking acquire - if we can't get the lock, another request is handling it

backend/onyx/server/kg/api.py

@@ -5,7 +5,6 @@ from sqlalchemy.orm import Session
 from onyx.auth.users import current_admin_user
 from onyx.configs.constants import TMP_DRALPHA_PERSONA_NAME
 from onyx.configs.kg_configs import KG_BETA_ASSISTANT_DESCRIPTION
-from onyx.context.search.enums import RecencyBiasSetting
 from onyx.db.engine.sql_engine import get_session
 from onyx.db.entities import get_entity_stats_by_grounded_source_name
 from onyx.db.entity_type import get_configured_entity_types
@@ -134,11 +133,7 @@ def enable_or_disable_kg(
         system_prompt=KG_BETA_ASSISTANT_SYSTEM_PROMPT,
         task_prompt=KG_BETA_ASSISTANT_TASK_PROMPT,
         datetime_aware=False,
-        num_chunks=25,
-        llm_relevance_filter=False,
         is_public=False,
-        llm_filter_extraction=False,
-        recency_bias=RecencyBiasSetting.NO_DECAY,
         document_set_ids=[],
         tool_ids=[search_tool.id, kg_tool.id],
         llm_model_provider_override=None,
@@ -147,7 +142,7 @@ def enable_or_disable_kg(
         users=[user.id],
         groups=[],
         label_ids=[],
-        is_default_persona=False,
+        featured=False,
         display_priority=0,
         user_file_ids=[],
     )

backend/onyx/server/manage/embedding/api.py

@@ -1,6 +1,5 @@
 from fastapi import APIRouter
 from fastapi import Depends
-from fastapi import HTTPException
 from sqlalchemy.orm import Session
 
 from onyx.auth.users import current_admin_user
@@ -11,6 +10,8 @@ from onyx.db.llm import upsert_cloud_embedding_provider
 from onyx.db.models import User
 from onyx.db.search_settings import get_all_search_settings
 from onyx.db.search_settings import get_current_db_embedding_provider
+from onyx.error_handling.error_codes import OnyxErrorCode
+from onyx.error_handling.exceptions import OnyxError
 from onyx.indexing.models import EmbeddingModelDetail
 from onyx.natural_language_processing.search_nlp_models import EmbeddingModel
 from onyx.server.manage.embedding.models import CloudEmbeddingProvider
@@ -59,7 +60,7 @@ def test_embedding_configuration(
     except Exception as e:
         error_msg = "An error occurred while testing your embedding model. Please check your configuration."
         logger.error(f"{error_msg} Error message: {e}", exc_info=True)
-        raise HTTPException(status_code=400, detail=error_msg)
+        raise OnyxError(OnyxErrorCode.VALIDATION_ERROR, error_msg)
 
 
 @admin_router.get("", response_model=list[EmbeddingModelDetail])
@@ -93,8 +94,9 @@ def delete_embedding_provider(
         embedding_provider is not None
         and provider_type == embedding_provider.provider_type
     ):
-        raise HTTPException(
-            status_code=400, detail="You can't delete a currently active model"
+        raise OnyxError(
+            OnyxErrorCode.VALIDATION_ERROR,
+            "You can't delete a currently active model",
         )
 
     remove_embedding_provider(db_session, provider_type=provider_type)

backend/onyx/server/manage/llm/api.py

@@ -11,7 +11,6 @@ from botocore.exceptions import ClientError
 from botocore.exceptions import NoCredentialsError
 from fastapi import APIRouter
 from fastapi import Depends
-from fastapi import HTTPException
 from fastapi import Query
 from pydantic import ValidationError
 from sqlalchemy.orm import Session
@@ -38,6 +37,8 @@ from onyx.db.llm import upsert_llm_provider
 from onyx.db.llm import validate_persona_ids_exist
 from onyx.db.models import User
 from onyx.db.persona import user_can_access_persona
+from onyx.error_handling.error_codes import OnyxErrorCode
+from onyx.error_handling.exceptions import OnyxError
 from onyx.llm.factory import get_default_llm
 from onyx.llm.factory import get_llm
 from onyx.llm.factory import get_max_input_tokens_from_llm_provider
@@ -186,7 +187,7 @@ def _validate_llm_provider_change(
     Only enforced in MULTI_TENANT mode.
 
     Raises:
-        HTTPException: If api_base or custom_config changed without changing API key
+        OnyxError: If api_base or custom_config changed without changing API key
     """
     if not MULTI_TENANT or api_key_changed:
         return
@@ -200,9 +201,9 @@ def _validate_llm_provider_change(
     )
 
     if api_base_changed or custom_config_changed:
-        raise HTTPException(
-            status_code=400,
-            detail="API base and/or custom config cannot be changed without changing the API key",
+        raise OnyxError(
+            OnyxErrorCode.VALIDATION_ERROR,
+            "API base and/or custom config cannot be changed without changing the API key",
         )
 
 
@@ -222,7 +223,7 @@ def fetch_llm_provider_options(
     for well_known_llm in well_known_llms:
         if well_known_llm.name == provider_name:
             return well_known_llm
-    raise HTTPException(status_code=404, detail=f"Provider {provider_name} not found")
+    raise OnyxError(OnyxErrorCode.NOT_FOUND, f"Provider {provider_name} not found")
 
 
 @admin_router.post("/test")
@@ -281,7 +282,7 @@ def test_llm_configuration(
     error_msg = test_llm(llm)
 
     if error_msg:
-        raise HTTPException(status_code=400, detail=error_msg)
+        raise OnyxError(OnyxErrorCode.VALIDATION_ERROR, error_msg)
 
 
 @admin_router.post("/test/default")
@@ -292,11 +293,11 @@ def test_default_provider(
         llm = get_default_llm()
     except ValueError:
         logger.exception("Failed to fetch default LLM Provider")
-        raise HTTPException(status_code=400, detail="No LLM Provider setup")
+        raise OnyxError(OnyxErrorCode.VALIDATION_ERROR, "No LLM Provider setup")
 
     error = test_llm(llm)
     if error:
-        raise HTTPException(status_code=400, detail=str(error))
+        raise OnyxError(OnyxErrorCode.VALIDATION_ERROR, str(error))
 
 
 @admin_router.get("/provider")
@@ -362,35 +363,31 @@ def put_llm_provider(
     # Check name constraints
     # TODO: Once port from name to id is complete, unique name will no longer be required
     if existing_provider and llm_provider_upsert_request.name != existing_provider.name:
-        raise HTTPException(
-            status_code=400,
-            detail="Renaming providers is not currently supported",
+        raise OnyxError(
+            OnyxErrorCode.VALIDATION_ERROR,
+            "Renaming providers is not currently supported",
         )
 
     found_provider = fetch_existing_llm_provider(
         name=llm_provider_upsert_request.name, db_session=db_session
     )
     if found_provider is not None and found_provider is not existing_provider:
-        raise HTTPException(
-            status_code=400,
-            detail=f"Provider with name={llm_provider_upsert_request.name} already exists",
+        raise OnyxError(
+            OnyxErrorCode.DUPLICATE_RESOURCE,
+            f"Provider with name={llm_provider_upsert_request.name} already exists",
         )
 
     if existing_provider and is_creation:
-        raise HTTPException(
-            status_code=400,
-            detail=(
-                f"LLM Provider with name {llm_provider_upsert_request.name} and "
-                f"id={llm_provider_upsert_request.id} already exists"
-            ),
+        raise OnyxError(
+            OnyxErrorCode.DUPLICATE_RESOURCE,
+            f"LLM Provider with name {llm_provider_upsert_request.name} and "
+            f"id={llm_provider_upsert_request.id} already exists",
         )
     elif not existing_provider and not is_creation:
-        raise HTTPException(
-            status_code=400,
-            detail=(
-                f"LLM Provider with name {llm_provider_upsert_request.name} and "
-                f"id={llm_provider_upsert_request.id} does not exist"
-            ),
+        raise OnyxError(
+            OnyxErrorCode.NOT_FOUND,
+            f"LLM Provider with name {llm_provider_upsert_request.name} and "
+            f"id={llm_provider_upsert_request.id} does not exist",
         )
 
     # SSRF Protection: Validate api_base and custom_config match stored values
@@ -415,9 +412,9 @@ def put_llm_provider(
             db_session, persona_ids
         )
         if missing_personas:
-            raise HTTPException(
-                status_code=400,
-                detail=f"Invalid persona IDs: {', '.join(map(str, missing_personas))}",
+            raise OnyxError(
+                OnyxErrorCode.VALIDATION_ERROR,
+                f"Invalid persona IDs: {', '.join(map(str, missing_personas))}",
             )
         # Remove duplicates while preserving order
         seen: set[int] = set()
@@ -473,19 +470,29 @@ def put_llm_provider(
         return result
     except ValueError as e:
         logger.exception("Failed to upsert LLM Provider")
-        raise HTTPException(status_code=400, detail=str(e))
+        raise OnyxError(OnyxErrorCode.VALIDATION_ERROR, str(e))
 
 
 @admin_router.delete("/provider/{provider_id}")
 def delete_llm_provider(
     provider_id: int,
+    force: bool = Query(False),
     _: User = Depends(current_admin_user),
     db_session: Session = Depends(get_session),
 ) -> None:
+    if not force:
+        model = fetch_default_llm_model(db_session)
+
+        if model and model.llm_provider_id == provider_id:
+            raise OnyxError(
+                OnyxErrorCode.VALIDATION_ERROR,
+                "Cannot delete the default LLM provider",
+            )
+
     try:
         remove_llm_provider(db_session, provider_id)
     except ValueError as e:
-        raise HTTPException(status_code=404, detail=str(e))
+        raise OnyxError(OnyxErrorCode.NOT_FOUND, str(e))
 
 
 @admin_router.post("/default")
@@ -525,9 +532,9 @@ def get_auto_config(
     """
     config = fetch_llm_recommendations_from_github()
     if not config:
-        raise HTTPException(
-            status_code=502,
-            detail="Failed to fetch configuration from GitHub",
+        raise OnyxError(
+            OnyxErrorCode.BAD_GATEWAY,
+            "Failed to fetch configuration from GitHub",
         )
     return config.model_dump()
 
@@ -603,9 +610,9 @@ def list_llm_provider_basics(
     for provider in all_providers:
         # Use centralized access control logic with persona=None since we're
         # listing providers without a specific persona context. This correctly:
-        # - Includes all public providers
+        # - Includes public providers WITHOUT persona restrictions
         # - Includes providers user can access via group membership
-        # - Excludes persona-only restricted providers (requires specific persona)
+        # - Excludes providers with persona restrictions (requires specific persona)
         # - Excludes non-public providers with no restrictions (admin-only)
         if can_user_access_llm_provider(
             provider, user_group_ids, persona=None, is_admin=is_admin
@@ -638,7 +645,7 @@ def get_valid_model_names_for_persona(
 
     Returns a list of model names (e.g., ["gpt-4o", "claude-3-5-sonnet"]) that are
     available to the user when using this persona, respecting all RBAC restrictions.
-    Public providers are always included.
+    Public providers are included unless they have persona restrictions that exclude this persona.
     """
     persona = fetch_persona_with_groups(db_session, persona_id)
     if not persona:
@@ -652,7 +659,7 @@ def get_valid_model_names_for_persona(
 
     valid_models = []
     for llm_provider_model in all_providers:
-        # Public providers always included, restricted checked via RBAC
+        # Check access with persona context — respects all RBAC restrictions
         if can_user_access_llm_provider(
             llm_provider_model, user_group_ids, persona, is_admin=is_admin
         ):
@@ -673,7 +680,7 @@ def list_llm_providers_for_persona(
     """Get LLM providers for a specific persona.
 
     Returns providers that the user can access when using this persona:
-    - All public providers (is_public=True) - ALWAYS included
+    - Public providers (respecting persona restrictions if set)
     - Restricted providers user can access via group/persona restrictions
 
     This endpoint is used for background fetching of restricted providers
@@ -684,13 +691,13 @@ def list_llm_providers_for_persona(
 
     persona = fetch_persona_with_groups(db_session, persona_id)
     if not persona:
-        raise HTTPException(status_code=404, detail="Persona not found")
+        raise OnyxError(OnyxErrorCode.PERSONA_NOT_FOUND, "Persona not found")
 
     # Verify user has access to this persona
     if not user_can_access_persona(db_session, persona_id, user, get_editable=False):
-        raise HTTPException(
-            status_code=403,
-            detail="You don't have access to this assistant",
+        raise OnyxError(
+            OnyxErrorCode.INSUFFICIENT_PERMISSIONS,
+            "You don't have access to this assistant",
         )
 
     is_admin = user.role == UserRole.ADMIN
@@ -702,7 +709,7 @@ def list_llm_providers_for_persona(
     llm_provider_list: list[LLMProviderDescriptor] = []
 
     for llm_provider_model in all_providers:
-        # Use simplified access check - public providers always included
+        # Check access with persona context — respects persona restrictions
         if can_user_access_llm_provider(
             llm_provider_model, user_group_ids, persona, is_admin=is_admin
         ):
@@ -844,9 +851,9 @@ def get_bedrock_available_models(
         try:
             bedrock = session.client("bedrock")
         except Exception as e:
-            raise HTTPException(
-                status_code=400,
-                detail=f"Failed to create Bedrock client: {e}. Check AWS credentials and region.",
+            raise OnyxError(
+                OnyxErrorCode.CREDENTIAL_INVALID,
+                f"Failed to create Bedrock client: {e}. Check AWS credentials and region.",
             )
 
         # Build model info dict from foundation models (modelId -> metadata)
@@ -965,14 +972,14 @@ def get_bedrock_available_models(
         return results
 
     except (ClientError, NoCredentialsError, BotoCoreError) as e:
-        raise HTTPException(
-            status_code=400,
-            detail=f"Failed to connect to AWS Bedrock: {e}",
+        raise OnyxError(
+            OnyxErrorCode.CREDENTIAL_INVALID,
+            f"Failed to connect to AWS Bedrock: {e}",
         )
     except Exception as e:
-        raise HTTPException(
-            status_code=500,
-            detail=f"Unexpected error fetching Bedrock models: {e}",
+        raise OnyxError(
+            OnyxErrorCode.INTERNAL_ERROR,
+            f"Unexpected error fetching Bedrock models: {e}",
         )
 
 
@@ -984,9 +991,9 @@ def _get_ollama_available_model_names(api_base: str) -> set[str]:
         response.raise_for_status()
         response_json = response.json()
     except Exception as e:
-        raise HTTPException(
-            status_code=400,
-            detail=f"Failed to fetch Ollama models: {e}",
+        raise OnyxError(
+            OnyxErrorCode.BAD_GATEWAY,
+            f"Failed to fetch Ollama models: {e}",
         )
 
     models = response_json.get("models", [])
@@ -1003,9 +1010,9 @@ def get_ollama_available_models(
 
     cleaned_api_base = request.api_base.strip().rstrip("/")
     if not cleaned_api_base:
-        raise HTTPException(
-            status_code=400,
-            detail="API base URL is required to fetch Ollama models.",
+        raise OnyxError(
+            OnyxErrorCode.VALIDATION_ERROR,
+            "API base URL is required to fetch Ollama models.",
         )
 
     # NOTE: most people run Ollama locally, so we don't disallow internal URLs
@@ -1014,9 +1021,9 @@ def get_ollama_available_models(
     # with the same response format
     model_names = _get_ollama_available_model_names(cleaned_api_base)
     if not model_names:
-        raise HTTPException(
-            status_code=400,
-            detail="No models found from your Ollama server",
+        raise OnyxError(
+            OnyxErrorCode.VALIDATION_ERROR,
+            "No models found from your Ollama server",
         )
 
     all_models_with_context_size_and_vision: list[OllamaFinalModelResponse] = []
@@ -1118,9 +1125,9 @@ def _get_openrouter_models_response(api_base: str, api_key: str) -> dict:
         response.raise_for_status()
         return response.json()
     except Exception as e:
-        raise HTTPException(
-            status_code=400,
-            detail=f"Failed to fetch OpenRouter models: {e}",
+        raise OnyxError(
+            OnyxErrorCode.BAD_GATEWAY,
+            f"Failed to fetch OpenRouter models: {e}",
         )
 
 
@@ -1141,9 +1148,9 @@ def get_openrouter_available_models(
 
     data = response_json.get("data", [])
     if not isinstance(data, list) or len(data) == 0:
-        raise HTTPException(
-            status_code=400,
-            detail="No models found from your OpenRouter endpoint",
+        raise OnyxError(
+            OnyxErrorCode.VALIDATION_ERROR,
+            "No models found from your OpenRouter endpoint",
         )
 
     results: list[OpenRouterFinalModelResponse] = []
@@ -1178,8 +1185,9 @@ def get_openrouter_available_models(
             )
 
     if not results:
-        raise HTTPException(
-            status_code=400, detail="No compatible models found from OpenRouter"
+        raise OnyxError(
+            OnyxErrorCode.VALIDATION_ERROR,
+            "No compatible models found from OpenRouter",
         )
 
     sorted_results = sorted(results, key=lambda m: m.name.lower())

backend/onyx/server/manage/slack_bot.py

@@ -198,7 +198,6 @@ def patch_slack_channel_config(
             channel_name=channel_config["channel_name"],
             document_set_ids=slack_channel_config_creation_request.document_sets,
             existing_persona_id=existing_persona_id,
-            enable_auto_filters=slack_channel_config_creation_request.enable_auto_filters,
         ).id
 
     slack_channel_config_model = update_slack_channel_config(

backend/onyx/server/query_and_chat/chat_backend.py

@@ -13,13 +13,13 @@ from fastapi import Request
 from fastapi import Response
 from fastapi.responses import StreamingResponse
 from pydantic import BaseModel
-from redis.client import Redis
 from sqlalchemy.orm import Session
 
 from onyx.auth.api_key import get_hashed_api_key_from_request
 from onyx.auth.pat import get_hashed_pat_from_request
 from onyx.auth.users import current_chat_accessible_user
 from onyx.auth.users import current_user
+from onyx.cache.factory import get_cache_backend
 from onyx.chat.chat_processing_checker import is_chat_session_processing
 from onyx.chat.chat_state import ChatStateContainer
 from onyx.chat.chat_utils import convert_chat_history_basic
@@ -67,7 +67,6 @@ from onyx.llm.constants import LlmProviderNames
 from onyx.llm.factory import get_default_llm
 from onyx.llm.factory import get_llm_for_persona
 from onyx.llm.factory import get_llm_token_counter
-from onyx.redis.redis_pool import get_redis_client
 from onyx.secondary_llm_flows.chat_session_naming import generate_chat_session_name
 from onyx.server.api_key_usage import check_api_key_usage
 from onyx.server.query_and_chat.models import ChatFeedbackRequest
@@ -152,10 +151,20 @@ def get_user_chat_sessions(
     project_id: int | None = None,
     only_non_project_chats: bool = True,
     include_failed_chats: bool = False,
+    page_size: int = Query(default=50, ge=1, le=100),
+    before: str | None = Query(default=None),
 ) -> ChatSessionsResponse:
     user_id = user.id
 
     try:
+        before_dt = (
+            datetime.datetime.fromisoformat(before) if before is not None else None
+        )
+    except ValueError:
+        raise HTTPException(status_code=422, detail="Invalid 'before' timestamp format")
+
+    try:
+        # Fetch one extra to determine if there are more results
         chat_sessions = get_chat_sessions_by_user(
             user_id=user_id,
             deleted=False,
@@ -163,11 +172,16 @@ def get_user_chat_sessions(
             project_id=project_id,
             only_non_project_chats=only_non_project_chats,
             include_failed_chats=include_failed_chats,
+            limit=page_size + 1,
+            before=before_dt,
         )
 
     except ValueError:
         raise ValueError("Chat session does not exist or has been deleted")
 
+    has_more = len(chat_sessions) > page_size
+    chat_sessions = chat_sessions[:page_size]
+
     return ChatSessionsResponse(
         sessions=[
             ChatSessionDetails(
@@ -181,7 +195,8 @@ def get_user_chat_sessions(
                 current_temperature_override=chat.temperature_override,
             )
             for chat in chat_sessions
-        ]
+        ],
+        has_more=has_more,
     )
 
 
@@ -314,7 +329,7 @@ def get_chat_session(
     ]
 
     try:
-        is_processing = is_chat_session_processing(session_id, get_redis_client())
+        is_processing = is_chat_session_processing(session_id, get_cache_backend())
         # Edit the last message to indicate loading (Overriding default message value)
         if is_processing and chat_message_details:
             last_msg = chat_message_details[-1]
@@ -911,11 +926,10 @@ async def search_chats(
 def stop_chat_session(
     chat_session_id: UUID,
     user: User = Depends(current_user),  # noqa: ARG001
-    redis_client: Redis = Depends(get_redis_client),
 ) -> dict[str, str]:
     """
-    Stop a chat session by setting a stop signal in Redis.
+    Stop a chat session by setting a stop signal.
     This endpoint is called by the frontend when the user clicks the stop button.
     """
-    set_fence(chat_session_id, redis_client, True)
+    set_fence(chat_session_id, get_cache_backend(), True)
     return {"message": "Chat session stopped"}

backend/onyx/server/query_and_chat/models.py

@@ -192,6 +192,7 @@ class ChatSessionDetails(BaseModel):
 
 class ChatSessionsResponse(BaseModel):
     sessions: list[ChatSessionDetails]
+    has_more: bool = False
 
 
 class ChatMessageDetail(BaseModel):

backend/onyx/server/settings/store.py

@@ -1,3 +1,4 @@
+from onyx.cache.factory import get_cache_backend
 from onyx.configs.app_configs import DISABLE_USER_KNOWLEDGE
 from onyx.configs.app_configs import ENABLE_OPENSEARCH_INDEXING_FOR_ONYX
 from onyx.configs.app_configs import ONYX_QUERY_HISTORY_TYPE
@@ -6,11 +7,8 @@ from onyx.configs.constants import KV_SETTINGS_KEY
 from onyx.configs.constants import OnyxRedisLocks
 from onyx.key_value_store.factory import get_kv_store
 from onyx.key_value_store.interface import KvKeyNotFoundError
-from onyx.redis.redis_pool import get_redis_client
 from onyx.server.settings.models import Settings
 from onyx.utils.logger import setup_logger
-from shared_configs.configs import MULTI_TENANT
-from shared_configs.contextvars import get_current_tenant_id
 
 logger = setup_logger()
 
@@ -33,30 +31,22 @@ def load_settings() -> Settings:
         logger.error(f"Error loading settings from KV store: {str(e)}")
         settings = Settings()
 
-    tenant_id = get_current_tenant_id() if MULTI_TENANT else None
-    redis_client = get_redis_client(tenant_id=tenant_id)
+    cache = get_cache_backend()
 
     try:
-        value = redis_client.get(OnyxRedisLocks.ANONYMOUS_USER_ENABLED)
+        value = cache.get(OnyxRedisLocks.ANONYMOUS_USER_ENABLED)
         if value is not None:
-            assert isinstance(value, bytes)
             anonymous_user_enabled = int(value.decode("utf-8")) == 1
         else:
-            # Default to False
             anonymous_user_enabled = False
-            # Optionally store the default back to Redis
-            redis_client.set(
-                OnyxRedisLocks.ANONYMOUS_USER_ENABLED, "0", ex=SETTINGS_TTL
-            )
+            cache.set(OnyxRedisLocks.ANONYMOUS_USER_ENABLED, "0", ex=SETTINGS_TTL)
     except Exception as e:
-        # Log the error and reset to default
-        logger.error(f"Error loading anonymous user setting from Redis: {str(e)}")
+        logger.error(f"Error loading anonymous user setting from cache: {str(e)}")
         anonymous_user_enabled = False
 
     settings.anonymous_user_enabled = anonymous_user_enabled
     settings.query_history_type = ONYX_QUERY_HISTORY_TYPE
 
-    # Override user knowledge setting if disabled via environment variable
     if DISABLE_USER_KNOWLEDGE:
         settings.user_knowledge_enabled = False
 
@@ -66,11 +56,10 @@ def load_settings() -> Settings:
 
 
 def store_settings(settings: Settings) -> None:
-    tenant_id = get_current_tenant_id() if MULTI_TENANT else None
-    redis_client = get_redis_client(tenant_id=tenant_id)
+    cache = get_cache_backend()
 
     if settings.anonymous_user_enabled is not None:
-        redis_client.set(
+        cache.set(
             OnyxRedisLocks.ANONYMOUS_USER_ENABLED,
             "1" if settings.anonymous_user_enabled else "0",
             ex=SETTINGS_TTL,

backend/onyx/tools/models.py

@@ -93,6 +93,8 @@ class ToolResponse(BaseModel):
         # | WebContentResponse
         # This comes from custom tools, tool result needs to be saved
         | CustomToolCallSummary
+        # This comes from code interpreter, carries generated files
+        | PythonToolRichResponse
         # If the rich response is a string, this is what's saved to the tool call in the DB
         | str
         | None  # If nothing needs to be persisted outside of the string value passed to the LLM
@@ -193,6 +195,12 @@ class ChatFile(BaseModel):
     model_config = ConfigDict(arbitrary_types_allowed=True)
 
 
+class PythonToolRichResponse(BaseModel):
+    """Rich response from the Python tool carrying generated files."""
+
+    generated_files: list[PythonExecutionFile] = []
+
+
 class PythonToolOverrideKwargs(BaseModel):
     """Override kwargs for the Python/Code Interpreter tool."""
 
@@ -245,6 +253,7 @@ class ToolCallInfo(BaseModel):
     tool_call_response: str
     search_docs: list[SearchDoc] | None = None
     generated_images: list[GeneratedImage] | None = None
+    generated_files: list[PythonExecutionFile] | None = None
 
 
 CHAT_SESSION_ID_PLACEHOLDER = "CHAT_SESSION_ID"

backend/onyx/tools/tool_implementations/python/code_interpreter_client.py

@@ -1,4 +1,7 @@
+from __future__ import annotations
+
 import json
+import time
 from collections.abc import Generator
 from typing import Literal
 from typing import TypedDict
@@ -12,6 +15,9 @@ from onyx.utils.logger import setup_logger
 
 logger = setup_logger()
 
+_HEALTH_CACHE_TTL_SECONDS = 30
+_health_cache: dict[str, tuple[float, bool]] = {}
+
 
 class FileInput(TypedDict):
     """Input file to be staged in execution workspace"""
@@ -80,6 +86,19 @@ class CodeInterpreterClient:
             raise ValueError("CODE_INTERPRETER_BASE_URL not configured")
         self.base_url = base_url.rstrip("/")
         self.session = requests.Session()
+        self._closed = False
+
+    def __enter__(self) -> CodeInterpreterClient:
+        return self
+
+    def __exit__(self, *args: object) -> None:
+        self.close()
+
+    def close(self) -> None:
+        if self._closed:
+            return
+        self.session.close()
+        self._closed = True
 
     def _build_payload(
         self,
@@ -98,16 +117,32 @@ class CodeInterpreterClient:
             payload["files"] = files
         return payload
 
-    def health(self) -> bool:
-        """Check if the Code Interpreter service is healthy"""
+    def health(self, use_cache: bool = False) -> bool:
+        """Check if the Code Interpreter service is healthy
+
+        Args:
+            use_cache: When True, return a cached result if available and
+                       within the TTL window. The cache is always populated
+                       after a live request regardless of this flag.
+        """
+        if use_cache:
+            cached = _health_cache.get(self.base_url)
+            if cached is not None:
+                cached_at, cached_result = cached
+                if time.monotonic() - cached_at < _HEALTH_CACHE_TTL_SECONDS:
+                    return cached_result
+
         url = f"{self.base_url}/health"
         try:
             response = self.session.get(url, timeout=5)
             response.raise_for_status()
-            return response.json().get("status") == "ok"
+            result = response.json().get("status") == "ok"
         except Exception as e:
             logger.warning(f"Exception caught when checking health, e={e}")
-            return False
+            result = False
+
+        _health_cache[self.base_url] = (time.monotonic(), result)
+        return result
 
     def execute(
         self,
@@ -157,8 +192,11 @@ class CodeInterpreterClient:
             yield from self._batch_as_stream(code, stdin, timeout_ms, files)
             return
 
-        response.raise_for_status()
-        yield from self._parse_sse(response)
+        try:
+            response.raise_for_status()
+            yield from self._parse_sse(response)
+        finally:
+            response.close()
 
     def _parse_sse(
         self, response: requests.Response

backend/onyx/tools/tool_implementations/python/python_tool.py

@@ -23,6 +23,7 @@ from onyx.tools.interface import Tool
 from onyx.tools.models import LlmPythonExecutionResult
 from onyx.tools.models import PythonExecutionFile
 from onyx.tools.models import PythonToolOverrideKwargs
+from onyx.tools.models import PythonToolRichResponse
 from onyx.tools.models import ToolCallException
 from onyx.tools.models import ToolResponse
 from onyx.tools.tool_implementations.python.code_interpreter_client import (
@@ -107,7 +108,11 @@ class PythonTool(Tool[PythonToolOverrideKwargs]):
         if not CODE_INTERPRETER_BASE_URL:
             return False
         server = fetch_code_interpreter_server(db_session)
-        return server.server_enabled
+        if not server.server_enabled:
+            return False
+
+        with CodeInterpreterClient() as client:
+            return client.health(use_cache=True)
 
     def tool_definition(self) -> dict:
         return {
@@ -171,194 +176,203 @@ class PythonTool(Tool[PythonToolOverrideKwargs]):
             )
         )
 
-        # Create Code Interpreter client
-        client = CodeInterpreterClient()
+        # Create Code Interpreter client — context manager ensures
+        # session.close() is called on every exit path.
+        with CodeInterpreterClient() as client:
+            # Stage chat files for execution
+            files_to_stage: list[FileInput] = []
+            for ind, chat_file in enumerate(chat_files):
+                file_name = chat_file.filename or f"file_{ind}"
+                try:
+                    # Upload to Code Interpreter
+                    ci_file_id = client.upload_file(chat_file.content, file_name)
+
+                    # Stage for execution
+                    files_to_stage.append({"path": file_name, "file_id": ci_file_id})
+
+                    logger.info(f"Staged file for Python execution: {file_name}")
+
+                except Exception as e:
+                    logger.warning(f"Failed to stage file {file_name}: {e}")
 
-        # Stage chat files for execution
-        files_to_stage: list[FileInput] = []
-        for ind, chat_file in enumerate(chat_files):
-            file_name = chat_file.filename or f"file_{ind}"
             try:
-                # Upload to Code Interpreter
-                ci_file_id = client.upload_file(chat_file.content, file_name)
+                logger.debug(f"Executing code: {code}")
 
-                # Stage for execution
-                files_to_stage.append({"path": file_name, "file_id": ci_file_id})
+                # Execute code with streaming (falls back to batch if unavailable)
+                stdout_parts: list[str] = []
+                stderr_parts: list[str] = []
+                result_event: StreamResultEvent | None = None
 
-                logger.info(f"Staged file for Python execution: {file_name}")
+                for event in client.execute_streaming(
+                    code=code,
+                    timeout_ms=CODE_INTERPRETER_DEFAULT_TIMEOUT_MS,
+                    files=files_to_stage or None,
+                ):
+                    if isinstance(event, StreamOutputEvent):
+                        if event.stream == "stdout":
+                            stdout_parts.append(event.data)
+                        else:
+                            stderr_parts.append(event.data)
+                        # Emit incremental delta to frontend
+                        self.emitter.emit(
+                            Packet(
+                                placement=placement,
+                                obj=PythonToolDelta(
+                                    stdout=(
+                                        event.data if event.stream == "stdout" else ""
+                                    ),
+                                    stderr=(
+                                        event.data if event.stream == "stderr" else ""
+                                    ),
+                                ),
+                            )
+                        )
+                    elif isinstance(event, StreamResultEvent):
+                        result_event = event
+                    elif isinstance(event, StreamErrorEvent):
+                        raise RuntimeError(f"Code interpreter error: {event.message}")
 
-            except Exception as e:
-                logger.warning(f"Failed to stage file {file_name}: {e}")
+                if result_event is None:
+                    raise RuntimeError(
+                        "Code interpreter stream ended without a result event"
+                    )
 
-        try:
-            logger.debug(f"Executing code: {code}")
+                full_stdout = "".join(stdout_parts)
+                full_stderr = "".join(stderr_parts)
 
-            # Execute code with streaming (falls back to batch if unavailable)
-            stdout_parts: list[str] = []
-            stderr_parts: list[str] = []
-            result_event: StreamResultEvent | None = None
+                # Truncate output for LLM consumption
+                truncated_stdout = _truncate_output(
+                    full_stdout, CODE_INTERPRETER_MAX_OUTPUT_LENGTH, "stdout"
+                )
+                truncated_stderr = _truncate_output(
+                    full_stderr, CODE_INTERPRETER_MAX_OUTPUT_LENGTH, "stderr"
+                )
 
-            for event in client.execute_streaming(
-                code=code,
-                timeout_ms=CODE_INTERPRETER_DEFAULT_TIMEOUT_MS,
-                files=files_to_stage or None,
-            ):
-                if isinstance(event, StreamOutputEvent):
-                    if event.stream == "stdout":
-                        stdout_parts.append(event.data)
-                    else:
-                        stderr_parts.append(event.data)
-                    # Emit incremental delta to frontend
+                # Handle generated files
+                generated_files: list[PythonExecutionFile] = []
+                generated_file_ids: list[str] = []
+                file_ids_to_cleanup: list[str] = []
+                file_store = get_default_file_store()
+
+                for workspace_file in result_event.files:
+                    if workspace_file.kind != "file" or not workspace_file.file_id:
+                        continue
+
+                    try:
+                        # Download file from Code Interpreter
+                        file_content = client.download_file(workspace_file.file_id)
+
+                        # Determine MIME type from file extension
+                        filename = workspace_file.path.split("/")[-1]
+                        mime_type, _ = mimetypes.guess_type(filename)
+                        # Default to binary if we can't determine the type
+                        mime_type = mime_type or "application/octet-stream"
+
+                        # Save to Onyx file store
+                        onyx_file_id = file_store.save_file(
+                            content=BytesIO(file_content),
+                            display_name=filename,
+                            file_origin=FileOrigin.CHAT_UPLOAD,
+                            file_type=mime_type,
+                        )
+
+                        generated_files.append(
+                            PythonExecutionFile(
+                                filename=filename,
+                                file_link=build_full_frontend_file_url(onyx_file_id),
+                            )
+                        )
+                        generated_file_ids.append(onyx_file_id)
+
+                        # Mark for cleanup
+                        file_ids_to_cleanup.append(workspace_file.file_id)
+
+                    except Exception as e:
+                        logger.error(
+                            f"Failed to handle generated file "
+                            f"{workspace_file.path}: {e}"
+                        )
+
+                # Cleanup Code Interpreter files (generated files)
+                for ci_file_id in file_ids_to_cleanup:
+                    try:
+                        client.delete_file(ci_file_id)
+                    except Exception as e:
+                        logger.error(
+                            f"Failed to delete Code Interpreter generated "
+                            f"file {ci_file_id}: {e}"
+                        )
+
+                # Cleanup staged input files
+                for file_mapping in files_to_stage:
+                    try:
+                        client.delete_file(file_mapping["file_id"])
+                    except Exception as e:
+                        logger.error(
+                            f"Failed to delete Code Interpreter staged "
+                            f"file {file_mapping['file_id']}: {e}"
+                        )
+
+                # Emit file_ids once files are processed
+                if generated_file_ids:
                     self.emitter.emit(
                         Packet(
                             placement=placement,
-                            obj=PythonToolDelta(
-                                stdout=event.data if event.stream == "stdout" else "",
-                                stderr=event.data if event.stream == "stderr" else "",
-                            ),
+                            obj=PythonToolDelta(file_ids=generated_file_ids),
                         )
                     )
-                elif isinstance(event, StreamResultEvent):
-                    result_event = event
-                elif isinstance(event, StreamErrorEvent):
-                    raise RuntimeError(f"Code interpreter error: {event.message}")
 
-            if result_event is None:
-                raise RuntimeError(
-                    "Code interpreter stream ended without a result event"
+                # Build result
+                result = LlmPythonExecutionResult(
+                    stdout=truncated_stdout,
+                    stderr=truncated_stderr,
+                    exit_code=result_event.exit_code,
+                    timed_out=result_event.timed_out,
+                    generated_files=generated_files,
+                    error=(None if result_event.exit_code == 0 else truncated_stderr),
                 )
 
-            full_stdout = "".join(stdout_parts)
-            full_stderr = "".join(stderr_parts)
+                # Serialize result for LLM
+                adapter = TypeAdapter(LlmPythonExecutionResult)
+                llm_response = adapter.dump_json(result).decode()
 
-            # Truncate output for LLM consumption
-            truncated_stdout = _truncate_output(
-                full_stdout, CODE_INTERPRETER_MAX_OUTPUT_LENGTH, "stdout"
-            )
-            truncated_stderr = _truncate_output(
-                full_stderr, CODE_INTERPRETER_MAX_OUTPUT_LENGTH, "stderr"
-            )
+                return ToolResponse(
+                    rich_response=PythonToolRichResponse(
+                        generated_files=generated_files,
+                    ),
+                    llm_facing_response=llm_response,
+                )
 
-            # Handle generated files
-            generated_files: list[PythonExecutionFile] = []
-            generated_file_ids: list[str] = []
-            file_ids_to_cleanup: list[str] = []
-            file_store = get_default_file_store()
+            except Exception as e:
+                logger.error(f"Python execution failed: {e}")
+                error_msg = str(e)
 
-            for workspace_file in result_event.files:
-                if workspace_file.kind != "file" or not workspace_file.file_id:
-                    continue
-
-                try:
-                    # Download file from Code Interpreter
-                    file_content = client.download_file(workspace_file.file_id)
-
-                    # Determine MIME type from file extension
-                    filename = workspace_file.path.split("/")[-1]
-                    mime_type, _ = mimetypes.guess_type(filename)
-                    # Default to binary if we can't determine the type
-                    mime_type = mime_type or "application/octet-stream"
-
-                    # Save to Onyx file store
-                    onyx_file_id = file_store.save_file(
-                        content=BytesIO(file_content),
-                        display_name=filename,
-                        file_origin=FileOrigin.CHAT_UPLOAD,
-                        file_type=mime_type,
-                    )
-
-                    generated_files.append(
-                        PythonExecutionFile(
-                            filename=filename,
-                            file_link=build_full_frontend_file_url(onyx_file_id),
-                        )
-                    )
-                    generated_file_ids.append(onyx_file_id)
-
-                    # Mark for cleanup
-                    file_ids_to_cleanup.append(workspace_file.file_id)
-
-                except Exception as e:
-                    logger.error(
-                        f"Failed to handle generated file {workspace_file.path}: {e}"
-                    )
-
-            # Cleanup Code Interpreter files (generated files)
-            for ci_file_id in file_ids_to_cleanup:
-                try:
-                    client.delete_file(ci_file_id)
-                except Exception as e:
-                    logger.error(
-                        f"Failed to delete Code Interpreter generated file {ci_file_id}: {e}"
-                    )
-
-            # Cleanup staged input files
-            for file_mapping in files_to_stage:
-                try:
-                    client.delete_file(file_mapping["file_id"])
-                except Exception as e:
-                    logger.error(
-                        f"Failed to delete Code Interpreter staged file {file_mapping['file_id']}: {e}"
-                    )
-
-            # Emit file_ids once files are processed
-            if generated_file_ids:
+                # Emit error delta
                 self.emitter.emit(
                     Packet(
                         placement=placement,
-                        obj=PythonToolDelta(file_ids=generated_file_ids),
+                        obj=PythonToolDelta(
+                            stdout="",
+                            stderr=error_msg,
+                            file_ids=[],
+                        ),
                     )
                 )
 
-            # Build result
-            result = LlmPythonExecutionResult(
-                stdout=truncated_stdout,
-                stderr=truncated_stderr,
-                exit_code=result_event.exit_code,
-                timed_out=result_event.timed_out,
-                generated_files=generated_files,
-                error=None if result_event.exit_code == 0 else truncated_stderr,
-            )
-
-            # Serialize result for LLM
-            adapter = TypeAdapter(LlmPythonExecutionResult)
-            llm_response = adapter.dump_json(result).decode()
-
-            return ToolResponse(
-                rich_response=None,  # No rich response needed for Python tool
-                llm_facing_response=llm_response,
-            )
-
-        except Exception as e:
-            logger.error(f"Python execution failed: {e}")
-            error_msg = str(e)
-
-            # Emit error delta
-            self.emitter.emit(
-                Packet(
-                    placement=placement,
-                    obj=PythonToolDelta(
-                        stdout="",
-                        stderr=error_msg,
-                        file_ids=[],
-                    ),
+                # Return error result
+                result = LlmPythonExecutionResult(
+                    stdout="",
+                    stderr=error_msg,
+                    exit_code=-1,
+                    timed_out=False,
+                    generated_files=[],
+                    error=error_msg,
                 )
-            )
 
-            # Return error result
-            result = LlmPythonExecutionResult(
-                stdout="",
-                stderr=error_msg,
-                exit_code=-1,
-                timed_out=False,
-                generated_files=[],
-                error=error_msg,
-            )
+                adapter = TypeAdapter(LlmPythonExecutionResult)
+                llm_response = adapter.dump_json(result).decode()
 
-            adapter = TypeAdapter(LlmPythonExecutionResult)
-            llm_response = adapter.dump_json(result).decode()
-
-            return ToolResponse(
-                rich_response=None,
-                llm_facing_response=llm_response,
-            )
+                return ToolResponse(
+                    rich_response=None,
+                    llm_facing_response=llm_response,
+                )