Make it possible to reenable LFC

The full build of all extensions takes a long time. When working locally
on parts that don't need extensions, you can iterate more quickly by
skipping the unnecessary extensions.

This adds a build argument to the dockerfile to specify extensions to
build. There are three options:

- EXTENSIONS=all (default)
- EXTENSIONS=minimal: Build only a few extensions that are listed in
shared_preload_libraries in the default neon config.
- EXTENSIONS=none: Build no extensions (except for the mandatory 'neon'
extension).

.github/ISSUE_TEMPLATE/bug-template.md

@@ -3,6 +3,7 @@ name: Bug Template
 about: Used for describing bugs
 title: ''
 labels: t/bug
+type: Bug
 assignees: ''
 
 ---

.github/ISSUE_TEMPLATE/epic-template.md

@@ -4,6 +4,7 @@ about: A set of related tasks contributing towards specific outcome, comprising
   more than 1 week of work.
 title: 'Epic: '
 labels: t/Epic
+type: Epic
 assignees: ''
 
 ---

.github/actionlint.yml

@@ -4,6 +4,7 @@ self-hosted-runner:
     - large
     - large-arm64
     - small
+    - small-metal
     - small-arm64
     - us-east-2
 config-variables:
@@ -26,3 +27,4 @@ config-variables:
   - SLACK_ON_CALL_QA_STAGING_STREAM
   - DEV_AWS_OIDC_ROLE_MANAGE_BENCHMARK_EC2_VMS_ARN
   - SLACK_ON_CALL_STORAGE_STAGING_STREAM
+  - SLACK_CICD_CHANNEL_ID

.github/actions/neon-project-create/action.yml

@@ -17,6 +17,34 @@ inputs:
   compute_units:
     description: '[Min, Max] compute units'
     default: '[1, 1]'
+  # settings below only needed if you want the project to be sharded from the beginning
+  shard_split_project:
+    description: 'by default new projects are not shard-split, specify true to shard-split'
+    required: false
+    default: 'false'
+  admin_api_key:
+    description: 'Admin API Key needed for shard-splitting. Must be specified if shard_split_project is true'
+    required: false
+  shard_count:
+    description: 'Number of shards to split the project into, only applies if shard_split_project is true'
+    required: false
+    default: '8'
+  stripe_size:
+    description: 'Stripe size, optional, in 8kiB pages.  e.g. set 2048 for 16MB stripes. Default is 128 MiB, only applies if shard_split_project is true'
+    required: false
+    default: '32768'
+  psql_path:
+    description: 'Path to psql binary - it is caller responsibility to provision the psql binary'
+    required: false
+    default: '/tmp/neon/pg_install/v16/bin/psql'
+  libpq_lib_path:
+    description: 'Path to directory containing libpq library - it is caller responsibility to provision the libpq library'
+    required: false
+    default: '/tmp/neon/pg_install/v16/lib'
+  project_settings:
+    description: 'A JSON object with project settings'
+    required: false
+    default: '{}'
 
 outputs:
   dsn:
@@ -48,7 +76,7 @@ runs:
               \"provisioner\": \"k8s-neonvm\",
               \"autoscaling_limit_min_cu\": ${MIN_CU},
               \"autoscaling_limit_max_cu\": ${MAX_CU},
-              \"settings\": { }
+              \"settings\": ${PROJECT_SETTINGS}
             }
           }")
 
@@ -63,6 +91,23 @@ runs:
         echo "project_id=${project_id}" >> $GITHUB_OUTPUT
 
         echo "Project ${project_id} has been created"
+
+        if [ "${SHARD_SPLIT_PROJECT}" = "true" ]; then
+          # determine tenant ID
+          TENANT_ID=`${PSQL} ${dsn} -t -A -c "SHOW neon.tenant_id"`
+
+          echo "Splitting project ${project_id} with tenant_id ${TENANT_ID} into $((SHARD_COUNT)) shards with stripe size $((STRIPE_SIZE))"
+
+          echo "Sending PUT request to https://${API_HOST}/regions/${REGION_ID}/api/v1/admin/storage/proxy/control/v1/tenant/${TENANT_ID}/shard_split"
+          echo "with body {\"new_shard_count\": $((SHARD_COUNT)), \"new_stripe_size\": $((STRIPE_SIZE))}"
+
+          # we need an ADMIN API KEY to invoke storage controller API for shard splitting (bash -u above checks that the variable is set)
+          curl -X PUT \
+            "https://${API_HOST}/regions/${REGION_ID}/api/v1/admin/storage/proxy/control/v1/tenant/${TENANT_ID}/shard_split" \
+            -H "Accept: application/json" -H "Content-Type: application/json" -H "Authorization: Bearer ${ADMIN_API_KEY}" \
+            -d "{\"new_shard_count\": $SHARD_COUNT, \"new_stripe_size\": $STRIPE_SIZE}"
+        fi
+
       env:
         API_HOST: ${{ inputs.api_host }}
         API_KEY: ${{ inputs.api_key }}
@@ -70,3 +115,10 @@ runs:
         POSTGRES_VERSION: ${{ inputs.postgres_version }}
         MIN_CU: ${{ fromJSON(inputs.compute_units)[0] }}
         MAX_CU: ${{ fromJSON(inputs.compute_units)[1] }}
+        SHARD_SPLIT_PROJECT: ${{ inputs.shard_split_project }}
+        ADMIN_API_KEY: ${{ inputs.admin_api_key }}
+        SHARD_COUNT: ${{ inputs.shard_count }}
+        STRIPE_SIZE: ${{ inputs.stripe_size }}
+        PSQL: ${{ inputs.psql_path }}
+        LD_LIBRARY_PATH: ${{ inputs.libpq_lib_path }}
+        PROJECT_SETTINGS: ${{ inputs.project_settings }}

.github/file-filters.yaml

@@ -1,4 +1,5 @@
 rust_code: ['**/*.rs', '**/Cargo.toml', '**/Cargo.lock']
+rust_dependencies: ['**/Cargo.lock']
 
 v14: ['vendor/postgres-v14/**', 'Makefile', 'pgxn/**']
 v15: ['vendor/postgres-v15/**', 'Makefile', 'pgxn/**']

.github/workflows/_benchmarking_preparation.yml

@@ -17,7 +17,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        platform: [ aws-rds-postgres, aws-aurora-serverless-v2-postgres, neon ]
+        platform: [ aws-rds-postgres, aws-aurora-serverless-v2-postgres, neon, neon_pg17 ]
         database: [ clickbench, tpch, userexample ]
 
     env:
@@ -41,6 +41,9 @@ jobs:
           neon)
             CONNSTR=${{ secrets.BENCHMARK_CAPTEST_CONNSTR }}
             ;;
+          neon_pg17)
+            CONNSTR=${{ secrets.BENCHMARK_CAPTEST_CONNSTR_PG17 }}
+            ;;
           aws-rds-postgres)
             CONNSTR=${{ secrets.BENCHMARK_RDS_POSTGRES_CONNSTR }}
             ;;

.github/workflows/_build-and-test-locally.yml

@@ -158,8 +158,6 @@ jobs:
 
       - name: Run cargo build
         run: |
-          PQ_LIB_DIR=$(pwd)/pg_install/v16/lib
-          export PQ_LIB_DIR
           ${cov_prefix} mold -run cargo build $CARGO_FLAGS $CARGO_FEATURES --bins --tests
 
       # Do install *before* running rust tests because they might recompile the
@@ -217,8 +215,6 @@ jobs:
         env:
           NEXTEST_RETRIES: 3
         run: |
-          PQ_LIB_DIR=$(pwd)/pg_install/v16/lib
-          export PQ_LIB_DIR
           LD_LIBRARY_PATH=$(pwd)/pg_install/v17/lib
           export LD_LIBRARY_PATH
 
@@ -229,8 +225,13 @@ jobs:
           ${cov_prefix} cargo nextest run $CARGO_FLAGS $CARGO_FEATURES -E '!package(pageserver)'
 
           # run pageserver tests with different settings
-          for io_engine in std-fs tokio-epoll-uring ; do
-            NEON_PAGESERVER_UNIT_TEST_VIRTUAL_FILE_IOENGINE=$io_engine ${cov_prefix} cargo nextest run $CARGO_FLAGS $CARGO_FEATURES  -E 'package(pageserver)'
+          for get_vectored_concurrent_io in sequential sidecar-task; do
+            for io_engine in std-fs tokio-epoll-uring ; do
+              NEON_PAGESERVER_UNIT_TEST_GET_VECTORED_CONCURRENT_IO=$get_vectored_concurrent_io \
+                NEON_PAGESERVER_UNIT_TEST_VIRTUAL_FILE_IOENGINE=$io_engine \
+                ${cov_prefix} \
+                cargo nextest run $CARGO_FLAGS $CARGO_FEATURES  -E 'package(pageserver)'
+            done
           done
 
           # Run separate tests for real S3
@@ -266,6 +267,26 @@ jobs:
           path: /tmp/neon
           aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
 
+      - name: Check diesel schema
+        if: inputs.build-type == 'release' && inputs.arch == 'x64'
+        env:
+          DATABASE_URL: postgresql://localhost:1235/storage_controller
+          POSTGRES_DISTRIB_DIR: /tmp/neon/pg_install
+        run: |
+          /tmp/neon/bin/neon_local init
+          /tmp/neon/bin/neon_local storage_controller start
+
+          diesel print-schema > storage_controller/src/schema.rs
+
+          if [ -n "$(git diff storage_controller/src/schema.rs)" ]; then
+            echo >&2 "Uncommitted changes in diesel schema"
+
+            git diff .
+            exit 1
+          fi
+
+          /tmp/neon/bin/neon_local storage_controller stop
+
       # XXX: keep this after the binaries.list is formed, so the coverage can properly work later
       - name: Merge and upload coverage data
         if: inputs.build-type == 'debug'
@@ -314,6 +335,7 @@ jobs:
           CHECK_ONDISK_DATA_COMPATIBILITY: nonempty
           BUILD_TAG: ${{ inputs.build-tag }}
           PAGESERVER_VIRTUAL_FILE_IO_ENGINE: tokio-epoll-uring
+          PAGESERVER_GET_VECTORED_CONCURRENT_IO: sidecar-task
           USE_LFC: ${{ matrix.lfc_state == 'with-lfc' && 'true' || 'false' }}
 
       # Temporary disable this step until we figure out why it's so flaky

.github/workflows/_check-codestyle-rust.yml

@@ -16,6 +16,9 @@ defaults:
   run:
     shell: bash -euxo pipefail {0}
 
+# No permission for GITHUB_TOKEN by default; the **minimal required** set of permissions should be granted in each job.
+permissions: {}
+
 jobs:
   check-codestyle-rust:
     strategy:
@@ -84,8 +87,3 @@ jobs:
         run: |
           cargo hakari generate --diff  # workspace-hack Cargo.toml is up-to-date
           cargo hakari manage-deps --dry-run  # all workspace crates depend on workspace-hack
-
-      # https://github.com/EmbarkStudios/cargo-deny
-      - name: Check rust licenses/bans/advisories/sources
-        if: ${{ !cancelled() }}
-        run: cargo deny check --hide-inclusion-graph

.github/workflows/approved-for-ci-run.yml

@@ -94,7 +94,9 @@ jobs:
           echo "LABELS_TO_ADD=${LABELS_TO_ADD}" >> ${GITHUB_OUTPUT}
           echo "LABELS_TO_REMOVE=${LABELS_TO_REMOVE}" >> ${GITHUB_OUTPUT}
 
-      - run: gh pr checkout "${PR_NUMBER}"
+      - uses: actions/checkout@v4
+        with:
+          ref: ${{ github.event.pull_request.head.sha }}
 
       - run: git checkout -b "${BRANCH}"
 

.github/workflows/benchmarking.yml

@@ -63,11 +63,15 @@ jobs:
       fail-fast: false
       matrix:
         include:
-          - DEFAULT_PG_VERSION: 16
+          - PG_VERSION: 16
             PLATFORM: "neon-staging"
             region_id: ${{ github.event.inputs.region_id || 'aws-us-east-2' }}
             RUNNER: [ self-hosted, us-east-2, x64 ]
-          - DEFAULT_PG_VERSION: 16
+          - PG_VERSION: 17
+            PLATFORM: "neon-staging"
+            region_id: ${{ github.event.inputs.region_id || 'aws-us-east-2' }}
+            RUNNER: [ self-hosted, us-east-2, x64 ]
+          - PG_VERSION: 16
             PLATFORM: "azure-staging"
             region_id: 'azure-eastus2'
             RUNNER: [ self-hosted, eastus2, x64 ]
@@ -75,7 +79,7 @@ jobs:
       TEST_PG_BENCH_DURATIONS_MATRIX: "300"
       TEST_PG_BENCH_SCALES_MATRIX: "10,100"
       POSTGRES_DISTRIB_DIR: /tmp/neon/pg_install
-      DEFAULT_PG_VERSION: ${{ matrix.DEFAULT_PG_VERSION }}
+      PG_VERSION: ${{ matrix.PG_VERSION }}
       TEST_OUTPUT: /tmp/test_output
       BUILD_TYPE: remote
       SAVE_PERF_REPORT: ${{ github.event.inputs.save_perf_report || ( github.ref_name == 'main' ) }}
@@ -112,7 +116,7 @@ jobs:
       uses: ./.github/actions/neon-project-create
       with:
         region_id: ${{ matrix.region_id }}
-        postgres_version: ${{ env.DEFAULT_PG_VERSION }}
+        postgres_version: ${{ env.PG_VERSION }}
         api_key: ${{ secrets.NEON_STAGING_API_KEY }}
 
     - name: Run benchmark
@@ -122,7 +126,7 @@ jobs:
         test_selection: performance
         run_in_parallel: false
         save_perf_report: ${{ env.SAVE_PERF_REPORT }}
-        pg_version: ${{ env.DEFAULT_PG_VERSION }}
+        pg_version: ${{ env.PG_VERSION }}
         aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
         # Set --sparse-ordering option of pytest-order plugin
         # to ensure tests are running in order of appears in the file.
@@ -313,7 +317,11 @@ jobs:
                       { "pg_version": 16, "region_id": "azure-eastus2",          "platform": "neonvm-azure-captest-freetier", "db_size": "3gb" ,"runner": '"$runner_azure"',   "image": "neondatabase/build-tools:pinned-bookworm" },
                       { "pg_version": 16, "region_id": "azure-eastus2",          "platform": "neonvm-azure-captest-new",      "db_size": "10gb","runner": '"$runner_azure"',   "image": "neondatabase/build-tools:pinned-bookworm" },
                       { "pg_version": 16, "region_id": "azure-eastus2",          "platform": "neonvm-azure-captest-new",      "db_size": "50gb","runner": '"$runner_azure"',   "image": "neondatabase/build-tools:pinned-bookworm" },
-                      { "pg_version": 16, "region_id": "'"$region_id_default"'", "platform": "neonvm-captest-sharding-reuse", "db_size": "50gb","runner": '"$runner_default"', "image": "'"$image_default"'" }]
+                      { "pg_version": 16, "region_id": "'"$region_id_default"'", "platform": "neonvm-captest-sharding-reuse", "db_size": "50gb","runner": '"$runner_default"', "image": "'"$image_default"'" },
+                      { "pg_version": 17, "region_id": "'"$region_id_default"'", "platform": "neonvm-captest-freetier",       "db_size": "3gb" ,"runner": '"$runner_default"', "image": "'"$image_default"'" },
+                      { "pg_version": 17, "region_id": "'"$region_id_default"'", "platform": "neonvm-captest-new",            "db_size": "10gb","runner": '"$runner_default"', "image": "'"$image_default"'" },
+                      { "pg_version": 17, "region_id": "'"$region_id_default"'", "platform": "neonvm-captest-new-many-tables","db_size": "10gb","runner": '"$runner_default"', "image": "'"$image_default"'" },
+                      { "pg_version": 17, "region_id": "'"$region_id_default"'", "platform": "neonvm-captest-new",            "db_size": "50gb","runner": '"$runner_default"', "image": "'"$image_default"'" }]
         }'
 
         if [ "$(date +%A)" = "Saturday" ] || [ ${RUN_AWS_RDS_AND_AURORA} = "true" ]; then
@@ -329,12 +337,15 @@ jobs:
         matrix='{
           "platform": [
             "neonvm-captest-reuse"
+          ],
+          "pg_version" : [
+            16,17
           ]
         }'
 
         if [ "$(date +%A)" = "Saturday" ] || [ ${RUN_AWS_RDS_AND_AURORA} = "true" ]; then
-          matrix=$(echo "$matrix" | jq '.include += [{ "platform": "rds-postgres" },
-                                                     { "platform": "rds-aurora"   }]')
+          matrix=$(echo "$matrix" | jq '.include += [{ "pg_version": 16, "platform": "rds-postgres" },
+                                                     { "pg_version": 16, "platform": "rds-aurora"   }]')
         fi
 
         echo "matrix=$(echo "$matrix" | jq --compact-output '.')" >> $GITHUB_OUTPUT
@@ -346,14 +357,14 @@ jobs:
           "platform": [
             "neonvm-captest-reuse"
           ],
-          "scale": [
-            "10"
+          "pg_version" : [
+            16,17
           ]
         }'
 
         if [ "$(date +%A)" = "Saturday" ] || [ ${RUN_AWS_RDS_AND_AURORA} = "true" ]; then
-          matrix=$(echo "$matrix" | jq '.include += [{ "platform": "rds-postgres", "scale": "10" },
-                                                     { "platform": "rds-aurora",   "scale": "10" }]')
+          matrix=$(echo "$matrix" | jq '.include += [{ "pg_version": 16, "platform": "rds-postgres" },
+                                                     { "pg_version": 16, "platform": "rds-aurora"   }]')
         fi
 
         echo "matrix=$(echo "$matrix" | jq --compact-output '.')" >> $GITHUB_OUTPUT
@@ -378,7 +389,7 @@ jobs:
       TEST_PG_BENCH_DURATIONS_MATRIX: "60m"
       TEST_PG_BENCH_SCALES_MATRIX: ${{ matrix.db_size }}
       POSTGRES_DISTRIB_DIR: /tmp/neon/pg_install
-      DEFAULT_PG_VERSION: ${{ matrix.pg_version }}
+      PG_VERSION: ${{ matrix.pg_version }}
       TEST_OUTPUT: /tmp/test_output
       BUILD_TYPE: remote
       SAVE_PERF_REPORT: ${{ github.event.inputs.save_perf_report || ( github.ref_name == 'main' ) }}
@@ -416,7 +427,7 @@ jobs:
       uses: ./.github/actions/neon-project-create
       with:
         region_id: ${{ matrix.region_id }}
-        postgres_version: ${{ env.DEFAULT_PG_VERSION }}
+        postgres_version: ${{ env.PG_VERSION }}
         api_key: ${{ secrets.NEON_STAGING_API_KEY }}
         compute_units: ${{ (contains(matrix.platform, 'captest-freetier') && '[0.25, 0.25]') || '[1, 1]' }}
 
@@ -447,7 +458,7 @@ jobs:
 
         echo "connstr=${CONNSTR}" >> $GITHUB_OUTPUT
 
-    # we want to compare Neon project OLTP throughput and latency at scale factor 10 GB 
+    # we want to compare Neon project OLTP throughput and latency at scale factor 10 GB
     # without (neonvm-captest-new)
     # and with (neonvm-captest-new-many-tables) many relations in the database
     - name: Create many relations before the run
@@ -459,7 +470,7 @@ jobs:
         run_in_parallel: false
         save_perf_report: ${{ env.SAVE_PERF_REPORT }}
         extra_params: -m remote_cluster --timeout 21600 -k test_perf_many_relations
-        pg_version: ${{ env.DEFAULT_PG_VERSION }}
+        pg_version: ${{ env.PG_VERSION }}
         aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
       env:
         BENCHMARK_CONNSTR: ${{ steps.set-up-connstr.outputs.connstr }}
@@ -475,7 +486,7 @@ jobs:
         run_in_parallel: false
         save_perf_report: ${{ env.SAVE_PERF_REPORT }}
         extra_params: -m remote_cluster --timeout 21600 -k test_pgbench_remote_init
-        pg_version: ${{ env.DEFAULT_PG_VERSION }}
+        pg_version: ${{ env.PG_VERSION }}
         aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
       env:
         BENCHMARK_CONNSTR: ${{ steps.set-up-connstr.outputs.connstr }}
@@ -490,7 +501,7 @@ jobs:
         run_in_parallel: false
         save_perf_report: ${{ env.SAVE_PERF_REPORT }}
         extra_params: -m remote_cluster --timeout 21600 -k test_pgbench_remote_simple_update
-        pg_version: ${{ env.DEFAULT_PG_VERSION }}
+        pg_version: ${{ env.PG_VERSION }}
         aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
       env:
         BENCHMARK_CONNSTR: ${{ steps.set-up-connstr.outputs.connstr }}
@@ -505,7 +516,7 @@ jobs:
         run_in_parallel: false
         save_perf_report: ${{ env.SAVE_PERF_REPORT }}
         extra_params: -m remote_cluster --timeout 21600 -k test_pgbench_remote_select_only
-        pg_version: ${{ env.DEFAULT_PG_VERSION }}
+        pg_version: ${{ env.PG_VERSION }}
         aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
       env:
         BENCHMARK_CONNSTR: ${{ steps.set-up-connstr.outputs.connstr }}
@@ -549,14 +560,19 @@ jobs:
         include:
           - PLATFORM: "neonvm-captest-pgvector"
             RUNNER: [ self-hosted, us-east-2, x64 ]
+            postgres_version: 16
+          - PLATFORM: "neonvm-captest-pgvector-pg17"
+            RUNNER: [ self-hosted, us-east-2, x64 ]
+            postgres_version: 17
           - PLATFORM: "azure-captest-pgvector"
             RUNNER: [ self-hosted, eastus2, x64 ]
+            postgres_version: 16
 
     env:
       TEST_PG_BENCH_DURATIONS_MATRIX: "15m"
       TEST_PG_BENCH_SCALES_MATRIX: "1"
       POSTGRES_DISTRIB_DIR: /tmp/neon/pg_install
-      DEFAULT_PG_VERSION: 16
+      PG_VERSION: ${{ matrix.postgres_version }}
       TEST_OUTPUT: /tmp/test_output
       BUILD_TYPE: remote
 
@@ -574,32 +590,20 @@ jobs:
     steps:
     - uses: actions/checkout@v4
 
-    # until https://github.com/neondatabase/neon/issues/8275 is fixed we temporarily install postgresql-16
-    # instead of using Neon artifacts containing pgbench
-    - name: Install postgresql-16 where pytest expects it
-      run: |
-        # Just to make it easier to test things locally on macOS (with arm64)
-        arch=$(uname -m | sed 's/x86_64/amd64/g' | sed 's/aarch64/arm64/g')
+    - name: Configure AWS credentials
+      uses: aws-actions/configure-aws-credentials@v4
+      with:
+        aws-region: eu-central-1
+        role-to-assume: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
+        role-duration-seconds: 18000 # 5 hours
 
-        cd /home/nonroot
-        wget -q "https://apt.postgresql.org/pub/repos/apt/pool/main/p/postgresql-17/libpq5_17.2-1.pgdg120+1_${arch}.deb"
-        wget -q "https://apt.postgresql.org/pub/repos/apt/pool/main/p/postgresql-16/postgresql-client-16_16.6-1.pgdg120+1_${arch}.deb"
-        wget -q "https://apt.postgresql.org/pub/repos/apt/pool/main/p/postgresql-16/postgresql-16_16.6-1.pgdg120+1_${arch}.deb"
-        dpkg -x libpq5_17.2-1.pgdg120+1_${arch}.deb pg
-        dpkg -x postgresql-16_16.6-1.pgdg120+1_${arch}.deb pg
-        dpkg -x postgresql-client-16_16.6-1.pgdg120+1_${arch}.deb pg
-
-        mkdir -p /tmp/neon/pg_install/v16/bin
-        ln -s /home/nonroot/pg/usr/lib/postgresql/16/bin/pgbench /tmp/neon/pg_install/v16/bin/pgbench
-        ln -s /home/nonroot/pg/usr/lib/postgresql/16/bin/psql    /tmp/neon/pg_install/v16/bin/psql
-        ln -s /home/nonroot/pg/usr/lib/$(uname -m)-linux-gnu     /tmp/neon/pg_install/v16/lib
-
-        LD_LIBRARY_PATH="/home/nonroot/pg/usr/lib/$(uname -m)-linux-gnu:${LD_LIBRARY_PATH:-}"
-        export LD_LIBRARY_PATH
-        echo "LD_LIBRARY_PATH=${LD_LIBRARY_PATH}" >> ${GITHUB_ENV}
-
-        /tmp/neon/pg_install/v16/bin/pgbench --version
-        /tmp/neon/pg_install/v16/bin/psql --version
+    - name: Download Neon artifact
+      uses: ./.github/actions/download
+      with:
+        name: neon-${{ runner.os }}-${{ runner.arch }}-release-artifact
+        path: /tmp/neon/
+        prefix: latest
+        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
 
     - name: Set up Connection String
       id: set-up-connstr
@@ -608,6 +612,9 @@ jobs:
           neonvm-captest-pgvector)
             CONNSTR=${{ secrets.BENCHMARK_PGVECTOR_CONNSTR }}
             ;;
+          neonvm-captest-pgvector-pg17)
+            CONNSTR=${{ secrets.BENCHMARK_PGVECTOR_CONNSTR_PG17 }}
+            ;;
           azure-captest-pgvector)
             CONNSTR=${{ secrets.BENCHMARK_PGVECTOR_CONNSTR_AZURE }}
             ;;
@@ -619,13 +626,6 @@ jobs:
 
         echo "connstr=${CONNSTR}" >> $GITHUB_OUTPUT
 
-    - name: Configure AWS credentials
-      uses: aws-actions/configure-aws-credentials@v4
-      with:
-        aws-region: eu-central-1
-        role-to-assume: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
-        role-duration-seconds: 18000 # 5 hours
-
     - name: Benchmark pgvector hnsw indexing
       uses: ./.github/actions/run-python-test-set
       with:
@@ -634,7 +634,7 @@ jobs:
         run_in_parallel: false
         save_perf_report: ${{ env.SAVE_PERF_REPORT }}
         extra_params: -m remote_cluster --timeout 21600 -k test_pgvector_indexing
-        pg_version: ${{ env.DEFAULT_PG_VERSION }}
+        pg_version: ${{ env.PG_VERSION }}
         aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
       env:
         VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
@@ -649,7 +649,7 @@ jobs:
         run_in_parallel: false
         save_perf_report: ${{ env.SAVE_PERF_REPORT }}
         extra_params: -m remote_cluster --timeout 21600
-        pg_version: ${{ env.DEFAULT_PG_VERSION }}
+        pg_version: ${{ env.PG_VERSION }}
         aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
       env:
         BENCHMARK_CONNSTR: ${{ steps.set-up-connstr.outputs.connstr }}
@@ -696,7 +696,7 @@ jobs:
 
     env:
       POSTGRES_DISTRIB_DIR: /tmp/neon/pg_install
-      DEFAULT_PG_VERSION: 16
+      PG_VERSION: ${{ matrix.pg_version }}
       TEST_OUTPUT: /tmp/test_output
       TEST_OLAP_COLLECT_EXPLAIN: ${{ github.event.inputs.collect_olap_explain }}
       TEST_OLAP_COLLECT_PG_STAT_STATEMENTS: ${{ github.event.inputs.collect_pg_stat_statements }}
@@ -739,7 +739,18 @@ jobs:
       run: |
         case "${PLATFORM}" in
           neonvm-captest-reuse)
-            CONNSTR=${{ secrets.BENCHMARK_CAPTEST_CLICKBENCH_10M_CONNSTR }}
+            case "${PG_VERSION}" in
+              16)
+                CONNSTR=${{ secrets.BENCHMARK_CAPTEST_CLICKBENCH_10M_CONNSTR }}
+                ;;
+              17)
+                CONNSTR=${{ secrets.BENCHMARK_CAPTEST_CLICKBENCH_CONNSTR_PG17 }}
+                ;;
+              *)
+                echo >&2 "Unsupported PG_VERSION=${PG_VERSION} for PLATFORM=${PLATFORM}"
+                exit 1
+                ;;
+            esac
             ;;
           rds-aurora)
             CONNSTR=${{ secrets.BENCHMARK_RDS_AURORA_CLICKBENCH_10M_CONNSTR }}
@@ -763,7 +774,7 @@ jobs:
         run_in_parallel: false
         save_perf_report: ${{ env.SAVE_PERF_REPORT }}
         extra_params: -m remote_cluster --timeout 43200 -k test_clickbench
-        pg_version: ${{ env.DEFAULT_PG_VERSION }}
+        pg_version: ${{ env.PG_VERSION }}
         aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
       env:
         VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
@@ -799,7 +810,7 @@ jobs:
     # We might change it after https://github.com/neondatabase/neon/issues/2900.
     #
     # *_TPCH_S10_CONNSTR: DB generated with scale factor 10 (~10 GB)
-    if: ${{ !cancelled() && (github.event.inputs.run_only_pgvector_tests == 'false' || github.event.inputs.run_only_pgvector_tests == null) }}
+    # if: ${{ !cancelled() && (github.event.inputs.run_only_pgvector_tests == 'false' || github.event.inputs.run_only_pgvector_tests == null) }}
     permissions:
       contents: write
       statuses: write
@@ -812,12 +823,11 @@ jobs:
 
     env:
       POSTGRES_DISTRIB_DIR: /tmp/neon/pg_install
-      DEFAULT_PG_VERSION: 16
+      PG_VERSION: ${{ matrix.pg_version }}
       TEST_OUTPUT: /tmp/test_output
       BUILD_TYPE: remote
       SAVE_PERF_REPORT: ${{ github.event.inputs.save_perf_report || ( github.ref_name == 'main' ) }}
       PLATFORM: ${{ matrix.platform }}
-      TEST_OLAP_SCALE: ${{ matrix.scale }}
 
     runs-on: [ self-hosted, us-east-2, x64 ]
     container:
@@ -849,13 +859,24 @@ jobs:
       run: |
         case "${PLATFORM}" in
           neonvm-captest-reuse)
-            ENV_PLATFORM=CAPTEST_TPCH
+            case "${PG_VERSION}" in
+              16)
+                CONNSTR_SECRET_NAME="BENCHMARK_CAPTEST_TPCH_S10_CONNSTR"
+                ;;
+              17)
+                CONNSTR_SECRET_NAME="BENCHMARK_CAPTEST_TPCH_CONNSTR_PG17"
+                ;;
+              *)
+                echo >&2 "Unsupported PG_VERSION=${PG_VERSION} for PLATFORM=${PLATFORM}"
+                exit 1
+                ;;
+            esac
             ;;
           rds-aurora)
-            ENV_PLATFORM=RDS_AURORA_TPCH
+            CONNSTR_SECRET_NAME="BENCHMARK_RDS_AURORA_TPCH_S10_CONNSTR"
             ;;
           rds-postgres)
-            ENV_PLATFORM=RDS_POSTGRES_TPCH
+            CONNSTR_SECRET_NAME="BENCHMARK_RDS_POSTGRES_TPCH_S10_CONNSTR"
             ;;
           *)
             echo >&2 "Unknown PLATFORM=${PLATFORM}. Allowed only 'neonvm-captest-reuse', 'rds-aurora', or 'rds-postgres'"
@@ -863,7 +884,6 @@ jobs:
             ;;
         esac
 
-        CONNSTR_SECRET_NAME="BENCHMARK_${ENV_PLATFORM}_S${TEST_OLAP_SCALE}_CONNSTR"
         echo "CONNSTR_SECRET_NAME=${CONNSTR_SECRET_NAME}" >> $GITHUB_ENV
 
     - name: Set up Connection String
@@ -881,13 +901,13 @@ jobs:
         run_in_parallel: false
         save_perf_report: ${{ env.SAVE_PERF_REPORT }}
         extra_params: -m remote_cluster --timeout 21600 -k test_tpch
-        pg_version: ${{ env.DEFAULT_PG_VERSION }}
+        pg_version: ${{ env.PG_VERSION }}
         aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
       env:
         VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
         PERF_TEST_RESULT_CONNSTR: "${{ secrets.PERF_TEST_RESULT_CONNSTR }}"
         BENCHMARK_CONNSTR: ${{ steps.set-up-connstr.outputs.connstr }}
-        TEST_OLAP_SCALE: ${{ matrix.scale }}
+        TEST_OLAP_SCALE: 10
 
     - name: Create Allure report
       id: create-allure-report
@@ -909,7 +929,7 @@ jobs:
         SLACK_BOT_TOKEN: ${{ secrets.SLACK_BOT_TOKEN }}
 
   user-examples-compare:
-    if: ${{ !cancelled() && (github.event.inputs.run_only_pgvector_tests == 'false' || github.event.inputs.run_only_pgvector_tests == null) }}
+    # if: ${{ !cancelled() && (github.event.inputs.run_only_pgvector_tests == 'false' || github.event.inputs.run_only_pgvector_tests == null) }}
     permissions:
       contents: write
       statuses: write
@@ -922,7 +942,7 @@ jobs:
 
     env:
       POSTGRES_DISTRIB_DIR: /tmp/neon/pg_install
-      DEFAULT_PG_VERSION: 16
+      PG_VERSION: ${{ matrix.pg_version }}
       TEST_OUTPUT: /tmp/test_output
       BUILD_TYPE: remote
       SAVE_PERF_REPORT: ${{ github.event.inputs.save_perf_report || ( github.ref_name == 'main' ) }}
@@ -959,7 +979,18 @@ jobs:
       run: |
         case "${PLATFORM}" in
           neonvm-captest-reuse)
-            CONNSTR=${{ secrets.BENCHMARK_USER_EXAMPLE_CAPTEST_CONNSTR }}
+            case "${PG_VERSION}" in
+              16)
+                CONNSTR=${{ secrets.BENCHMARK_USER_EXAMPLE_CAPTEST_CONNSTR }}
+                ;;
+              17)
+                CONNSTR=${{ secrets.BENCHMARK_CAPTEST_USER_EXAMPLE_CONNSTR_PG17 }}
+                ;;
+              *)
+                echo >&2 "Unsupported PG_VERSION=${PG_VERSION} for PLATFORM=${PLATFORM}"
+                exit 1
+                ;;
+            esac
             ;;
           rds-aurora)
             CONNSTR=${{ secrets.BENCHMARK_USER_EXAMPLE_RDS_AURORA_CONNSTR }}
@@ -983,7 +1014,7 @@ jobs:
         run_in_parallel: false
         save_perf_report: ${{ env.SAVE_PERF_REPORT }}
         extra_params: -m remote_cluster --timeout 21600 -k test_user_examples
-        pg_version: ${{ env.DEFAULT_PG_VERSION }}
+        pg_version: ${{ env.PG_VERSION }}
         aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
       env:
         VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"

.github/workflows/build-macos.yml

@@ -235,7 +235,7 @@ jobs:
           echo 'CPPFLAGS=-I/usr/local/opt/openssl@3/include' >> $GITHUB_ENV
 
       - name: Run cargo build (only for v17)
-        run: PQ_LIB_DIR=$(pwd)/pg_install/v17/lib cargo build --all --release -j$(sysctl -n hw.ncpu)
+        run: cargo build --all --release -j$(sysctl -n hw.ncpu)
 
       - name: Check that no warnings are produced (only for v17)
         run: ./run_clippy.sh

.github/workflows/build_and_test.yml

@@ -45,6 +45,26 @@ jobs:
             run cancel-previous-in-concurrency-group.yml \
               --field concurrency_group="${{ env.E2E_CONCURRENCY_GROUP }}"
 
+  files-changed:
+    needs: [ check-permissions ]
+    runs-on: [ self-hosted, small ]
+    timeout-minutes: 3
+    outputs:
+      check-rust-dependencies: ${{ steps.files-changed.outputs.rust_dependencies }}
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          submodules: true
+
+      - name: Check for file changes
+        uses: dorny/paths-filter@de90cc6fb38fc0963ad72b210f1f284cd68cea36  # v3.0.2
+        id: files-changed
+        with:
+          token: ${{ secrets.GITHUB_TOKEN }}
+          filters: .github/file-filters.yaml
+
   tag:
     needs: [ check-permissions ]
     runs-on: [ self-hosted, small ]
@@ -170,6 +190,14 @@ jobs:
       archs: '["x64", "arm64"]'
     secrets: inherit
 
+  check-dependencies-rust:
+    needs: [ files-changed, build-build-tools-image ]
+    if: ${{ needs.files-changed.outputs.check-rust-dependencies == 'true' }}
+    uses: ./.github/workflows/cargo-deny.yml
+    with:
+      build-tools-image: ${{ needs.build-build-tools-image.outputs.image }}-bookworm
+    secrets: inherit
+
   build-and-test-locally:
     needs: [ tag, build-build-tools-image ]
     strategy:
@@ -242,7 +270,7 @@ jobs:
       statuses: write
       contents: write
       pull-requests: write
-    runs-on: [ self-hosted, small ]
+    runs-on: [ self-hosted, small-metal ]
     container:
       image: ${{ needs.build-build-tools-image.outputs.image }}-bookworm
       credentials:
@@ -654,7 +682,7 @@ jobs:
           push: true
           pull: true
           file: compute/compute-node.Dockerfile
-          target: neon-pg-ext-test
+          target: extension-tests
           cache-from: type=registry,ref=cache.neon.build/compute-node-${{ matrix.version.pg }}:cache-${{ matrix.version.debian }}-${{ matrix.arch }}
           tags: |
             neondatabase/neon-test-extensions-${{ matrix.version.pg }}:${{needs.tag.outputs.build-tag}}-${{ matrix.version.debian }}-${{ matrix.arch }}
@@ -786,6 +814,17 @@ jobs:
           username: ${{ secrets.NEON_DOCKERHUB_USERNAME }}
           password: ${{ secrets.NEON_DOCKERHUB_PASSWORD }}
 
+      - name: Get the last compute release tag
+        id: get-last-compute-release-tag
+        env:
+          GH_TOKEN: ${{ secrets.CI_ACCESS_TOKEN }}
+        run: |
+          tag=$(gh api -q '[.[].tag_name | select(startswith("release-compute"))][0]'\
+            -H "Accept: application/vnd.github+json" \
+            -H "X-GitHub-Api-Version: 2022-11-28" \
+            "/repos/${{ github.repository }}/releases")
+          echo tag=${tag} >> ${GITHUB_OUTPUT}
+
       # `neondatabase/neon` contains multiple binaries, all of them use the same input for the version into the same version formatting library.
       # Pick pageserver as currently the only binary with extra "version" features printed in the string to verify.
       # Regular pageserver version string looks like
@@ -817,14 +856,28 @@ jobs:
           TEST_VERSION_ONLY: ${{ matrix.pg_version }}
         run: ./docker-compose/docker_compose_test.sh
 
+      - name: Print logs and clean up docker-compose test
+        if: always()
+        run: |
+          docker compose --profile test-extensions -f ./docker-compose/docker-compose.yml logs || true
+          docker compose --profile test-extensions -f ./docker-compose/docker-compose.yml down
+
+      - name: Test extension upgrade
+        timeout-minutes: 20
+        if: ${{ needs.tag.outputs.build-tag == github.run_id }}
+        env:
+          NEWTAG: ${{ needs.tag.outputs.build-tag }}
+          OLDTAG: ${{ steps.get-last-compute-release-tag.outputs.tag }}
+        run: ./docker-compose/test_extensions_upgrade.sh
+
       - name: Print logs and clean up
         if: always()
         run: |
-          docker compose -f ./docker-compose/docker-compose.yml logs || 0
-          docker compose -f ./docker-compose/docker-compose.yml down
+          docker compose --profile test-extensions -f ./docker-compose/docker-compose.yml logs || true
+          docker compose --profile test-extensions -f ./docker-compose/docker-compose.yml down
 
   promote-images-dev:
-    needs: [ check-permissions, tag, vm-compute-node-image ]
+    needs: [ check-permissions, tag, vm-compute-node-image, neon-image ]
     runs-on: ubuntu-22.04
 
     permissions:
@@ -859,7 +912,7 @@ jobs:
           done
 
   promote-images-prod:
-    needs: [ check-permissions, tag, test-images, vm-compute-node-image ]
+    needs: [ check-permissions, tag, test-images, promote-images-dev ]
     runs-on: ubuntu-22.04
     if: github.ref_name == 'main' || github.ref_name == 'release' || github.ref_name == 'release-proxy' || github.ref_name == 'release-compute'
 
@@ -1050,6 +1103,7 @@ jobs:
           retries: 5
           script: |
             const tag = "${{ needs.tag.outputs.build-tag }}";
+            const branch = "${{ github.ref_name }}";
 
             try {
               const existingRef = await github.rest.git.getRef({
@@ -1078,12 +1132,6 @@ jobs:
               console.log(`Tag ${tag} created successfully.`);
             }
 
-            // TODO: check how GitHub releases looks for proxy/compute releases and enable them if they're ok
-            if (context.ref !== 'refs/heads/release') {
-              console.log(`GitHub release skipped for ${context.ref}.`);
-              return;
-            }
-
             try {
               const existingRelease = await github.rest.repos.getReleaseByTag({
                 owner: context.repo.owner,
@@ -1098,11 +1146,48 @@ jobs:
               }
 
               console.log(`Release for tag ${tag} does not exist. Creating it...`);
+
+              // Find the PR number using the commit SHA
+              const pullRequests = await github.rest.pulls.list({
+                owner: context.repo.owner,
+                repo: context.repo.repo,
+                state: 'closed',
+                base: branch,
+              });
+
+              const pr = pullRequests.data.find(pr => pr.merge_commit_sha === context.sha);
+              const prNumber = pr ? pr.number : null;
+
+              // Find the previous release on the branch
+              const releases = await github.rest.repos.listReleases({
+                owner: context.repo.owner,
+                repo: context.repo.repo,
+                per_page: 100,
+              });
+
+              const branchReleases = releases.data
+                .filter((release) => {
+                  const regex = new RegExp(`^${branch}-\\d+$`);
+                  return regex.test(release.tag_name) && !release.draft && !release.prerelease;
+                })
+                .sort((a, b) => new Date(b.created_at) - new Date(a.created_at));
+
+              const previousTag = branchReleases.length > 0 ? branchReleases[0].tag_name : null;
+
+              const releaseNotes = [
+                prNumber
+                  ? `Release PR https://github.com/${context.repo.owner}/${context.repo.repo}/pull/${prNumber}.`
+                  : 'Release PR not found.',
+                previousTag
+                  ? `Diff with the previous release https://github.com/${context.repo.owner}/${context.repo.repo}/compare/${previousTag}...${tag}.`
+                  : `No previous release found on branch ${branch}.`,
+              ].join('\n\n');
+
               await github.rest.repos.createRelease({
                 owner: context.repo.owner,
                 repo: context.repo.repo,
                 tag_name: tag,
-                generate_release_notes: true,
+                body: releaseNotes,
               });
               console.log(`Release for tag ${tag} created successfully.`);
             }
@@ -1275,6 +1360,8 @@ jobs:
       - build-and-test-locally
       - check-codestyle-python
       - check-codestyle-rust
+      - check-dependencies-rust
+      - files-changed
       - promote-images-dev
       - test-images
       - trigger-custom-extensions-build-and-wait
@@ -1287,4 +1374,11 @@ jobs:
         if: |
           contains(needs.*.result, 'failure')
           || contains(needs.*.result, 'cancelled')
-          || contains(needs.*.result, 'skipped')
+          || (needs.check-dependencies-rust.result == 'skipped' && needs.files-changed.outputs.check-rust-dependencies == 'true')
+          || needs.build-and-test-locally.result == 'skipped'
+          || needs.check-codestyle-python.result == 'skipped'
+          || needs.check-codestyle-rust.result == 'skipped'
+          || needs.files-changed.result == 'skipped'
+          || needs.promote-images-dev.result == 'skipped'
+          || needs.test-images.result == 'skipped'
+          || needs.trigger-custom-extensions-build-and-wait.result == 'skipped'

.github/workflows/cargo-deny.yml

@@ -0,0 +1,57 @@
+name: cargo deny checks
+
+on:
+  workflow_call:
+    inputs:
+      build-tools-image:
+        required: false
+        type: string
+  schedule:
+    - cron: '0 0 * * *'
+
+jobs:
+  cargo-deny:
+    strategy:
+      matrix:
+        ref: >-
+          ${{
+            fromJSON(
+              github.event_name == 'schedule'
+                && '["main","release","release-proxy","release-compute"]'
+                || format('["{0}"]', github.sha)
+            )
+          }}
+
+    runs-on: [self-hosted, small]
+
+    container:
+      image: ${{ inputs.build-tools-image || 'neondatabase/build-tools:pinned' }}
+      credentials:
+        username: ${{ secrets.NEON_DOCKERHUB_USERNAME }}
+        password: ${{ secrets.NEON_DOCKERHUB_PASSWORD }}
+      options: --init
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          ref: ${{ matrix.ref }}
+
+      - name: Check rust licenses/bans/advisories/sources
+        env:
+          CARGO_DENY_TARGET: >-
+            ${{ github.event_name == 'schedule' && 'advisories' || 'all' }}
+        run: cargo deny check --hide-inclusion-graph $CARGO_DENY_TARGET
+
+      - name: Post to a Slack channel
+        if: ${{ github.event_name == 'schedule' && failure() }}
+        uses: slackapi/slack-github-action@v2
+        with:
+          method: chat.postMessage
+          token: ${{ secrets.SLACK_BOT_TOKEN }}
+          payload: |
+            channel: ${{ vars.SLACK_CICD_CHANNEL_ID }}
+            text: |
+              Periodic cargo-deny on ${{ matrix.ref }}: ${{ job.status }}
+              <${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}|GitHub Run>
+              Pinging @oncall-devprod.

.github/workflows/ingest_benchmark.yml

@@ -28,7 +28,24 @@ jobs:
     strategy:
       fail-fast: false # allow other variants to continue even if one fails
       matrix:
-        target_project: [new_empty_project, large_existing_project]
+        include:
+          - target_project: new_empty_project_stripe_size_2048 
+            stripe_size: 2048 # 16 MiB
+            postgres_version: 16
+          - target_project: new_empty_project_stripe_size_32768
+            stripe_size: 32768 # 256 MiB # note that this is different from null because using null will shard_split the project only if it reaches the threshold
+                               # while here it is sharded from the beginning with a shard size of 256 MiB
+            postgres_version: 16
+          - target_project: new_empty_project
+            stripe_size: null # run with neon defaults which will shard split only when reaching the threshold
+            postgres_version: 16
+          - target_project: new_empty_project
+            stripe_size: null # run with neon defaults which will shard split only when reaching the threshold
+            postgres_version: 17
+          - target_project: large_existing_project
+            stripe_size: null # cannot re-shared or choose different stripe size for existing, already sharded project
+            postgres_version: 16
+      max-parallel: 1 # we want to run each stripe size sequentially to be able to compare the results
     permissions:
       contents: write
       statuses: write
@@ -67,17 +84,21 @@ jobs:
         aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
 
     - name: Create Neon Project
-      if: ${{ matrix.target_project == 'new_empty_project' }}
+      if: ${{ startsWith(matrix.target_project, 'new_empty_project') }}
       id: create-neon-project-ingest-target
       uses: ./.github/actions/neon-project-create
       with:
         region_id: aws-us-east-2
-        postgres_version: 16
+        postgres_version: ${{ matrix.postgres_version }}
         compute_units: '[7, 7]' # we want to test large compute here to avoid compute-side bottleneck
         api_key: ${{ secrets.NEON_STAGING_API_KEY }}
+        shard_split_project: ${{ matrix.stripe_size != null && 'true' || 'false' }}
+        admin_api_key: ${{ secrets.NEON_STAGING_ADMIN_API_KEY }} 
+        shard_count: 8
+        stripe_size: ${{ matrix.stripe_size }}
 
     - name: Initialize Neon project
-      if: ${{ matrix.target_project == 'new_empty_project' }}
+      if: ${{ startsWith(matrix.target_project, 'new_empty_project') }}
       env:
           BENCHMARK_INGEST_TARGET_CONNSTR: ${{ steps.create-neon-project-ingest-target.outputs.dsn }}
           NEW_PROJECT_ID: ${{ steps.create-neon-project-ingest-target.outputs.project_id }}
@@ -130,7 +151,7 @@ jobs:
         test_selection: performance/test_perf_ingest_using_pgcopydb.py
         run_in_parallel: false
         extra_params: -s -m remote_cluster --timeout 86400 -k test_ingest_performance_using_pgcopydb
-        pg_version: v16
+        pg_version: v${{ matrix.postgres_version }}
         save_perf_report: true
         aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
       env:
@@ -146,7 +167,7 @@ jobs:
         ${PSQL} "${BENCHMARK_INGEST_TARGET_CONNSTR}" -c "\dt+"
 
     - name: Delete Neon Project
-      if: ${{ always() && matrix.target_project == 'new_empty_project' }}
+      if: ${{ always() && startsWith(matrix.target_project, 'new_empty_project') }}
       uses: ./.github/actions/neon-project-delete
       with:
         project_id: ${{ steps.create-neon-project-ingest-target.outputs.project_id }}

.github/workflows/neon_extra_builds.yml

@@ -114,7 +114,7 @@ jobs:
         run: make walproposer-lib -j$(nproc)
 
       - name: Produce the build stats
-        run: PQ_LIB_DIR=$(pwd)/pg_install/v17/lib cargo build --all --release --timings -j$(nproc)
+        run: cargo build --all --release --timings -j$(nproc)
 
       - name: Configure AWS credentials
         uses: aws-actions/configure-aws-credentials@v4

.github/workflows/pg-clients.yml

@@ -12,8 +12,8 @@ on:
   pull_request:
     paths:
       - '.github/workflows/pg-clients.yml'
-      - 'test_runner/pg_clients/**'
-      - 'test_runner/logical_repl/**'
+      - 'test_runner/pg_clients/**/*.py'
+      - 'test_runner/logical_repl/**/*.py'
       - 'poetry.lock'
   workflow_dispatch:
 
@@ -104,6 +104,8 @@ jobs:
         with:
           api_key: ${{ secrets.NEON_STAGING_API_KEY }}
           postgres_version: ${{ env.DEFAULT_PG_VERSION }}
+          project_settings: >-
+            {"enable_logical_replication": true}
 
       - name: Run tests
         uses: ./.github/actions/run-python-test-set

.github/workflows/pre-merge-checks.yml

@@ -59,7 +59,10 @@ jobs:
           echo "${RUST_CHANGED_FILES}"
 
   build-build-tools-image:
-    if: needs.get-changed-files.outputs.python-changed == 'true'
+    if: |
+      false
+      || needs.get-changed-files.outputs.python-changed == 'true'
+      || needs.get-changed-files.outputs.rust-changed == 'true'
     needs: [ get-changed-files ]
     uses: ./.github/workflows/build-build-tools-image.yml
     with:
@@ -92,7 +95,8 @@ jobs:
   # - conclusion
   # - neon-cloud-e2e
   conclusion:
-    if: always()
+    # Do not run job on Pull Requests as it interferes with the `conclusion` job from the `build_and_test` workflow
+    if: always() && github.event_name == 'merge_group'
     permissions:
       statuses: write # for `github.repos.createCommitStatus(...)`
       contents: write
@@ -124,6 +128,8 @@ jobs:
       - name: Fail the job if any of the dependencies do not succeed or skipped
         run: exit 1
         if: |
-          (contains(needs.check-codestyle-python.result, 'skipped') && needs.get-changed-files.outputs.python-changed == 'true')
+          false
+          || (needs.check-codestyle-python.result == 'skipped' && needs.get-changed-files.outputs.python-changed == 'true')
+          || (needs.check-codestyle-rust.result   == 'skipped' && needs.get-changed-files.outputs.rust-changed   == 'true')
           || contains(needs.*.result, 'failure')
           || contains(needs.*.result, 'cancelled')

.github/workflows/release.yml

@@ -3,8 +3,9 @@ name: Create Release Branch
 on:
   schedule:
     # It should be kept in sync with if-condition in jobs
-    - cron: '0 6 * * FRI' # Storage release
     - cron: '0 6 * * THU' # Proxy release
+    - cron: '0 6 * * FRI' # Storage release
+    - cron: '0 7 * * FRI' # Compute release
   workflow_dispatch:
     inputs:
       create-storage-release-branch:
@@ -55,7 +56,7 @@ jobs:
       ci-access-token: ${{ secrets.CI_ACCESS_TOKEN }}
 
   create-compute-release-branch:
-    if: inputs.create-compute-release-branch
+    if: ${{ github.event.schedule == '0 7 * * FRI' || inputs.create-compute-release-branch }}
 
     permissions:
       contents: write

Cargo.toml

@@ -54,6 +54,7 @@ async-compression = { version = "0.4.0", features = ["tokio", "gzip", "zstd"] }
 atomic-take = "1.1.0"
 backtrace = "0.3.74"
 flate2 = "1.0.26"
+assert-json-diff = "2"
 async-stream = "0.3"
 async-trait = "0.1"
 aws-config = { version = "1.5", default-features = false, features=["rustls", "sso"] }
@@ -65,7 +66,7 @@ aws-smithy-types = "1.2"
 aws-credential-types = "1.2.0"
 aws-sigv4 = { version = "1.2", features = ["sign-http"] }
 aws-types = "1.3"
-axum = { version = "0.7.9", features = ["ws"] }
+axum = { version = "0.8.1", features = ["ws"] }
 base64 = "0.13.0"
 bincode = "1.3"
 bindgen = "0.70"
@@ -77,10 +78,10 @@ camino = "1.1.6"
 cfg-if = "1.0.0"
 chrono = { version = "0.4", default-features = false, features = ["clock"] }
 clap = { version = "4.0", features = ["derive", "env"] }
+clashmap = { version = "1.0", features = ["raw-api"] }
 comfy-table = "7.1"
 const_format = "0.2"
 crc32c = "0.6"
-dashmap = { version = "5.5.0", features = ["raw-api"] }
 diatomic-waker = { version = "0.2.3" }
 either = "1.8"
 enum-map = "2.4.2"
@@ -123,7 +124,7 @@ measured = { version = "0.0.22", features=["lasso"] }
 measured-process = { version = "0.0.22" }
 memoffset = "0.9"
 nix = { version = "0.27", features = ["dir", "fs", "process", "socket", "signal", "poll"] }
-notify = "6.0.0"
+notify = "8.0.0"
 num_cpus = "1.15"
 num-traits = "0.2.15"
 once_cell = "1.13"
@@ -177,7 +178,7 @@ test-context = "0.3"
 thiserror = "1.0"
 tikv-jemallocator = { version = "0.6", features = ["profiling", "stats", "unprefixed_malloc_on_supported_platforms"] }
 tikv-jemalloc-ctl = { version = "0.6", features = ["stats"] }
-tokio = { version = "1.17", features = ["macros"] }
+tokio = { version = "1.41", features = ["macros"] }
 tokio-epoll-uring = { git = "https://github.com/neondatabase/tokio-epoll-uring.git" , branch = "main" }
 tokio-io-timeout = "1.2.0"
 tokio-postgres-rustls = "0.12.0"
@@ -187,13 +188,15 @@ tokio-tar = "0.3"
 tokio-util = { version = "0.7.10", features = ["io", "rt"] }
 toml = "0.8"
 toml_edit = "0.22"
-tonic = {version = "0.12.3", features = ["tls", "tls-roots"]}
+tonic = {version = "0.12.3", default-features = false, features = ["channel", "tls", "tls-roots"]}
 tower = { version = "0.5.2", default-features = false }
 tower-http = { version = "0.6.2", features = ["request-id", "trace"] }
 tower-service = "0.3.3"
 tracing = "0.1"
 tracing-error = "0.2"
+tracing-log = "0.2"
 tracing-opentelemetry = "0.28"
+tracing-serde = "0.2.0"
 tracing-subscriber = { version = "0.3", default-features = false, features = ["smallvec", "fmt", "tracing-log", "std", "env-filter", "json"] }
 try-lock = "0.2.5"
 twox-hash = { version = "1.6.3", default-features = false }

Dockerfile

@@ -45,7 +45,7 @@ COPY --chown=nonroot . .
 
 ARG ADDITIONAL_RUSTFLAGS
 RUN set -e \
-    && PQ_LIB_DIR=$(pwd)/pg_install/v${STABLE_PG_VERSION}/lib RUSTFLAGS="-Clinker=clang -Clink-arg=-fuse-ld=mold -Clink-arg=-Wl,--no-rosegment -Cforce-frame-pointers=yes ${ADDITIONAL_RUSTFLAGS}" cargo build \
+    && RUSTFLAGS="-Clinker=clang -Clink-arg=-fuse-ld=mold -Clink-arg=-Wl,--no-rosegment -Cforce-frame-pointers=yes ${ADDITIONAL_RUSTFLAGS}" cargo build \
       --bin pg_sni_router  \
       --bin pageserver  \
       --bin pagectl  \
@@ -64,6 +64,7 @@ ARG DEFAULT_PG_VERSION
 WORKDIR /data
 
 RUN set -e \
+    && echo 'Acquire::Retries "5";' > /etc/apt/apt.conf.d/80-retries \
     && apt update \
     && apt install -y \
         libreadline-dev \
@@ -72,6 +73,7 @@ RUN set -e \
 	# System postgres for use with client libraries (e.g. in storage controller)
         postgresql-15 \
         openssl \
+    && rm -f /etc/apt/apt.conf.d/80-retries \
     && rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/* \
     && useradd -d /data neon \
     && chown -R neon:neon /data

Makefile

@@ -64,8 +64,6 @@ CARGO_BUILD_FLAGS += $(filter -j1,$(MAKEFLAGS))
 CARGO_CMD_PREFIX += $(if $(filter n,$(MAKEFLAGS)),,+)
 # Force cargo not to print progress bar
 CARGO_CMD_PREFIX += CARGO_TERM_PROGRESS_WHEN=never CI=1
-# Set PQ_LIB_DIR to make sure `storage_controller` get linked with bundled libpq (through diesel)
-CARGO_CMD_PREFIX += PQ_LIB_DIR=$(POSTGRES_INSTALL_DIR)/v16/lib
 
 CACHEDIR_TAG_CONTENTS := "Signature: 8a477f597d28d172789f06886806bc55"
 

README.md

@@ -21,8 +21,10 @@ The Neon storage engine consists of two major components:
 
 See developer documentation in [SUMMARY.md](/docs/SUMMARY.md) for more information.
 
-## Running local installation
+## Running a local development environment
 
+Neon can be run on a workstation for small experiments and to test code changes, by
+following these instructions.
 
 #### Installing dependencies on Linux
 1. Install build dependencies and other applicable packages
@@ -238,7 +240,7 @@ postgres=# select * from t;
 > cargo neon stop
 ```
 
-More advanced usages can be found at [Control Plane and Neon Local](./control_plane/README.md).
+More advanced usages can be found at [Local Development Control Plane (`neon_local`))](./control_plane/README.md).
 
 #### Handling build failures
 

build-tools.Dockerfile

@@ -3,6 +3,15 @@ ARG DEBIAN_VERSION=bookworm
 FROM debian:bookworm-slim AS pgcopydb_builder
 ARG DEBIAN_VERSION
 
+# Use strict mode for bash to catch errors early
+SHELL ["/bin/bash", "-euo", "pipefail", "-c"]
+
+# By default, /bin/sh used in debian images will treat '\n' as eol,
+# but as we use bash as SHELL, and built-in echo in bash requires '-e' flag for that.
+RUN echo 'Acquire::Retries "5";' > /etc/apt/apt.conf.d/80-retries && \
+    echo -e "retry_connrefused = on\ntimeout=15\ntries=5\n" > /root/.wgetrc && \
+    echo -e "--retry-connrefused\n--connect-timeout 15\n--retry 5\n--max-time 300\n" > /root/.curlrc
+
 RUN if [ "${DEBIAN_VERSION}" = "bookworm" ]; then \
         set -e && \
         apt update && \
@@ -51,7 +60,8 @@ ARG DEBIAN_VERSION
 
 # Add nonroot user
 RUN useradd -ms /bin/bash nonroot -b /home
-SHELL ["/bin/bash", "-c"]
+# Use strict mode for bash to catch errors early
+SHELL ["/bin/bash", "-euo", "pipefail", "-c"]
 
 RUN mkdir -p /pgcopydb/bin && \
     mkdir -p /pgcopydb/lib && \
@@ -61,6 +71,10 @@ RUN mkdir -p /pgcopydb/bin && \
 COPY --from=pgcopydb_builder /usr/lib/postgresql/16/bin/pgcopydb /pgcopydb/bin/pgcopydb
 COPY --from=pgcopydb_builder /pgcopydb/lib/libpq.so.5 /pgcopydb/lib/libpq.so.5
 
+RUN echo 'Acquire::Retries "5";' > /etc/apt/apt.conf.d/80-retries && \
+    echo -e "retry_connrefused = on\ntimeout=15\ntries=5\n" > /root/.wgetrc && \
+    echo -e "--retry-connrefused\n--connect-timeout 15\n--retry 5\n--max-time 300\n" > /root/.curlrc
+
 # System deps
 #
 # 'gdb' is included so that we get backtraces of core dumps produced in
@@ -182,8 +196,14 @@ RUN set -e \
 # It includes several bug fixes on top on v2.0 release (https://github.com/linux-test-project/lcov/compare/v2.0...master)
 # And patches from us:
 # - Generates json file with code coverage summary (https://github.com/neondatabase/lcov/commit/426e7e7a22f669da54278e9b55e6d8caabd00af0.tar.gz)
-RUN for package in Capture::Tiny DateTime Devel::Cover Digest::MD5 File::Spec JSON::XS Memory::Process Time::HiRes JSON; do yes | perl -MCPAN -e "CPAN::Shell->notest('install', '$package')"; done \
-    && wget https://github.com/neondatabase/lcov/archive/426e7e7a22f669da54278e9b55e6d8caabd00af0.tar.gz -O lcov.tar.gz \
+RUN set +o pipefail && \
+	 for package in Capture::Tiny DateTime Devel::Cover Digest::MD5 File::Spec JSON::XS Memory::Process Time::HiRes JSON; do \
+		yes | perl -MCPAN -e "CPAN::Shell->notest('install', '$package')";\
+	 done && \
+	set -o pipefail
+# Split into separate step to debug flaky failures here
+RUN wget https://github.com/neondatabase/lcov/archive/426e7e7a22f669da54278e9b55e6d8caabd00af0.tar.gz -O lcov.tar.gz \
+    && ls -laht lcov.tar.gz && sha256sum lcov.tar.gz \
     && echo "61a22a62e20908b8b9e27d890bd0ea31f567a7b9668065589266371dcbca0992  lcov.tar.gz" | sha256sum --check \
     && mkdir -p lcov && tar -xzf lcov.tar.gz -C lcov --strip-components=1 \
     && cd lcov \
@@ -218,6 +238,8 @@ RUN wget -O /tmp/libicu-${ICU_VERSION}.tgz https://github.com/unicode-org/icu/re
 USER nonroot:nonroot
 WORKDIR /home/nonroot
 
+RUN echo -e "--retry-connrefused\n--connect-timeout 15\n--retry 5\n--max-time 300\n" > /home/nonroot/.curlrc
+
 # Python
 ENV PYTHON_VERSION=3.11.10 \
     PYENV_ROOT=/home/nonroot/.pyenv \
@@ -243,7 +265,7 @@ WORKDIR /home/nonroot
 
 # Rust
 # Please keep the version of llvm (installed above) in sync with rust llvm (`rustc --version --verbose | grep LLVM`)
-ENV RUSTC_VERSION=1.84.0
+ENV RUSTC_VERSION=1.84.1
 ENV RUSTUP_HOME="/home/nonroot/.rustup"
 ENV PATH="/home/nonroot/.cargo/bin:${PATH}"
 ARG RUSTFILT_VERSION=0.2.1
@@ -251,6 +273,7 @@ ARG CARGO_HAKARI_VERSION=0.9.33
 ARG CARGO_DENY_VERSION=0.16.2
 ARG CARGO_HACK_VERSION=0.6.33
 ARG CARGO_NEXTEST_VERSION=0.9.85
+ARG CARGO_DIESEL_CLI_VERSION=2.2.6
 RUN curl -sSO https://static.rust-lang.org/rustup/dist/$(uname -m)-unknown-linux-gnu/rustup-init && whoami && \
 	chmod +x rustup-init && \
 	./rustup-init -y --default-toolchain ${RUSTC_VERSION} && \
@@ -264,6 +287,8 @@ RUN curl -sSO https://static.rust-lang.org/rustup/dist/$(uname -m)-unknown-linux
     cargo install cargo-deny --locked --version ${CARGO_DENY_VERSION} && \
     cargo install cargo-hack          --version ${CARGO_HACK_VERSION} && \
     cargo install cargo-nextest       --version ${CARGO_NEXTEST_VERSION} && \
+    cargo install diesel_cli          --version ${CARGO_DIESEL_CLI_VERSION} \
+                                      --features postgres-bundled --no-default-features && \
     rm -rf /home/nonroot/.cargo/registry && \
     rm -rf /home/nonroot/.cargo/git
 

compute/etc/pgbouncer.ini

@@ -19,6 +19,8 @@ max_prepared_statements=0
 admin_users=postgres
 unix_socket_dir=/tmp/
 unix_socket_mode=0777
+; required for pgbouncer_exporter
+ignore_startup_parameters=extra_float_digits
 
 ;; Disable connection logging. It produces a lot of logs that no one looks at,
 ;; and we can get similar log entries from the proxy too. We had incidents in

compute/patches/contrib_pg16.patch

@@ -0,0 +1,242 @@
+diff --git a/contrib/amcheck/expected/check_heap.out b/contrib/amcheck/expected/check_heap.out
+index 979e5e8..2375b45 100644
+--- a/contrib/amcheck/expected/check_heap.out
++++ b/contrib/amcheck/expected/check_heap.out
+@@ -80,12 +80,9 @@ INSERT INTO heaptest (a, b)
+ -- same transaction.  The heaptest table is smaller than the default
+ -- wal_skip_threshold, so a wal_level=minimal commit reads the table into
+ -- shared_buffers.  A transaction delays that and excludes any autovacuum.
+-SET allow_in_place_tablespaces = true;
+-CREATE TABLESPACE regress_test_stats_tblspc LOCATION '';
+ SELECT sum(reads) AS stats_bulkreads_before
+   FROM pg_stat_io WHERE context = 'bulkread' \gset
+ BEGIN;
+-ALTER TABLE heaptest SET TABLESPACE regress_test_stats_tblspc;
+ -- Check that valid options are not rejected nor corruption reported
+ -- for a non-empty table
+ SELECT * FROM verify_heapam(relation := 'heaptest', skip := 'none');
+@@ -118,14 +115,6 @@ SELECT pg_stat_force_next_flush();
+  
+ (1 row)
+ 
+-SELECT sum(reads) AS stats_bulkreads_after
+-  FROM pg_stat_io WHERE context = 'bulkread' \gset
+-SELECT :stats_bulkreads_after > :stats_bulkreads_before;
+- ?column? 
+-----------
+- t
+-(1 row)
+-
+ CREATE ROLE regress_heaptest_role;
+ -- verify permissions are checked (error due to function not callable)
+ SET ROLE regress_heaptest_role;
+@@ -233,7 +222,6 @@ ERROR:  cannot check relation "test_foreign_table"
+ DETAIL:  This operation is not supported for foreign tables.
+ -- cleanup
+ DROP TABLE heaptest;
+-DROP TABLESPACE regress_test_stats_tblspc;
+ DROP TABLE test_partition;
+ DROP TABLE test_partitioned;
+ DROP OWNED BY regress_heaptest_role; -- permissions
+diff --git a/contrib/amcheck/sql/check_heap.sql b/contrib/amcheck/sql/check_heap.sql
+index 1745bae..3b429c3 100644
+--- a/contrib/amcheck/sql/check_heap.sql
++++ b/contrib/amcheck/sql/check_heap.sql
+@@ -40,12 +40,9 @@ INSERT INTO heaptest (a, b)
+ -- same transaction.  The heaptest table is smaller than the default
+ -- wal_skip_threshold, so a wal_level=minimal commit reads the table into
+ -- shared_buffers.  A transaction delays that and excludes any autovacuum.
+-SET allow_in_place_tablespaces = true;
+-CREATE TABLESPACE regress_test_stats_tblspc LOCATION '';
+ SELECT sum(reads) AS stats_bulkreads_before
+   FROM pg_stat_io WHERE context = 'bulkread' \gset
+ BEGIN;
+-ALTER TABLE heaptest SET TABLESPACE regress_test_stats_tblspc;
+ -- Check that valid options are not rejected nor corruption reported
+ -- for a non-empty table
+ SELECT * FROM verify_heapam(relation := 'heaptest', skip := 'none');
+@@ -58,9 +55,6 @@ COMMIT;
+ --   ALTER TABLE ... SET TABLESPACE ...
+ -- causing an additional bulkread, which should be reflected in pg_stat_io.
+ SELECT pg_stat_force_next_flush();
+-SELECT sum(reads) AS stats_bulkreads_after
+-  FROM pg_stat_io WHERE context = 'bulkread' \gset
+-SELECT :stats_bulkreads_after > :stats_bulkreads_before;
+ 
+ CREATE ROLE regress_heaptest_role;
+ 
+@@ -140,7 +134,6 @@ SELECT * FROM verify_heapam('test_foreign_table',
+ 
+ -- cleanup
+ DROP TABLE heaptest;
+-DROP TABLESPACE regress_test_stats_tblspc;
+ DROP TABLE test_partition;
+ DROP TABLE test_partitioned;
+ DROP OWNED BY regress_heaptest_role; -- permissions
+diff --git a/contrib/citext/expected/create_index_acl.out b/contrib/citext/expected/create_index_acl.out
+index 33be13a..70a406c 100644
+--- a/contrib/citext/expected/create_index_acl.out
++++ b/contrib/citext/expected/create_index_acl.out
+@@ -5,9 +5,6 @@
+ -- owner having as few applicable privileges as possible.  (The privileges.sql
+ -- regress_sro_user tests look for the opposite defect; they confirm that
+ -- DefineIndex() uses the table owner userid where necessary.)
+-SET allow_in_place_tablespaces = true;
+-CREATE TABLESPACE regress_create_idx_tblspace LOCATION '';
+-RESET allow_in_place_tablespaces;
+ BEGIN;
+ CREATE ROLE regress_minimal;
+ CREATE SCHEMA s;
+@@ -49,11 +46,9 @@ ALTER TABLE s.x OWNER TO regress_minimal;
+ -- Empty-table DefineIndex()
+ CREATE UNIQUE INDEX u0rows ON s.x USING btree
+   ((s.index_this_expr(y, s.const())) COLLATE s.coll s.citext_pattern_ops)
+-  TABLESPACE regress_create_idx_tblspace
+   WHERE s.index_row_if(y);
+ ALTER TABLE s.x ADD CONSTRAINT e0rows EXCLUDE USING btree
+   ((s.index_this_expr(y, s.const())) COLLATE s.coll WITH s.=)
+-  USING INDEX TABLESPACE regress_create_idx_tblspace
+   WHERE (s.index_row_if(y));
+ -- Make the table nonempty.
+ INSERT INTO s.x VALUES ('foo'), ('bar');
+@@ -66,11 +61,9 @@ RESET search_path;
+ GRANT EXECUTE ON FUNCTION s.index_this_expr TO regress_minimal;
+ CREATE UNIQUE INDEX u2rows ON s.x USING btree
+   ((s.index_this_expr(y, s.const())) COLLATE s.coll s.citext_pattern_ops)
+-  TABLESPACE regress_create_idx_tblspace
+   WHERE s.index_row_if(y);
+ ALTER TABLE s.x ADD CONSTRAINT e2rows EXCLUDE USING btree
+   ((s.index_this_expr(y, s.const())) COLLATE s.coll WITH s.=)
+-  USING INDEX TABLESPACE regress_create_idx_tblspace
+   WHERE (s.index_row_if(y));
+ -- Shall not find s.coll via search_path, despite the s.const->public.setter
+ -- call having set search_path=s during expression planning.  Suppress the
+@@ -78,9 +71,7 @@ ALTER TABLE s.x ADD CONSTRAINT e2rows EXCLUDE USING btree
+ \set VERBOSITY sqlstate
+ ALTER TABLE s.x ADD CONSTRAINT underqualified EXCLUDE USING btree
+   ((s.index_this_expr(y, s.const())) COLLATE coll WITH s.=)
+-  USING INDEX TABLESPACE regress_create_idx_tblspace
+   WHERE (s.index_row_if(y));
+ ERROR:  42704
+ \set VERBOSITY default
+ ROLLBACK;
+-DROP TABLESPACE regress_create_idx_tblspace;
+diff --git a/contrib/citext/sql/create_index_acl.sql b/contrib/citext/sql/create_index_acl.sql
+index 10b5225..ae442e1 100644
+--- a/contrib/citext/sql/create_index_acl.sql
++++ b/contrib/citext/sql/create_index_acl.sql
+@@ -6,10 +6,6 @@
+ -- regress_sro_user tests look for the opposite defect; they confirm that
+ -- DefineIndex() uses the table owner userid where necessary.)
+ 
+-SET allow_in_place_tablespaces = true;
+-CREATE TABLESPACE regress_create_idx_tblspace LOCATION '';
+-RESET allow_in_place_tablespaces;
+-
+ BEGIN;
+ CREATE ROLE regress_minimal;
+ CREATE SCHEMA s;
+@@ -51,11 +47,9 @@ ALTER TABLE s.x OWNER TO regress_minimal;
+ -- Empty-table DefineIndex()
+ CREATE UNIQUE INDEX u0rows ON s.x USING btree
+   ((s.index_this_expr(y, s.const())) COLLATE s.coll s.citext_pattern_ops)
+-  TABLESPACE regress_create_idx_tblspace
+   WHERE s.index_row_if(y);
+ ALTER TABLE s.x ADD CONSTRAINT e0rows EXCLUDE USING btree
+   ((s.index_this_expr(y, s.const())) COLLATE s.coll WITH s.=)
+-  USING INDEX TABLESPACE regress_create_idx_tblspace
+   WHERE (s.index_row_if(y));
+ -- Make the table nonempty.
+ INSERT INTO s.x VALUES ('foo'), ('bar');
+@@ -68,11 +62,9 @@ RESET search_path;
+ GRANT EXECUTE ON FUNCTION s.index_this_expr TO regress_minimal;
+ CREATE UNIQUE INDEX u2rows ON s.x USING btree
+   ((s.index_this_expr(y, s.const())) COLLATE s.coll s.citext_pattern_ops)
+-  TABLESPACE regress_create_idx_tblspace
+   WHERE s.index_row_if(y);
+ ALTER TABLE s.x ADD CONSTRAINT e2rows EXCLUDE USING btree
+   ((s.index_this_expr(y, s.const())) COLLATE s.coll WITH s.=)
+-  USING INDEX TABLESPACE regress_create_idx_tblspace
+   WHERE (s.index_row_if(y));
+ -- Shall not find s.coll via search_path, despite the s.const->public.setter
+ -- call having set search_path=s during expression planning.  Suppress the
+@@ -80,9 +72,7 @@ ALTER TABLE s.x ADD CONSTRAINT e2rows EXCLUDE USING btree
+ \set VERBOSITY sqlstate
+ ALTER TABLE s.x ADD CONSTRAINT underqualified EXCLUDE USING btree
+   ((s.index_this_expr(y, s.const())) COLLATE coll WITH s.=)
+-  USING INDEX TABLESPACE regress_create_idx_tblspace
+   WHERE (s.index_row_if(y));
+ \set VERBOSITY default
+ ROLLBACK;
+ 
+-DROP TABLESPACE regress_create_idx_tblspace;
+diff --git a/contrib/file_fdw/expected/file_fdw.out b/contrib/file_fdw/expected/file_fdw.out
+index 72304e0..ebe131b 100644
+--- a/contrib/file_fdw/expected/file_fdw.out
++++ b/contrib/file_fdw/expected/file_fdw.out
+@@ -4,6 +4,7 @@
+ -- directory paths are passed to us in environment variables
+ \getenv abs_srcdir PG_ABS_SRCDIR
+ -- Clean up in case a prior regression run failed
++SET compute_query_id TO 'off';
+ SET client_min_messages TO 'warning';
+ DROP ROLE IF EXISTS regress_file_fdw_superuser, regress_file_fdw_user, regress_no_priv_user;
+ RESET client_min_messages;
+diff --git a/contrib/file_fdw/sql/file_fdw.sql b/contrib/file_fdw/sql/file_fdw.sql
+index f0548e1..848a08c 100644
+--- a/contrib/file_fdw/sql/file_fdw.sql
++++ b/contrib/file_fdw/sql/file_fdw.sql
+@@ -6,6 +6,7 @@
+ \getenv abs_srcdir PG_ABS_SRCDIR
+ 
+ -- Clean up in case a prior regression run failed
++SET compute_query_id TO 'off';
+ SET client_min_messages TO 'warning';
+ DROP ROLE IF EXISTS regress_file_fdw_superuser, regress_file_fdw_user, regress_no_priv_user;
+ RESET client_min_messages;
+diff --git a/contrib/pageinspect/expected/gist.out b/contrib/pageinspect/expected/gist.out
+index d1adbab..38b52ac 100644
+--- a/contrib/pageinspect/expected/gist.out
++++ b/contrib/pageinspect/expected/gist.out
+@@ -10,25 +10,6 @@ BEGIN;
+ CREATE TABLE test_gist AS SELECT point(i,i) p, i::text t FROM
+     generate_series(1,1000) i;
+ CREATE INDEX test_gist_idx ON test_gist USING gist (p);
+--- Page 0 is the root, the rest are leaf pages
+-SELECT * FROM gist_page_opaque_info(get_raw_page('test_gist_idx', 0));
+- lsn | nsn | rightlink  | flags 
+------+-----+------------+-------
+- 0/1 | 0/0 | 4294967295 | {}
+-(1 row)
+-
+-SELECT * FROM gist_page_opaque_info(get_raw_page('test_gist_idx', 1));
+- lsn | nsn | rightlink  | flags  
+------+-----+------------+--------
+- 0/1 | 0/0 | 4294967295 | {leaf}
+-(1 row)
+-
+-SELECT * FROM gist_page_opaque_info(get_raw_page('test_gist_idx', 2));
+- lsn | nsn | rightlink | flags  
+------+-----+-----------+--------
+- 0/1 | 0/0 |         1 | {leaf}
+-(1 row)
+-
+ COMMIT;
+ SELECT * FROM gist_page_items(get_raw_page('test_gist_idx', 0), 'test_gist_idx');
+  itemoffset |   ctid    | itemlen | dead |             keys              
+diff --git a/contrib/pageinspect/sql/gist.sql b/contrib/pageinspect/sql/gist.sql
+index d263542..607992f 100644
+--- a/contrib/pageinspect/sql/gist.sql
++++ b/contrib/pageinspect/sql/gist.sql
+@@ -12,11 +12,6 @@ CREATE TABLE test_gist AS SELECT point(i,i) p, i::text t FROM
+     generate_series(1,1000) i;
+ CREATE INDEX test_gist_idx ON test_gist USING gist (p);
+ 
+--- Page 0 is the root, the rest are leaf pages
+-SELECT * FROM gist_page_opaque_info(get_raw_page('test_gist_idx', 0));
+-SELECT * FROM gist_page_opaque_info(get_raw_page('test_gist_idx', 1));
+-SELECT * FROM gist_page_opaque_info(get_raw_page('test_gist_idx', 2));
+-
+ COMMIT;
+ 
+ SELECT * FROM gist_page_items(get_raw_page('test_gist_idx', 0), 'test_gist_idx');

compute/patches/contrib_pg17.patch

@@ -0,0 +1,196 @@
+diff --git a/contrib/amcheck/expected/check_heap.out b/contrib/amcheck/expected/check_heap.out
+index 979e5e8..2375b45 100644
+--- a/contrib/amcheck/expected/check_heap.out
++++ b/contrib/amcheck/expected/check_heap.out
+@@ -80,12 +80,9 @@ INSERT INTO heaptest (a, b)
+ -- same transaction.  The heaptest table is smaller than the default
+ -- wal_skip_threshold, so a wal_level=minimal commit reads the table into
+ -- shared_buffers.  A transaction delays that and excludes any autovacuum.
+-SET allow_in_place_tablespaces = true;
+-CREATE TABLESPACE regress_test_stats_tblspc LOCATION '';
+ SELECT sum(reads) AS stats_bulkreads_before
+   FROM pg_stat_io WHERE context = 'bulkread' \gset
+ BEGIN;
+-ALTER TABLE heaptest SET TABLESPACE regress_test_stats_tblspc;
+ -- Check that valid options are not rejected nor corruption reported
+ -- for a non-empty table
+ SELECT * FROM verify_heapam(relation := 'heaptest', skip := 'none');
+@@ -118,14 +115,6 @@ SELECT pg_stat_force_next_flush();
+  
+ (1 row)
+ 
+-SELECT sum(reads) AS stats_bulkreads_after
+-  FROM pg_stat_io WHERE context = 'bulkread' \gset
+-SELECT :stats_bulkreads_after > :stats_bulkreads_before;
+- ?column? 
+-----------
+- t
+-(1 row)
+-
+ CREATE ROLE regress_heaptest_role;
+ -- verify permissions are checked (error due to function not callable)
+ SET ROLE regress_heaptest_role;
+@@ -233,7 +222,6 @@ ERROR:  cannot check relation "test_foreign_table"
+ DETAIL:  This operation is not supported for foreign tables.
+ -- cleanup
+ DROP TABLE heaptest;
+-DROP TABLESPACE regress_test_stats_tblspc;
+ DROP TABLE test_partition;
+ DROP TABLE test_partitioned;
+ DROP OWNED BY regress_heaptest_role; -- permissions
+diff --git a/contrib/amcheck/sql/check_heap.sql b/contrib/amcheck/sql/check_heap.sql
+index 1745bae..3b429c3 100644
+--- a/contrib/amcheck/sql/check_heap.sql
++++ b/contrib/amcheck/sql/check_heap.sql
+@@ -40,12 +40,9 @@ INSERT INTO heaptest (a, b)
+ -- same transaction.  The heaptest table is smaller than the default
+ -- wal_skip_threshold, so a wal_level=minimal commit reads the table into
+ -- shared_buffers.  A transaction delays that and excludes any autovacuum.
+-SET allow_in_place_tablespaces = true;
+-CREATE TABLESPACE regress_test_stats_tblspc LOCATION '';
+ SELECT sum(reads) AS stats_bulkreads_before
+   FROM pg_stat_io WHERE context = 'bulkread' \gset
+ BEGIN;
+-ALTER TABLE heaptest SET TABLESPACE regress_test_stats_tblspc;
+ -- Check that valid options are not rejected nor corruption reported
+ -- for a non-empty table
+ SELECT * FROM verify_heapam(relation := 'heaptest', skip := 'none');
+@@ -58,9 +55,6 @@ COMMIT;
+ --   ALTER TABLE ... SET TABLESPACE ...
+ -- causing an additional bulkread, which should be reflected in pg_stat_io.
+ SELECT pg_stat_force_next_flush();
+-SELECT sum(reads) AS stats_bulkreads_after
+-  FROM pg_stat_io WHERE context = 'bulkread' \gset
+-SELECT :stats_bulkreads_after > :stats_bulkreads_before;
+ 
+ CREATE ROLE regress_heaptest_role;
+ 
+@@ -140,7 +134,6 @@ SELECT * FROM verify_heapam('test_foreign_table',
+ 
+ -- cleanup
+ DROP TABLE heaptest;
+-DROP TABLESPACE regress_test_stats_tblspc;
+ DROP TABLE test_partition;
+ DROP TABLE test_partitioned;
+ DROP OWNED BY regress_heaptest_role; -- permissions
+diff --git a/contrib/citext/expected/create_index_acl.out b/contrib/citext/expected/create_index_acl.out
+index 33be13a..70a406c 100644
+--- a/contrib/citext/expected/create_index_acl.out
++++ b/contrib/citext/expected/create_index_acl.out
+@@ -5,9 +5,6 @@
+ -- owner having as few applicable privileges as possible.  (The privileges.sql
+ -- regress_sro_user tests look for the opposite defect; they confirm that
+ -- DefineIndex() uses the table owner userid where necessary.)
+-SET allow_in_place_tablespaces = true;
+-CREATE TABLESPACE regress_create_idx_tblspace LOCATION '';
+-RESET allow_in_place_tablespaces;
+ BEGIN;
+ CREATE ROLE regress_minimal;
+ CREATE SCHEMA s;
+@@ -49,11 +46,9 @@ ALTER TABLE s.x OWNER TO regress_minimal;
+ -- Empty-table DefineIndex()
+ CREATE UNIQUE INDEX u0rows ON s.x USING btree
+   ((s.index_this_expr(y, s.const())) COLLATE s.coll s.citext_pattern_ops)
+-  TABLESPACE regress_create_idx_tblspace
+   WHERE s.index_row_if(y);
+ ALTER TABLE s.x ADD CONSTRAINT e0rows EXCLUDE USING btree
+   ((s.index_this_expr(y, s.const())) COLLATE s.coll WITH s.=)
+-  USING INDEX TABLESPACE regress_create_idx_tblspace
+   WHERE (s.index_row_if(y));
+ -- Make the table nonempty.
+ INSERT INTO s.x VALUES ('foo'), ('bar');
+@@ -66,11 +61,9 @@ RESET search_path;
+ GRANT EXECUTE ON FUNCTION s.index_this_expr TO regress_minimal;
+ CREATE UNIQUE INDEX u2rows ON s.x USING btree
+   ((s.index_this_expr(y, s.const())) COLLATE s.coll s.citext_pattern_ops)
+-  TABLESPACE regress_create_idx_tblspace
+   WHERE s.index_row_if(y);
+ ALTER TABLE s.x ADD CONSTRAINT e2rows EXCLUDE USING btree
+   ((s.index_this_expr(y, s.const())) COLLATE s.coll WITH s.=)
+-  USING INDEX TABLESPACE regress_create_idx_tblspace
+   WHERE (s.index_row_if(y));
+ -- Shall not find s.coll via search_path, despite the s.const->public.setter
+ -- call having set search_path=s during expression planning.  Suppress the
+@@ -78,9 +71,7 @@ ALTER TABLE s.x ADD CONSTRAINT e2rows EXCLUDE USING btree
+ \set VERBOSITY sqlstate
+ ALTER TABLE s.x ADD CONSTRAINT underqualified EXCLUDE USING btree
+   ((s.index_this_expr(y, s.const())) COLLATE coll WITH s.=)
+-  USING INDEX TABLESPACE regress_create_idx_tblspace
+   WHERE (s.index_row_if(y));
+ ERROR:  42704
+ \set VERBOSITY default
+ ROLLBACK;
+-DROP TABLESPACE regress_create_idx_tblspace;
+diff --git a/contrib/citext/sql/create_index_acl.sql b/contrib/citext/sql/create_index_acl.sql
+index 10b5225..ae442e1 100644
+--- a/contrib/citext/sql/create_index_acl.sql
++++ b/contrib/citext/sql/create_index_acl.sql
+@@ -6,10 +6,6 @@
+ -- regress_sro_user tests look for the opposite defect; they confirm that
+ -- DefineIndex() uses the table owner userid where necessary.)
+ 
+-SET allow_in_place_tablespaces = true;
+-CREATE TABLESPACE regress_create_idx_tblspace LOCATION '';
+-RESET allow_in_place_tablespaces;
+-
+ BEGIN;
+ CREATE ROLE regress_minimal;
+ CREATE SCHEMA s;
+@@ -51,11 +47,9 @@ ALTER TABLE s.x OWNER TO regress_minimal;
+ -- Empty-table DefineIndex()
+ CREATE UNIQUE INDEX u0rows ON s.x USING btree
+   ((s.index_this_expr(y, s.const())) COLLATE s.coll s.citext_pattern_ops)
+-  TABLESPACE regress_create_idx_tblspace
+   WHERE s.index_row_if(y);
+ ALTER TABLE s.x ADD CONSTRAINT e0rows EXCLUDE USING btree
+   ((s.index_this_expr(y, s.const())) COLLATE s.coll WITH s.=)
+-  USING INDEX TABLESPACE regress_create_idx_tblspace
+   WHERE (s.index_row_if(y));
+ -- Make the table nonempty.
+ INSERT INTO s.x VALUES ('foo'), ('bar');
+@@ -68,11 +62,9 @@ RESET search_path;
+ GRANT EXECUTE ON FUNCTION s.index_this_expr TO regress_minimal;
+ CREATE UNIQUE INDEX u2rows ON s.x USING btree
+   ((s.index_this_expr(y, s.const())) COLLATE s.coll s.citext_pattern_ops)
+-  TABLESPACE regress_create_idx_tblspace
+   WHERE s.index_row_if(y);
+ ALTER TABLE s.x ADD CONSTRAINT e2rows EXCLUDE USING btree
+   ((s.index_this_expr(y, s.const())) COLLATE s.coll WITH s.=)
+-  USING INDEX TABLESPACE regress_create_idx_tblspace
+   WHERE (s.index_row_if(y));
+ -- Shall not find s.coll via search_path, despite the s.const->public.setter
+ -- call having set search_path=s during expression planning.  Suppress the
+@@ -80,9 +72,7 @@ ALTER TABLE s.x ADD CONSTRAINT e2rows EXCLUDE USING btree
+ \set VERBOSITY sqlstate
+ ALTER TABLE s.x ADD CONSTRAINT underqualified EXCLUDE USING btree
+   ((s.index_this_expr(y, s.const())) COLLATE coll WITH s.=)
+-  USING INDEX TABLESPACE regress_create_idx_tblspace
+   WHERE (s.index_row_if(y));
+ \set VERBOSITY default
+ ROLLBACK;
+ 
+-DROP TABLESPACE regress_create_idx_tblspace;
+diff --git a/contrib/file_fdw/expected/file_fdw.out b/contrib/file_fdw/expected/file_fdw.out
+index 86c148a..81bdb2c 100644
+--- a/contrib/file_fdw/expected/file_fdw.out
++++ b/contrib/file_fdw/expected/file_fdw.out
+@@ -4,6 +4,7 @@
+ -- directory paths are passed to us in environment variables
+ \getenv abs_srcdir PG_ABS_SRCDIR
+ -- Clean up in case a prior regression run failed
++SET compute_query_id TO 'off';
+ SET client_min_messages TO 'warning';
+ DROP ROLE IF EXISTS regress_file_fdw_superuser, regress_file_fdw_user, regress_no_priv_user;
+ RESET client_min_messages;
+diff --git a/contrib/file_fdw/sql/file_fdw.sql b/contrib/file_fdw/sql/file_fdw.sql
+index f0548e1..848a08c 100644
+--- a/contrib/file_fdw/sql/file_fdw.sql
++++ b/contrib/file_fdw/sql/file_fdw.sql
+@@ -6,6 +6,7 @@
+ \getenv abs_srcdir PG_ABS_SRCDIR
+ 
+ -- Clean up in case a prior regression run failed
++SET compute_query_id TO 'off';
+ SET client_min_messages TO 'warning';
+ DROP ROLE IF EXISTS regress_file_fdw_superuser, regress_file_fdw_user, regress_no_priv_user;
+ RESET client_min_messages;

compute/patches/pg_graphql.patch

@@ -0,0 +1,19 @@
+commit ec6a491d126882966a696f9ad5d3698935361d55
+Author: Alexey Masterov <alexeymasterov@neon.tech>
+Date:   Tue Dec 17 10:25:00 2024 +0100
+
+    Changes required to run tests on Neon
+
+diff --git a/test/expected/permissions_functions.out b/test/expected/permissions_functions.out
+index 1e9fbc2..94cbe25 100644
+--- a/test/expected/permissions_functions.out
++++ b/test/expected/permissions_functions.out
+@@ -64,7 +64,7 @@ begin;
+     select current_user;
+  current_user 
+ --------------
+- postgres
++ cloud_admin
+ (1 row)
+ 
+     -- revoke default access from the public role for new functions

compute/patches/pgvector.patch

@@ -1,8 +1,24 @@
+diff --git a/Makefile b/Makefile
+index 7a4b88c..56678af 100644
+--- a/Makefile
++++ b/Makefile
+@@ -3,7 +3,10 @@ EXTVERSION = 0.8.0
+ 
+ MODULE_big = vector
+ DATA = $(wildcard sql/*--*--*.sql)
+-DATA_built = sql/$(EXTENSION)--$(EXTVERSION).sql
++# This change is needed to install different per-version SQL files
++# like pgvector--0.8.0.sql and pgvector--0.7.4.sql
++# The corresponding file is downloaded during the Docker image build process
++DATA_built = sql/$(EXTENSION)--$(EXTVERSION).sql sql/vector--0.7.4.sql
+ OBJS = src/bitutils.o src/bitvec.o src/halfutils.o src/halfvec.o src/hnsw.o src/hnswbuild.o src/hnswinsert.o src/hnswscan.o src/hnswutils.o src/hnswvacuum.o src/ivfbuild.o src/ivfflat.o src/ivfinsert.o src/ivfkmeans.o src/ivfscan.o src/ivfutils.o src/ivfvacuum.o src/sparsevec.o src/vector.o
+ HEADERS = src/halfvec.h src/sparsevec.h src/vector.h
+ 
 diff --git a/src/hnswbuild.c b/src/hnswbuild.c
-index dcfb2bd..d5189ee 100644
+index b667478..fc1897c 100644
 --- a/src/hnswbuild.c
 +++ b/src/hnswbuild.c
-@@ -860,9 +860,17 @@ HnswParallelBuildMain(dsm_segment *seg, shm_toc *toc)
+@@ -843,9 +843,17 @@ HnswParallelBuildMain(dsm_segment *seg, shm_toc *toc)
  
  	hnswarea = shm_toc_lookup(toc, PARALLEL_KEY_HNSW_AREA, false);
  
@@ -20,7 +36,7 @@ index dcfb2bd..d5189ee 100644
  	/* Close relations within worker */
  	index_close(indexRel, indexLockmode);
  	table_close(heapRel, heapLockmode);
-@@ -1117,12 +1125,38 @@ BuildIndex(Relation heap, Relation index, IndexInfo *indexInfo,
+@@ -1100,12 +1108,38 @@ BuildIndex(Relation heap, Relation index, IndexInfo *indexInfo,
  	SeedRandom(42);
  #endif
  

compute/vm-image-spec-bookworm.yaml

@@ -27,6 +27,10 @@ commands:
     user: nobody
     sysvInitAction: respawn
     shell: 'DATA_SOURCE_NAME="user=cloud_admin sslmode=disable dbname=postgres application_name=postgres-exporter" /bin/postgres_exporter --config.file=/etc/postgres_exporter.yml'
+  - name: pgbouncer-exporter
+    user: postgres
+    sysvInitAction: respawn
+    shell: '/bin/pgbouncer_exporter --pgBouncer.connectionString="postgres:///pgbouncer?host=/tmp&port=6432&dbname=pgbouncer&user=pgbouncer"'
   - name: sql-exporter
     user: nobody
     sysvInitAction: respawn

compute/vm-image-spec-bullseye.yaml

@@ -27,6 +27,10 @@ commands:
     user: nobody
     sysvInitAction: respawn
     shell: 'DATA_SOURCE_NAME="user=cloud_admin sslmode=disable dbname=postgres application_name=postgres-exporter" /bin/postgres_exporter --config.file=/etc/postgres_exporter.yml'
+  - name: pgbouncer-exporter
+    user: postgres
+    sysvInitAction: respawn
+    shell: '/bin/pgbouncer_exporter --pgBouncer.connectionString="postgres:///pgbouncer?host=/tmp&port=6432&dbname=pgbouncer&user=pgbouncer"'
   - name: sql-exporter
     user: nobody
     sysvInitAction: respawn

compute_tools/src/bin/compute_ctl.rs

@@ -34,6 +34,7 @@
 //!             -r http://pg-ext-s3-gateway \
 //! ```
 use std::collections::HashMap;
+use std::ffi::OsString;
 use std::fs::File;
 use std::path::Path;
 use std::process::exit;
@@ -44,7 +45,7 @@ use std::{thread, time::Duration};
 
 use anyhow::{Context, Result};
 use chrono::Utc;
-use clap::Arg;
+use clap::Parser;
 use compute_tools::disk_quota::set_disk_quota;
 use compute_tools::lsn_lease::launch_lsn_lease_bg_task_for_static;
 use signal_hook::consts::{SIGQUIT, SIGTERM};
@@ -73,10 +74,75 @@ use utils::failpoint_support;
 // in-case of not-set environment var
 const BUILD_TAG_DEFAULT: &str = "latest";
 
-fn main() -> Result<()> {
-    let scenario = failpoint_support::init();
+// Compatibility hack: if the control plane specified any remote-ext-config
+// use the default value for extension storage proxy gateway.
+// Remove this once the control plane is updated to pass the gateway URL
+fn parse_remote_ext_config(arg: &str) -> Result<String> {
+    if arg.starts_with("http") {
+        Ok(arg.trim_end_matches('/').to_string())
+    } else {
+        Ok("http://pg-ext-s3-gateway".to_string())
+    }
+}
 
-    let (build_tag, clap_args) = init()?;
+#[derive(Parser)]
+#[command(rename_all = "kebab-case")]
+struct Cli {
+    #[arg(short = 'b', long, default_value = "postgres", env = "POSTGRES_PATH")]
+    pub pgbin: String,
+
+    #[arg(short = 'r', long, value_parser = parse_remote_ext_config)]
+    pub remote_ext_config: Option<String>,
+
+    #[arg(long, default_value_t = 3080)]
+    pub http_port: u16,
+
+    #[arg(short = 'D', long, value_name = "DATADIR")]
+    pub pgdata: String,
+
+    #[arg(short = 'C', long, value_name = "DATABASE_URL")]
+    pub connstr: String,
+
+    #[cfg(target_os = "linux")]
+    #[arg(long, default_value = "neon-postgres")]
+    pub cgroup: String,
+
+    #[cfg(target_os = "linux")]
+    #[arg(
+        long,
+        default_value = "host=localhost port=5432 dbname=postgres user=cloud_admin sslmode=disable application_name=vm-monitor"
+    )]
+    pub filecache_connstr: String,
+
+    #[cfg(target_os = "linux")]
+    #[arg(long, default_value = "0.0.0.0:10301")]
+    pub vm_monitor_addr: String,
+
+    #[arg(long, action = clap::ArgAction::SetTrue)]
+    pub resize_swap_on_bind: bool,
+
+    #[arg(long)]
+    pub set_disk_quota_for_fs: Option<String>,
+
+    #[arg(short = 's', long = "spec", group = "spec")]
+    pub spec_json: Option<String>,
+
+    #[arg(short = 'S', long, group = "spec-path")]
+    pub spec_path: Option<OsString>,
+
+    #[arg(short = 'i', long, group = "compute-id", conflicts_with_all = ["spec", "spec-path"])]
+    pub compute_id: Option<String>,
+
+    #[arg(short = 'p', long, conflicts_with_all = ["spec", "spec-path"], requires = "compute-id", value_name = "CONTROL_PLANE_API_BASE_URL")]
+    pub control_plane_uri: Option<String>,
+}
+
+fn main() -> Result<()> {
+    let cli = Cli::parse();
+
+    let build_tag = init()?;
+
+    let scenario = failpoint_support::init();
 
     // enable core dumping for all child processes
     setrlimit(Resource::CORE, rlimit::INFINITY, rlimit::INFINITY)?;
@@ -85,13 +151,11 @@ fn main() -> Result<()> {
         // Enter startup tracing context
         let _startup_context_guard = startup_context_from_env();
 
-        let cli_args = process_cli(&clap_args)?;
+        let cli_spec = try_spec_from_cli(&cli)?;
 
-        let cli_spec = try_spec_from_cli(&clap_args, &cli_args)?;
+        let compute = wait_spec(build_tag, &cli, cli_spec)?;
 
-        let wait_spec_result = wait_spec(build_tag, cli_args, cli_spec)?;
-
-        start_postgres(&clap_args, wait_spec_result)?
+        start_postgres(&cli, compute)?
 
         // Startup is finished, exit the startup tracing span
     };
@@ -108,7 +172,7 @@ fn main() -> Result<()> {
     deinit_and_exit(wait_pg_result);
 }
 
-fn init() -> Result<(String, clap::ArgMatches)> {
+fn init() -> Result<String> {
     init_tracing_and_logging(DEFAULT_LOG_LEVEL)?;
 
     let mut signals = Signals::new([SIGINT, SIGTERM, SIGQUIT])?;
@@ -123,66 +187,7 @@ fn init() -> Result<(String, clap::ArgMatches)> {
         .to_string();
     info!("build_tag: {build_tag}");
 
-    Ok((build_tag, cli().get_matches()))
-}
-
-fn process_cli(matches: &clap::ArgMatches) -> Result<ProcessCliResult> {
-    let pgbin_default = "postgres";
-    let pgbin = matches
-        .get_one::<String>("pgbin")
-        .map(|s| s.as_str())
-        .unwrap_or(pgbin_default);
-
-    let ext_remote_storage = matches
-        .get_one::<String>("remote-ext-config")
-        // Compatibility hack: if the control plane specified any remote-ext-config
-        // use the default value for extension storage proxy gateway.
-        // Remove this once the control plane is updated to pass the gateway URL
-        .map(|conf| {
-            if conf.starts_with("http") {
-                conf.trim_end_matches('/')
-            } else {
-                "http://pg-ext-s3-gateway"
-            }
-        });
-
-    let http_port = *matches
-        .get_one::<u16>("http-port")
-        .expect("http-port is required");
-    let pgdata = matches
-        .get_one::<String>("pgdata")
-        .expect("PGDATA path is required");
-    let connstr = matches
-        .get_one::<String>("connstr")
-        .expect("Postgres connection string is required");
-    let spec_json = matches.get_one::<String>("spec");
-    let spec_path = matches.get_one::<String>("spec-path");
-    let resize_swap_on_bind = matches.get_flag("resize-swap-on-bind");
-    let set_disk_quota_for_fs = matches.get_one::<String>("set-disk-quota-for-fs");
-
-    Ok(ProcessCliResult {
-        connstr,
-        pgdata,
-        pgbin,
-        ext_remote_storage,
-        http_port,
-        spec_json,
-        spec_path,
-        resize_swap_on_bind,
-        set_disk_quota_for_fs,
-    })
-}
-
-struct ProcessCliResult<'clap> {
-    connstr: &'clap str,
-    pgdata: &'clap str,
-    pgbin: &'clap str,
-    ext_remote_storage: Option<&'clap str>,
-    http_port: u16,
-    spec_json: Option<&'clap String>,
-    spec_path: Option<&'clap String>,
-    resize_swap_on_bind: bool,
-    set_disk_quota_for_fs: Option<&'clap String>,
+    Ok(build_tag)
 }
 
 fn startup_context_from_env() -> Option<opentelemetry::ContextGuard> {
@@ -235,19 +240,9 @@ fn startup_context_from_env() -> Option<opentelemetry::ContextGuard> {
     }
 }
 
-fn try_spec_from_cli(
-    matches: &clap::ArgMatches,
-    ProcessCliResult {
-        spec_json,
-        spec_path,
-        ..
-    }: &ProcessCliResult,
-) -> Result<CliSpecParams> {
-    let compute_id = matches.get_one::<String>("compute-id");
-    let control_plane_uri = matches.get_one::<String>("control-plane-uri");
-
+fn try_spec_from_cli(cli: &Cli) -> Result<CliSpecParams> {
     // First, try to get cluster spec from the cli argument
-    if let Some(spec_json) = spec_json {
+    if let Some(ref spec_json) = cli.spec_json {
         info!("got spec from cli argument {}", spec_json);
         return Ok(CliSpecParams {
             spec: Some(serde_json::from_str(spec_json)?),
@@ -256,7 +251,7 @@ fn try_spec_from_cli(
     }
 
     // Second, try to read it from the file if path is provided
-    if let Some(spec_path) = spec_path {
+    if let Some(ref spec_path) = cli.spec_path {
         let file = File::open(Path::new(spec_path))?;
         return Ok(CliSpecParams {
             spec: Some(serde_json::from_reader(file)?),
@@ -264,17 +259,20 @@ fn try_spec_from_cli(
         });
     }
 
-    let Some(compute_id) = compute_id else {
+    if cli.compute_id.is_none() {
         panic!(
             "compute spec should be provided by one of the following ways: \
                 --spec OR --spec-path OR --control-plane-uri and --compute-id"
         );
     };
-    let Some(control_plane_uri) = control_plane_uri else {
+    if cli.control_plane_uri.is_none() {
         panic!("must specify both --control-plane-uri and --compute-id or none");
     };
 
-    match get_spec_from_control_plane(control_plane_uri, compute_id) {
+    match get_spec_from_control_plane(
+        cli.control_plane_uri.as_ref().unwrap(),
+        cli.compute_id.as_ref().unwrap(),
+    ) {
         Ok(spec) => Ok(CliSpecParams {
             spec,
             live_config_allowed: true,
@@ -298,21 +296,12 @@ struct CliSpecParams {
 
 fn wait_spec(
     build_tag: String,
-    ProcessCliResult {
-        connstr,
-        pgdata,
-        pgbin,
-        ext_remote_storage,
-        resize_swap_on_bind,
-        set_disk_quota_for_fs,
-        http_port,
-        ..
-    }: ProcessCliResult,
+    cli: &Cli,
     CliSpecParams {
         spec,
         live_config_allowed,
     }: CliSpecParams,
-) -> Result<WaitSpecResult> {
+) -> Result<Arc<ComputeNode>> {
     let mut new_state = ComputeState::new();
     let spec_set;
 
@@ -324,7 +313,7 @@ fn wait_spec(
     } else {
         spec_set = false;
     }
-    let connstr = Url::parse(connstr).context("cannot parse connstr as a URL")?;
+    let connstr = Url::parse(&cli.connstr).context("cannot parse connstr as a URL")?;
     let conn_conf = postgres::config::Config::from_str(connstr.as_str())
         .context("cannot build postgres config from connstr")?;
     let tokio_conn_conf = tokio_postgres::config::Config::from_str(connstr.as_str())
@@ -333,14 +322,14 @@ fn wait_spec(
         connstr,
         conn_conf,
         tokio_conn_conf,
-        pgdata: pgdata.to_string(),
-        pgbin: pgbin.to_string(),
-        pgversion: get_pg_version_string(pgbin),
-        http_port,
+        pgdata: cli.pgdata.clone(),
+        pgbin: cli.pgbin.clone(),
+        pgversion: get_pg_version_string(&cli.pgbin),
+        http_port: cli.http_port,
         live_config_allowed,
         state: Mutex::new(new_state),
         state_changed: Condvar::new(),
-        ext_remote_storage: ext_remote_storage.map(|s| s.to_string()),
+        ext_remote_storage: cli.remote_ext_config.clone(),
         ext_download_progress: RwLock::new(HashMap::new()),
         build_tag,
     };
@@ -357,7 +346,7 @@ fn wait_spec(
     // Launch http service first, so that we can serve control-plane requests
     // while configuration is still in progress.
     let _http_handle =
-        launch_http_server(http_port, &compute).expect("cannot launch http endpoint thread");
+        launch_http_server(cli.http_port, &compute).expect("cannot launch http endpoint thread");
 
     if !spec_set {
         // No spec provided, hang waiting for it.
@@ -389,27 +378,12 @@ fn wait_spec(
 
     launch_lsn_lease_bg_task_for_static(&compute);
 
-    Ok(WaitSpecResult {
-        compute,
-        resize_swap_on_bind,
-        set_disk_quota_for_fs: set_disk_quota_for_fs.cloned(),
-    })
-}
-
-struct WaitSpecResult {
-    compute: Arc<ComputeNode>,
-    resize_swap_on_bind: bool,
-    set_disk_quota_for_fs: Option<String>,
+    Ok(compute)
 }
 
 fn start_postgres(
-    // need to allow unused because `matches` is only used if target_os = "linux"
-    #[allow(unused_variables)] matches: &clap::ArgMatches,
-    WaitSpecResult {
-        compute,
-        resize_swap_on_bind,
-        set_disk_quota_for_fs,
-    }: WaitSpecResult,
+    cli: &Cli,
+    compute: Arc<ComputeNode>,
 ) -> Result<(Option<PostgresHandle>, StartPostgresResult)> {
     // We got all we need, update the state.
     let mut state = compute.state.lock().unwrap();
@@ -437,7 +411,7 @@ fn start_postgres(
     let mut delay_exit = false;
 
     // Resize swap to the desired size if the compute spec says so
-    if let (Some(size_bytes), true) = (swap_size_bytes, resize_swap_on_bind) {
+    if let (Some(size_bytes), true) = (swap_size_bytes, cli.resize_swap_on_bind) {
         // To avoid 'swapoff' hitting postgres startup, we need to run resize-swap to completion
         // *before* starting postgres.
         //
@@ -464,9 +438,9 @@ fn start_postgres(
 
     // Set disk quota if the compute spec says so
     if let (Some(disk_quota_bytes), Some(disk_quota_fs_mountpoint)) =
-        (disk_quota_bytes, set_disk_quota_for_fs)
+        (disk_quota_bytes, cli.set_disk_quota_for_fs.as_ref())
     {
-        match set_disk_quota(disk_quota_bytes, &disk_quota_fs_mountpoint) {
+        match set_disk_quota(disk_quota_bytes, disk_quota_fs_mountpoint) {
             Ok(()) => {
                 let size_mib = disk_quota_bytes as f32 / (1 << 20) as f32; // just for more coherent display.
                 info!(%disk_quota_bytes, %size_mib, "set disk quota");
@@ -509,13 +483,7 @@ fn start_postgres(
         if #[cfg(target_os = "linux")] {
             use std::env;
             use tokio_util::sync::CancellationToken;
-            let vm_monitor_addr = matches
-                .get_one::<String>("vm-monitor-addr")
-                .expect("--vm-monitor-addr should always be set because it has a default arg");
-            let file_cache_connstr = matches.get_one::<String>("filecache-connstr");
-            let cgroup = matches.get_one::<String>("cgroup");
 
-            // Only make a runtime if we need to.
             // Note: it seems like you can make a runtime in an inner scope and
             // if you start a task in it it won't be dropped. However, make it
             // in the outermost scope just to be safe.
@@ -538,15 +506,15 @@ fn start_postgres(
             let pgconnstr = if disable_lfc_resizing.unwrap_or(false) {
                 None
             } else {
-                file_cache_connstr.cloned()
+                Some(cli.filecache_connstr.clone())
             };
 
             let vm_monitor = rt.as_ref().map(|rt| {
                 rt.spawn(vm_monitor::start(
                     Box::leak(Box::new(vm_monitor::Args {
-                        cgroup: cgroup.cloned(),
+                        cgroup: Some(cli.cgroup.clone()),
                         pgconnstr,
-                        addr: vm_monitor_addr.clone(),
+                        addr: cli.vm_monitor_addr.clone(),
                     })),
                     token.clone(),
                 ))
@@ -702,105 +670,6 @@ fn deinit_and_exit(WaitPostgresResult { exit_code }: WaitPostgresResult) -> ! {
     exit(exit_code.unwrap_or(1))
 }
 
-fn cli() -> clap::Command {
-    // Env variable is set by `cargo`
-    let version = option_env!("CARGO_PKG_VERSION").unwrap_or("unknown");
-    clap::Command::new("compute_ctl")
-        .version(version)
-        .arg(
-            Arg::new("http-port")
-                .long("http-port")
-                .value_name("HTTP_PORT")
-                .default_value("3080")
-                .value_parser(clap::value_parser!(u16))
-                .required(false),
-        )
-        .arg(
-            Arg::new("connstr")
-                .short('C')
-                .long("connstr")
-                .value_name("DATABASE_URL")
-                .required(true),
-        )
-        .arg(
-            Arg::new("pgdata")
-                .short('D')
-                .long("pgdata")
-                .value_name("DATADIR")
-                .required(true),
-        )
-        .arg(
-            Arg::new("pgbin")
-                .short('b')
-                .long("pgbin")
-                .default_value("postgres")
-                .value_name("POSTGRES_PATH"),
-        )
-        .arg(
-            Arg::new("spec")
-                .short('s')
-                .long("spec")
-                .value_name("SPEC_JSON"),
-        )
-        .arg(
-            Arg::new("spec-path")
-                .short('S')
-                .long("spec-path")
-                .value_name("SPEC_PATH"),
-        )
-        .arg(
-            Arg::new("compute-id")
-                .short('i')
-                .long("compute-id")
-                .value_name("COMPUTE_ID"),
-        )
-        .arg(
-            Arg::new("control-plane-uri")
-                .short('p')
-                .long("control-plane-uri")
-                .value_name("CONTROL_PLANE_API_BASE_URI"),
-        )
-        .arg(
-            Arg::new("remote-ext-config")
-                .short('r')
-                .long("remote-ext-config")
-                .value_name("REMOTE_EXT_CONFIG"),
-        )
-        // TODO(fprasx): we currently have default arguments because the cloud PR
-        // to pass them in hasn't been merged yet. We should get rid of them once
-        // the PR is merged.
-        .arg(
-            Arg::new("vm-monitor-addr")
-                .long("vm-monitor-addr")
-                .default_value("0.0.0.0:10301")
-                .value_name("VM_MONITOR_ADDR"),
-        )
-        .arg(
-            Arg::new("cgroup")
-                .long("cgroup")
-                .default_value("neon-postgres")
-                .value_name("CGROUP"),
-        )
-        .arg(
-            Arg::new("filecache-connstr")
-                .long("filecache-connstr")
-                .default_value(
-                    "host=localhost port=5432 dbname=postgres user=cloud_admin sslmode=disable application_name=vm-monitor",
-                )
-                .value_name("FILECACHE_CONNSTR"),
-        )
-        .arg(
-            Arg::new("resize-swap-on-bind")
-                .long("resize-swap-on-bind")
-                .action(clap::ArgAction::SetTrue),
-        )
-        .arg(
-            Arg::new("set-disk-quota-for-fs")
-                .long("set-disk-quota-for-fs")
-                .value_name("SET_DISK_QUOTA_FOR_FS")
-        )
-}
-
 /// When compute_ctl is killed, send also termination signal to sync-safekeepers
 /// to prevent leakage. TODO: it is better to convert compute_ctl to async and
 /// wait for termination which would be easy then.
@@ -810,7 +679,14 @@ fn handle_exit_signal(sig: i32) {
     exit(1);
 }
 
-#[test]
-fn verify_cli() {
-    cli().debug_assert()
+#[cfg(test)]
+mod test {
+    use clap::CommandFactory;
+
+    use super::Cli;
+
+    #[test]
+    fn verify_cli() {
+        Cli::command().debug_assert()
+    }
 }

compute_tools/src/bin/fast_import.rs

@@ -58,6 +58,8 @@ struct Args {
     pg_bin_dir: Utf8PathBuf,
     #[clap(long)]
     pg_lib_dir: Utf8PathBuf,
+    #[clap(long)]
+    pg_port: Option<u16>, // port to run postgres on, 5432 is default
 }
 
 #[serde_with::serde_as]
@@ -74,6 +76,13 @@ enum EncryptionSecret {
     KMS { key_id: String },
 }
 
+// copied from pageserver_api::config::defaults::DEFAULT_LOCALE to avoid dependency just for a constant
+const DEFAULT_LOCALE: &str = if cfg!(target_os = "macos") {
+    "C"
+} else {
+    "C.UTF-8"
+};
+
 #[tokio::main]
 pub(crate) async fn main() -> anyhow::Result<()> {
     utils::logging::init(
@@ -97,6 +106,10 @@ pub(crate) async fn main() -> anyhow::Result<()> {
     let working_directory = args.working_directory;
     let pg_bin_dir = args.pg_bin_dir;
     let pg_lib_dir = args.pg_lib_dir;
+    let pg_port = args.pg_port.unwrap_or_else(|| {
+        info!("pg_port not specified, using default 5432");
+        5432
+    });
 
     // Initialize AWS clients only if s3_prefix is specified
     let (aws_config, kms_client) = if args.s3_prefix.is_some() {
@@ -180,7 +193,7 @@ pub(crate) async fn main() -> anyhow::Result<()> {
     let superuser = "cloud_admin"; // XXX: this shouldn't be hard-coded
     postgres_initdb::do_run_initdb(postgres_initdb::RunInitdbArgs {
         superuser,
-        locale: "en_US.UTF-8", // XXX: this shouldn't be hard-coded,
+        locale: DEFAULT_LOCALE, // XXX: this shouldn't be hard-coded,
         pg_version,
         initdb_bin: pg_bin_dir.join("initdb").as_ref(),
         library_search_path: &pg_lib_dir, // TODO: is this right? Prob works in compute image, not sure about neon_local.
@@ -197,6 +210,7 @@ pub(crate) async fn main() -> anyhow::Result<()> {
     let mut postgres_proc = tokio::process::Command::new(pgbin)
         .arg("-D")
         .arg(&pgdata_dir)
+        .args(["-p", &format!("{pg_port}")])
         .args(["-c", "wal_level=minimal"])
         .args(["-c", "shared_buffers=10GB"])
         .args(["-c", "max_wal_senders=0"])
@@ -216,6 +230,7 @@ pub(crate) async fn main() -> anyhow::Result<()> {
             ),
         ])
         .env_clear()
+        .env("LD_LIBRARY_PATH", &pg_lib_dir)
         .stdout(std::process::Stdio::piped())
         .stderr(std::process::Stdio::piped())
         .spawn()
@@ -232,7 +247,7 @@ pub(crate) async fn main() -> anyhow::Result<()> {
 
     // Create neondb database in the running postgres
     let restore_pg_connstring =
-        format!("host=localhost port=5432 user={superuser} dbname=postgres");
+        format!("host=localhost port={pg_port} user={superuser} dbname=postgres");
 
     let start_time = std::time::Instant::now();
 
@@ -314,6 +329,7 @@ pub(crate) async fn main() -> anyhow::Result<()> {
             .arg(&source_connection_string)
             // how we run it
             .env_clear()
+            .env("LD_LIBRARY_PATH", &pg_lib_dir)
             .kill_on_drop(true)
             .stdout(std::process::Stdio::piped())
             .stderr(std::process::Stdio::piped())
@@ -347,6 +363,7 @@ pub(crate) async fn main() -> anyhow::Result<()> {
             .arg(&dumpdir)
             // how we run it
             .env_clear()
+            .env("LD_LIBRARY_PATH", &pg_lib_dir)
             .kill_on_drop(true)
             .stdout(std::process::Stdio::piped())
             .stderr(std::process::Stdio::piped())

compute_tools/src/compute.rs

@@ -41,14 +41,14 @@ use crate::local_proxy;
 use crate::pg_helpers::*;
 use crate::spec::*;
 use crate::spec_apply::ApplySpecPhase::{
-    CreateAndAlterDatabases, CreateAndAlterRoles, CreateAvailabilityCheck, CreateSuperUser,
-    DropInvalidDatabases, DropRoles, HandleNeonExtension, HandleOtherExtensions,
-    RenameAndDeleteDatabases, RenameRoles, RunInEachDatabase,
+    CreateAndAlterDatabases, CreateAndAlterRoles, CreateAvailabilityCheck, CreateSchemaNeon,
+    CreateSuperUser, DropInvalidDatabases, DropRoles, FinalizeDropLogicalSubscriptions,
+    HandleNeonExtension, HandleOtherExtensions, RenameAndDeleteDatabases, RenameRoles,
+    RunInEachDatabase,
 };
 use crate::spec_apply::PerDatabasePhase;
 use crate::spec_apply::PerDatabasePhase::{
-    ChangeSchemaPerms, DeleteDBRoleReferences, DropSubscriptionsForDeletedDatabases,
-    HandleAnonExtension,
+    ChangeSchemaPerms, DeleteDBRoleReferences, DropLogicalSubscriptions, HandleAnonExtension,
 };
 use crate::spec_apply::{apply_operations, MutableApplyContext, DB};
 use crate::sync_sk::{check_if_synced, ping_safekeeper};
@@ -340,6 +340,15 @@ impl ComputeNode {
         self.state.lock().unwrap().status
     }
 
+    pub fn get_timeline_id(&self) -> Option<TimelineId> {
+        self.state
+            .lock()
+            .unwrap()
+            .pspec
+            .as_ref()
+            .map(|s| s.timeline_id)
+    }
+
     // Remove `pgdata` directory and create it again with right permissions.
     fn create_pgdata(&self) -> Result<()> {
         // Ignore removal error, likely it is a 'No such file or directory (os error 2)'.
@@ -929,6 +938,48 @@ impl ComputeNode {
                 .map(|role| (role.name.clone(), role))
                 .collect::<HashMap<String, Role>>();
 
+            // Check if we need to drop subscriptions before starting the endpoint.
+            //
+            // It is important to do this operation exactly once when endpoint starts on a new branch.
+            // Otherwise, we may drop not inherited, but newly created subscriptions.
+            //
+            // We cannot rely only on spec.drop_subscriptions_before_start flag,
+            // because if for some reason compute restarts inside VM,
+            // it will start again with the same spec and flag value.
+            //
+            // To handle this, we save the fact of the operation in the database
+            // in the neon.drop_subscriptions_done table.
+            // If the table does not exist, we assume that the operation was never performed, so we must do it.
+            // If table exists, we check if the operation was performed on the current timelilne.
+            //
+            let mut drop_subscriptions_done = false;
+
+            if spec.drop_subscriptions_before_start {
+                let timeline_id = self.get_timeline_id().context("timeline_id must be set")?;
+                let query = format!("select 1 from neon.drop_subscriptions_done where timeline_id = '{}'", timeline_id);
+
+                info!("Checking if drop subscription operation was already performed for timeline_id: {}", timeline_id);
+
+                drop_subscriptions_done =  match
+                    client.simple_query(&query).await {
+                    Ok(result) => {
+                        matches!(&result[0], postgres::SimpleQueryMessage::Row(_))
+                    },
+                    Err(e) =>
+                    {
+                        match e.code() {
+                            Some(&SqlState::UNDEFINED_TABLE) => false,
+                            _ => {
+                                // We don't expect any other error here, except for the schema/table not existing
+                                error!("Error checking if drop subscription operation was already performed: {}", e);
+                                return Err(e.into());
+                            }
+                        }
+                    }
+                }
+            };
+
+
             let jwks_roles = Arc::new(
                 spec.as_ref()
                     .local_proxy_config
@@ -996,7 +1047,7 @@ impl ComputeNode {
                         jwks_roles.clone(),
                         concurrency_token.clone(),
                         db,
-                        [DropSubscriptionsForDeletedDatabases].to_vec(),
+                        [DropLogicalSubscriptions].to_vec(),
                     );
 
                     Ok(spawn(fut))
@@ -1024,6 +1075,7 @@ impl ComputeNode {
                 CreateAndAlterRoles,
                 RenameAndDeleteDatabases,
                 CreateAndAlterDatabases,
+                CreateSchemaNeon,
             ] {
                 info!("Applying phase {:?}", &phase);
                 apply_operations(
@@ -1064,6 +1116,17 @@ impl ComputeNode {
                     }
 
                     let conf = Arc::new(conf);
+                    let mut phases = vec![
+                        DeleteDBRoleReferences,
+                        ChangeSchemaPerms,
+                        HandleAnonExtension,
+                    ];
+
+                    if spec.drop_subscriptions_before_start && !drop_subscriptions_done {
+                        info!("Adding DropLogicalSubscriptions phase because drop_subscriptions_before_start is set");
+                        phases.push(DropLogicalSubscriptions);
+                    }
+
                     let fut = Self::apply_spec_sql_db(
                         spec.clone(),
                         conf,
@@ -1071,12 +1134,7 @@ impl ComputeNode {
                         jwks_roles.clone(),
                         concurrency_token.clone(),
                         db,
-                        [
-                            DeleteDBRoleReferences,
-                            ChangeSchemaPerms,
-                            HandleAnonExtension,
-                        ]
-                        .to_vec(),
+                        phases,
                     );
 
                     Ok(spawn(fut))
@@ -1088,12 +1146,20 @@ impl ComputeNode {
                 handle.await??;
             }
 
-            for phase in vec![
+            let mut phases = vec![
                 HandleOtherExtensions,
-                HandleNeonExtension,
+                HandleNeonExtension, // This step depends on CreateSchemaNeon
                 CreateAvailabilityCheck,
                 DropRoles,
-            ] {
+            ];
+
+            // This step depends on CreateSchemaNeon
+            if spec.drop_subscriptions_before_start && !drop_subscriptions_done {
+                info!("Adding FinalizeDropLogicalSubscriptions phase because drop_subscriptions_before_start is set");
+                phases.push(FinalizeDropLogicalSubscriptions);
+            }
+
+            for phase in phases {
                 debug!("Applying phase {:?}", &phase);
                 apply_operations(
                     spec.clone(),
@@ -1463,6 +1529,14 @@ impl ComputeNode {
                         Ok(())
                     },
                 )?;
+
+                let postgresql_conf_path = pgdata_path.join("postgresql.conf");
+                if config::line_in_file(
+                    &postgresql_conf_path,
+                    "neon.disable_logical_replication_subscribers=false",
+                )? {
+                    info!("updated postgresql.conf to set neon.disable_logical_replication_subscribers=false");
+                }
                 self.pg_reload_conf()?;
             }
             self.post_apply_config()?;

compute_tools/src/config.rs

@@ -129,6 +129,13 @@ pub fn write_postgres_conf(
 
     writeln!(file, "neon.extension_server_port={}", extension_server_port)?;
 
+    if spec.drop_subscriptions_before_start {
+        writeln!(file, "neon.disable_logical_replication_subscribers=true")?;
+    } else {
+        // be explicit about the default value
+        writeln!(file, "neon.disable_logical_replication_subscribers=false")?;
+    }
+
     // This is essential to keep this line at the end of the file,
     // because it is intended to override any settings above.
     writeln!(file, "include_if_exists = 'compute_ctl_temp_override.conf'")?;

compute_tools/src/extension_server.rs

@@ -85,6 +85,8 @@ use tracing::info;
 use tracing::log::warn;
 use zstd::stream::read::Decoder;
 
+use crate::metrics::{REMOTE_EXT_REQUESTS_TOTAL, UNKNOWN_HTTP_STATUS};
+
 fn get_pg_config(argument: &str, pgbin: &str) -> String {
     // gives the result of `pg_config [argument]`
     // where argument is a flag like `--version` or `--sharedir`
@@ -256,23 +258,60 @@ pub fn create_control_files(remote_extensions: &RemoteExtSpec, pgbin: &str) {
 async fn download_extension_tar(ext_remote_storage: &str, ext_path: &str) -> Result<Bytes> {
     let uri = format!("{}/{}", ext_remote_storage, ext_path);
 
-    info!("Download extension {:?} from uri {:?}", ext_path, uri);
+    info!("Download extension {} from uri {}", ext_path, uri);
 
-    let resp = reqwest::get(uri).await?;
+    match do_extension_server_request(&uri).await {
+        Ok(resp) => {
+            info!("Successfully downloaded remote extension data {}", ext_path);
+            REMOTE_EXT_REQUESTS_TOTAL
+                .with_label_values(&[&StatusCode::OK.to_string()])
+                .inc();
+            Ok(resp)
+        }
+        Err((msg, status)) => {
+            REMOTE_EXT_REQUESTS_TOTAL
+                .with_label_values(&[&status])
+                .inc();
+            bail!(msg);
+        }
+    }
+}
 
-    match resp.status() {
+// Do a single remote extensions server request.
+// Return result or (error message + stringified status code) in case of any failures.
+async fn do_extension_server_request(uri: &str) -> Result<Bytes, (String, String)> {
+    let resp = reqwest::get(uri).await.map_err(|e| {
+        (
+            format!(
+                "could not perform remote extensions server request: {:?}",
+                e
+            ),
+            UNKNOWN_HTTP_STATUS.to_string(),
+        )
+    })?;
+    let status = resp.status();
+
+    match status {
         StatusCode::OK => match resp.bytes().await {
-            Ok(resp) => {
-                info!("Download extension {:?} completed successfully", ext_path);
-                Ok(resp)
-            }
-            Err(e) => bail!("could not deserialize remote extension response: {}", e),
+            Ok(resp) => Ok(resp),
+            Err(e) => Err((
+                format!("could not read remote extensions server response: {:?}", e),
+                // It's fine to return and report error with status as 200 OK,
+                // because we still failed to read the response.
+                status.to_string(),
+            )),
         },
-        StatusCode::SERVICE_UNAVAILABLE => bail!("remote extension is temporarily unavailable"),
-        _ => bail!(
-            "unexpected remote extension response status code: {}",
-            resp.status()
-        ),
+        StatusCode::SERVICE_UNAVAILABLE => Err((
+            "remote extensions server is temporarily unavailable".to_string(),
+            status.to_string(),
+        )),
+        _ => Err((
+            format!(
+                "unexpected remote extensions server response status code: {}",
+                status
+            ),
+            status.to_string(),
+        )),
     }
 }
 

compute_tools/src/http/extract/json.rs

@@ -1,9 +1,6 @@
 use std::ops::{Deref, DerefMut};
 
-use axum::{
-    async_trait,
-    extract::{rejection::JsonRejection, FromRequest, Request},
-};
+use axum::extract::{rejection::JsonRejection, FromRequest, Request};
 use compute_api::responses::GenericAPIError;
 use http::StatusCode;
 
@@ -12,7 +9,6 @@ use http::StatusCode;
 #[derive(Debug, Clone, Copy, Default)]
 pub(crate) struct Json<T>(pub T);
 
-#[async_trait]
 impl<S, T> FromRequest<S> for Json<T>
 where
     axum::Json<T>: FromRequest<S, Rejection = JsonRejection>,

compute_tools/src/http/extract/path.rs

@@ -1,9 +1,6 @@
 use std::ops::{Deref, DerefMut};
 
-use axum::{
-    async_trait,
-    extract::{rejection::PathRejection, FromRequestParts},
-};
+use axum::extract::{rejection::PathRejection, FromRequestParts};
 use compute_api::responses::GenericAPIError;
 use http::{request::Parts, StatusCode};
 
@@ -12,7 +9,6 @@ use http::{request::Parts, StatusCode};
 #[derive(Debug, Clone, Copy, Default)]
 pub(crate) struct Path<T>(pub T);
 
-#[async_trait]
 impl<S, T> FromRequestParts<S> for Path<T>
 where
     axum::extract::Path<T>: FromRequestParts<S, Rejection = PathRejection>,

compute_tools/src/http/extract/query.rs

@@ -1,9 +1,6 @@
 use std::ops::{Deref, DerefMut};
 
-use axum::{
-    async_trait,
-    extract::{rejection::QueryRejection, FromRequestParts},
-};
+use axum::extract::{rejection::QueryRejection, FromRequestParts};
 use compute_api::responses::GenericAPIError;
 use http::{request::Parts, StatusCode};
 
@@ -12,7 +9,6 @@ use http::{request::Parts, StatusCode};
 #[derive(Debug, Clone, Copy, Default)]
 pub(crate) struct Query<T>(pub T);
 
-#[async_trait]
 impl<S, T> FromRequestParts<S> for Query<T>
 where
     axum::extract::Query<T>: FromRequestParts<S, Rejection = QueryRejection>,

compute_tools/src/http/openapi_spec.yaml

@@ -68,35 +68,6 @@ paths:
               schema:
                 $ref: "#/components/schemas/ComputeInsights"
 
-  /installed_extensions:
-    get:
-      tags:
-      - Info
-      summary: Get installed extensions.
-      description: ""
-      operationId: getInstalledExtensions
-      responses:
-        200:
-          description: List of installed extensions
-          content:
-            application/json:
-              schema:
-                $ref: "#/components/schemas/InstalledExtensions"
-  /info:
-    get:
-      tags:
-      - Info
-      summary: Get info about the compute pod / VM.
-      description: ""
-      operationId: getInfo
-      responses:
-        200:
-          description: Info
-          content:
-            application/json:
-              schema:
-                $ref: "#/components/schemas/Info"
-
   /dbs_and_roles:
     get:
       tags:

compute_tools/src/http/routes/info.rs

@@ -1,11 +0,0 @@
-use axum::response::Response;
-use compute_api::responses::InfoResponse;
-use http::StatusCode;
-
-use crate::http::JsonResponse;
-
-/// Get information about the physical characteristics about the compute.
-pub(in crate::http) async fn get_info() -> Response {
-    let num_cpus = num_cpus::get_physical();
-    JsonResponse::success(StatusCode::OK, &InfoResponse { num_cpus })
-}

compute_tools/src/http/routes/installed_extensions.rs

@@ -1,33 +0,0 @@
-use std::sync::Arc;
-
-use axum::{extract::State, response::Response};
-use compute_api::responses::ComputeStatus;
-use http::StatusCode;
-use tokio::task;
-
-use crate::{compute::ComputeNode, http::JsonResponse, installed_extensions};
-
-/// Get a list of installed extensions.
-pub(in crate::http) async fn get_installed_extensions(
-    State(compute): State<Arc<ComputeNode>>,
-) -> Response {
-    let status = compute.get_status();
-    if status != ComputeStatus::Running {
-        return JsonResponse::invalid_status(status);
-    }
-
-    let conf = compute.get_conn_conf(None);
-    let res = task::spawn_blocking(move || installed_extensions::get_installed_extensions(conf))
-        .await
-        .unwrap();
-
-    match res {
-        Ok(installed_extensions) => {
-            JsonResponse::success(StatusCode::OK, Some(installed_extensions))
-        }
-        Err(e) => JsonResponse::error(
-            StatusCode::INTERNAL_SERVER_ERROR,
-            format!("failed to get list of installed extensions: {e}"),
-        ),
-    }
-}

compute_tools/src/http/routes/metrics.rs

@@ -2,17 +2,16 @@ use axum::{body::Body, response::Response};
 use http::header::CONTENT_TYPE;
 use http::StatusCode;
 use metrics::proto::MetricFamily;
-use metrics::Encoder;
-use metrics::TextEncoder;
+use metrics::{Encoder, TextEncoder};
 
-use crate::{http::JsonResponse, installed_extensions};
+use crate::{http::JsonResponse, metrics::collect};
 
 /// Expose Prometheus metrics.
 pub(in crate::http) async fn get_metrics() -> Response {
     // When we call TextEncoder::encode() below, it will immediately return an
     // error if a metric family has no metrics, so we need to preemptively
     // filter out metric families with no metrics.
-    let metrics = installed_extensions::collect()
+    let metrics = collect()
         .into_iter()
         .filter(|m| !m.get_metric().is_empty())
         .collect::<Vec<MetricFamily>>();

compute_tools/src/http/routes/mod.rs

@@ -10,9 +10,7 @@ pub(in crate::http) mod extension_server;
 pub(in crate::http) mod extensions;
 pub(in crate::http) mod failpoints;
 pub(in crate::http) mod grants;
-pub(in crate::http) mod info;
 pub(in crate::http) mod insights;
-pub(in crate::http) mod installed_extensions;
 pub(in crate::http) mod metrics;
 pub(in crate::http) mod metrics_json;
 pub(in crate::http) mod status;

compute_tools/src/http/server.rs

@@ -22,8 +22,7 @@ use uuid::Uuid;
 
 use super::routes::{
     check_writability, configure, database_schema, dbs_and_roles, extension_server, extensions,
-    grants, info as info_route, insights, installed_extensions, metrics, metrics_json, status,
-    terminate,
+    grants, insights, metrics, metrics_json, status, terminate,
 };
 use crate::compute::ComputeNode;
 
@@ -55,17 +54,12 @@ async fn serve(port: u16, compute: Arc<ComputeNode>) {
         .route("/database_schema", get(database_schema::get_schema_dump))
         .route("/dbs_and_roles", get(dbs_and_roles::get_catalog_objects))
         .route(
-            "/extension_server/*filename",
+            "/extension_server/{*filename}",
             post(extension_server::download_extension),
         )
         .route("/extensions", post(extensions::install_extension))
         .route("/grants", post(grants::add_grant))
-        .route("/info", get(info_route::get_info))
         .route("/insights", get(insights::get_insights))
-        .route(
-            "/installed_extensions",
-            get(installed_extensions::get_installed_extensions),
-        )
         .route("/metrics", get(metrics::get_metrics))
         .route("/metrics.json", get(metrics_json::get_metrics))
         .route("/status", get(status::get_status))

compute_tools/src/installed_extensions.rs

@@ -1,13 +1,10 @@
 use compute_api::responses::{InstalledExtension, InstalledExtensions};
-use metrics::proto::MetricFamily;
 use std::collections::HashMap;
 
 use anyhow::Result;
 use postgres::{Client, NoTls};
 
-use metrics::core::Collector;
-use metrics::{register_uint_gauge_vec, UIntGaugeVec};
-use once_cell::sync::Lazy;
+use crate::metrics::INSTALLED_EXTENSIONS;
 
 /// We don't reuse get_existing_dbs() just for code clarity
 /// and to make database listing query here more explicit.
@@ -102,16 +99,3 @@ pub fn get_installed_extensions(mut conf: postgres::config::Config) -> Result<In
         extensions: extensions_map.into_values().collect(),
     })
 }
-
-static INSTALLED_EXTENSIONS: Lazy<UIntGaugeVec> = Lazy::new(|| {
-    register_uint_gauge_vec!(
-        "compute_installed_extensions",
-        "Number of databases where the version of extension is installed",
-        &["extension_name", "version", "owned_by_superuser"]
-    )
-    .expect("failed to define a metric")
-});
-
-pub fn collect() -> Vec<MetricFamily> {
-    INSTALLED_EXTENSIONS.collect()
-}

compute_tools/src/lib.rs

@@ -16,6 +16,7 @@ pub mod extension_server;
 pub mod installed_extensions;
 pub mod local_proxy;
 pub mod lsn_lease;
+pub mod metrics;
 mod migration;
 pub mod monitor;
 pub mod params;

compute_tools/src/metrics.rs

@@ -0,0 +1,70 @@
+use metrics::core::Collector;
+use metrics::proto::MetricFamily;
+use metrics::{register_int_counter_vec, register_uint_gauge_vec, IntCounterVec, UIntGaugeVec};
+use once_cell::sync::Lazy;
+
+pub(crate) static INSTALLED_EXTENSIONS: Lazy<UIntGaugeVec> = Lazy::new(|| {
+    register_uint_gauge_vec!(
+        "compute_installed_extensions",
+        "Number of databases where the version of extension is installed",
+        &["extension_name", "version", "owned_by_superuser"]
+    )
+    .expect("failed to define a metric")
+});
+
+// Normally, any HTTP API request is described by METHOD (e.g. GET, POST, etc.) + PATH,
+// but for all our APIs we defined a 'slug'/method/operationId in the OpenAPI spec.
+// And it's fair to call it a 'RPC' (Remote Procedure Call).
+pub enum CPlaneRequestRPC {
+    GetSpec,
+}
+
+impl CPlaneRequestRPC {
+    pub fn as_str(&self) -> &str {
+        match self {
+            CPlaneRequestRPC::GetSpec => "GetSpec",
+        }
+    }
+}
+
+pub const UNKNOWN_HTTP_STATUS: &str = "unknown";
+
+pub(crate) static CPLANE_REQUESTS_TOTAL: Lazy<IntCounterVec> = Lazy::new(|| {
+    register_int_counter_vec!(
+        "compute_ctl_cplane_requests_total",
+        "Total number of control plane requests made by compute_ctl by status",
+        &["rpc", "http_status"]
+    )
+    .expect("failed to define a metric")
+});
+
+/// Total number of failed database migrations. Per-compute, this is actually a boolean metric,
+/// either empty or with a single value (1, migration_id) because we stop at the first failure.
+/// Yet, the sum over the fleet will provide the total number of failures.
+pub(crate) static DB_MIGRATION_FAILED: Lazy<IntCounterVec> = Lazy::new(|| {
+    register_int_counter_vec!(
+        "compute_ctl_db_migration_failed_total",
+        "Total number of failed database migrations",
+        &["migration_id"]
+    )
+    .expect("failed to define a metric")
+});
+
+pub(crate) static REMOTE_EXT_REQUESTS_TOTAL: Lazy<IntCounterVec> = Lazy::new(|| {
+    register_int_counter_vec!(
+        "compute_ctl_remote_ext_requests_total",
+        "Total number of requests made by compute_ctl to download extensions from S3 proxy by status",
+        // Do not use any labels like extension name yet.
+        // We can add them later if needed.
+        &["http_status"]
+    )
+    .expect("failed to define a metric")
+});
+
+pub fn collect() -> Vec<MetricFamily> {
+    let mut metrics = INSTALLED_EXTENSIONS.collect();
+    metrics.extend(CPLANE_REQUESTS_TOTAL.collect());
+    metrics.extend(REMOTE_EXT_REQUESTS_TOTAL.collect());
+    metrics.extend(DB_MIGRATION_FAILED.collect());
+    metrics
+}

compute_tools/src/migration.rs

@@ -1,7 +1,9 @@
 use anyhow::{Context, Result};
 use fail::fail_point;
 use postgres::{Client, Transaction};
-use tracing::info;
+use tracing::{error, info};
+
+use crate::metrics::DB_MIGRATION_FAILED;
 
 /// Runs a series of migrations on a target database
 pub(crate) struct MigrationRunner<'m> {
@@ -78,24 +80,31 @@ impl<'m> MigrationRunner<'m> {
         Ok(())
     }
 
-    /// Run an individual migration
-    fn run_migration(txn: &mut Transaction, migration_id: i64, migration: &str) -> Result<()> {
+    /// Run an individual migration in a separate transaction block.
+    fn run_migration(client: &mut Client, migration_id: i64, migration: &str) -> Result<()> {
+        let mut txn = client
+            .transaction()
+            .with_context(|| format!("begin transaction for migration {migration_id}"))?;
+
         if migration.starts_with("-- SKIP") {
             info!("Skipping migration id={}", migration_id);
 
             // Even though we are skipping the migration, updating the
             // migration ID should help keep logic easy to understand when
             // trying to understand the state of a cluster.
-            Self::update_migration_id(txn, migration_id)?;
+            Self::update_migration_id(&mut txn, migration_id)?;
         } else {
             info!("Running migration id={}:\n{}\n", migration_id, migration);
 
             txn.simple_query(migration)
                 .with_context(|| format!("apply migration {migration_id}"))?;
 
-            Self::update_migration_id(txn, migration_id)?;
+            Self::update_migration_id(&mut txn, migration_id)?;
         }
 
+        txn.commit()
+            .with_context(|| format!("commit transaction for migration {migration_id}"))?;
+
         Ok(())
     }
 
@@ -109,19 +118,20 @@ impl<'m> MigrationRunner<'m> {
             // The index lags the migration ID by 1, so the current migration
             // ID is also the next index
             let migration_id = (current_migration + 1) as i64;
+            let migration = self.migrations[current_migration];
 
-            let mut txn = self
-                .client
-                .transaction()
-                .with_context(|| format!("begin transaction for migration {migration_id}"))?;
-
-            Self::run_migration(&mut txn, migration_id, self.migrations[current_migration])
-                .with_context(|| format!("running migration {migration_id}"))?;
-
-            txn.commit()
-                .with_context(|| format!("commit transaction for migration {migration_id}"))?;
-
-            info!("Finished migration id={}", migration_id);
+            match Self::run_migration(self.client, migration_id, migration) {
+                Ok(_) => {
+                    info!("Finished migration id={}", migration_id);
+                }
+                Err(e) => {
+                    error!("Failed to run migration id={}: {:?}", migration_id, e);
+                    DB_MIGRATION_FAILED
+                        .with_label_values(&[migration_id.to_string().as_str()])
+                        .inc();
+                    return Err(e);
+                }
+            }
 
             current_migration += 1;
         }

compute_tools/src/spec.rs

@@ -6,6 +6,7 @@ use std::path::Path;
 use tracing::{error, info, instrument, warn};
 
 use crate::config;
+use crate::metrics::{CPlaneRequestRPC, CPLANE_REQUESTS_TOTAL, UNKNOWN_HTTP_STATUS};
 use crate::migration::MigrationRunner;
 use crate::params::PG_HBA_ALL_MD5;
 use crate::pg_helpers::*;
@@ -19,7 +20,7 @@ use compute_api::spec::ComputeSpec;
 fn do_control_plane_request(
     uri: &str,
     jwt: &str,
-) -> Result<ControlPlaneSpecResponse, (bool, String)> {
+) -> Result<ControlPlaneSpecResponse, (bool, String, String)> {
     let resp = reqwest::blocking::Client::new()
         .get(uri)
         .header("Authorization", format!("Bearer {}", jwt))
@@ -27,35 +28,42 @@ fn do_control_plane_request(
         .map_err(|e| {
             (
                 true,
-                format!("could not perform spec request to control plane: {}", e),
+                format!("could not perform spec request to control plane: {:?}", e),
+                UNKNOWN_HTTP_STATUS.to_string(),
             )
         })?;
 
-    match resp.status() {
+    let status = resp.status();
+    match status {
         StatusCode::OK => match resp.json::<ControlPlaneSpecResponse>() {
             Ok(spec_resp) => Ok(spec_resp),
             Err(e) => Err((
                 true,
-                format!("could not deserialize control plane response: {}", e),
+                format!("could not deserialize control plane response: {:?}", e),
+                status.to_string(),
             )),
         },
-        StatusCode::SERVICE_UNAVAILABLE => {
-            Err((true, "control plane is temporarily unavailable".to_string()))
-        }
+        StatusCode::SERVICE_UNAVAILABLE => Err((
+            true,
+            "control plane is temporarily unavailable".to_string(),
+            status.to_string(),
+        )),
         StatusCode::BAD_GATEWAY => {
             // We have a problem with intermittent 502 errors now
             // https://github.com/neondatabase/cloud/issues/2353
             // It's fine to retry GET request in this case.
-            Err((true, "control plane request failed with 502".to_string()))
+            Err((
+                true,
+                "control plane request failed with 502".to_string(),
+                status.to_string(),
+            ))
         }
         // Another code, likely 500 or 404, means that compute is unknown to the control plane
         // or some internal failure happened. Doesn't make much sense to retry in this case.
         _ => Err((
             false,
-            format!(
-                "unexpected control plane response status code: {}",
-                resp.status()
-            ),
+            format!("unexpected control plane response status code: {}", status),
+            status.to_string(),
         )),
     }
 }
@@ -83,17 +91,28 @@ pub fn get_spec_from_control_plane(
     // - got spec -> return Ok(Some(spec))
     while attempt < 4 {
         spec = match do_control_plane_request(&cp_uri, &jwt) {
-            Ok(spec_resp) => match spec_resp.status {
-                ControlPlaneComputeStatus::Empty => Ok(None),
-                ControlPlaneComputeStatus::Attached => {
-                    if let Some(spec) = spec_resp.spec {
-                        Ok(Some(spec))
-                    } else {
-                        bail!("compute is attached, but spec is empty")
+            Ok(spec_resp) => {
+                CPLANE_REQUESTS_TOTAL
+                    .with_label_values(&[
+                        CPlaneRequestRPC::GetSpec.as_str(),
+                        &StatusCode::OK.to_string(),
+                    ])
+                    .inc();
+                match spec_resp.status {
+                    ControlPlaneComputeStatus::Empty => Ok(None),
+                    ControlPlaneComputeStatus::Attached => {
+                        if let Some(spec) = spec_resp.spec {
+                            Ok(Some(spec))
+                        } else {
+                            bail!("compute is attached, but spec is empty")
+                        }
                     }
                 }
-            },
-            Err((retry, msg)) => {
+            }
+            Err((retry, msg, status)) => {
+                CPLANE_REQUESTS_TOTAL
+                    .with_label_values(&[CPlaneRequestRPC::GetSpec.as_str(), &status])
+                    .inc();
                 if retry {
                     Err(anyhow!(msg))
                 } else {

compute_tools/src/spec_apply.rs

@@ -47,7 +47,7 @@ pub enum PerDatabasePhase {
     DeleteDBRoleReferences,
     ChangeSchemaPerms,
     HandleAnonExtension,
-    DropSubscriptionsForDeletedDatabases,
+    DropLogicalSubscriptions,
 }
 
 #[derive(Clone, Debug)]
@@ -58,11 +58,13 @@ pub enum ApplySpecPhase {
     CreateAndAlterRoles,
     RenameAndDeleteDatabases,
     CreateAndAlterDatabases,
+    CreateSchemaNeon,
     RunInEachDatabase { db: DB, subphase: PerDatabasePhase },
     HandleOtherExtensions,
     HandleNeonExtension,
     CreateAvailabilityCheck,
     DropRoles,
+    FinalizeDropLogicalSubscriptions,
 }
 
 pub struct Operation {
@@ -331,7 +333,7 @@ async fn get_operations<'a>(
                             // NB: there could be other db states, which prevent us from dropping
                             // the database. For example, if db is used by any active subscription
                             // or replication slot.
-                            // Such cases are handled in the DropSubscriptionsForDeletedDatabases
+                            // Such cases are handled in the DropLogicalSubscriptions
                             // phase. We do all the cleanup before actually dropping the database.
                             let drop_db_query: String = format!(
                                 "DROP DATABASE IF EXISTS {} WITH (FORCE)",
@@ -442,13 +444,19 @@ async fn get_operations<'a>(
 
             Ok(Box::new(operations))
         }
+        ApplySpecPhase::CreateSchemaNeon => Ok(Box::new(once(Operation {
+            query: String::from("CREATE SCHEMA IF NOT EXISTS neon"),
+            comment: Some(String::from(
+                "create schema for neon extension and utils tables",
+            )),
+        }))),
         ApplySpecPhase::RunInEachDatabase { db, subphase } => {
             match subphase {
-                PerDatabasePhase::DropSubscriptionsForDeletedDatabases => {
+                PerDatabasePhase::DropLogicalSubscriptions => {
                     match &db {
                         DB::UserDB(db) => {
                             let drop_subscription_query: String = format!(
-                                include_str!("sql/drop_subscription_for_drop_dbs.sql"),
+                                include_str!("sql/drop_subscriptions.sql"),
                                 datname_str = escape_literal(&db.name),
                             );
 
@@ -666,10 +674,6 @@ async fn get_operations<'a>(
         }
         ApplySpecPhase::HandleNeonExtension => {
             let operations = vec![
-                Operation {
-                    query: String::from("CREATE SCHEMA IF NOT EXISTS neon"),
-                    comment: Some(String::from("init: add schema for extension")),
-                },
                 Operation {
                     query: String::from("CREATE EXTENSION IF NOT EXISTS neon WITH SCHEMA neon"),
                     comment: Some(String::from(
@@ -712,5 +716,9 @@ async fn get_operations<'a>(
 
             Ok(Box::new(operations))
         }
+        ApplySpecPhase::FinalizeDropLogicalSubscriptions => Ok(Box::new(once(Operation {
+            query: String::from(include_str!("sql/finalize_drop_subscriptions.sql")),
+            comment: None,
+        }))),
     }
 }

compute_tools/src/sql/finalize_drop_subscriptions.sql

@@ -0,0 +1,21 @@
+DO $$
+BEGIN
+    IF NOT EXISTS(
+        SELECT 1
+        FROM pg_catalog.pg_tables
+        WHERE tablename = 'drop_subscriptions_done'
+        AND schemaname = 'neon'
+    )
+    THEN
+        CREATE TABLE neon.drop_subscriptions_done
+        (id serial primary key, timeline_id text);
+    END IF;
+
+    -- preserve the timeline_id of the last drop_subscriptions run
+    -- to ensure that the cleanup of a timeline is executed only once.
+    -- use upsert to avoid the table bloat in case of cascade branching (branch of a branch)
+    INSERT INTO neon.drop_subscriptions_done VALUES (1, current_setting('neon.timeline_id'))
+    ON CONFLICT (id) DO UPDATE
+    SET timeline_id = current_setting('neon.timeline_id');
+END
+$$

control_plane/README.md

@@ -1,6 +1,10 @@
-# Control Plane and Neon Local
+# Local Development Control Plane (`neon_local`)
 
-This crate contains tools to start a Neon development environment locally. This utility can be used with the `cargo neon` command.
+This crate contains tools to start a Neon development environment locally. This utility can be used with the `cargo neon` command.  This is a convenience to invoke
+the `neon_local` binary.
+
+**Note**: this is a dev/test tool -- a minimal control plane suitable for testing
+code changes locally, but not suitable for running production systems.
 
 ## Example: Start with Postgres 16
 

control_plane/src/bin/neon_local.rs

@@ -1357,6 +1357,7 @@ async fn handle_endpoint(subcmd: &EndpointCmd, env: &local_env::LocalEnv) -> Res
                 args.pg_version,
                 mode,
                 !args.update_catalog,
+                false,
             )?;
         }
         EndpointCmd::Start(args) => {

control_plane/src/endpoint.rs

@@ -76,6 +76,7 @@ pub struct EndpointConf {
     http_port: u16,
     pg_version: u32,
     skip_pg_catalog_updates: bool,
+    drop_subscriptions_before_start: bool,
     features: Vec<ComputeFeature>,
 }
 
@@ -143,6 +144,7 @@ impl ComputeControlPlane {
         pg_version: u32,
         mode: ComputeMode,
         skip_pg_catalog_updates: bool,
+        drop_subscriptions_before_start: bool,
     ) -> Result<Arc<Endpoint>> {
         let pg_port = pg_port.unwrap_or_else(|| self.get_port());
         let http_port = http_port.unwrap_or_else(|| self.get_port() + 1);
@@ -162,6 +164,7 @@ impl ComputeControlPlane {
             // with this we basically test a case of waking up an idle compute, where
             // we also skip catalog updates in the cloud.
             skip_pg_catalog_updates,
+            drop_subscriptions_before_start,
             features: vec![],
         });
 
@@ -177,6 +180,7 @@ impl ComputeControlPlane {
                 pg_port,
                 pg_version,
                 skip_pg_catalog_updates,
+                drop_subscriptions_before_start,
                 features: vec![],
             })?,
         )?;
@@ -240,6 +244,7 @@ pub struct Endpoint {
     // Optimizations
     skip_pg_catalog_updates: bool,
 
+    drop_subscriptions_before_start: bool,
     // Feature flags
     features: Vec<ComputeFeature>,
 }
@@ -291,6 +296,7 @@ impl Endpoint {
             tenant_id: conf.tenant_id,
             pg_version: conf.pg_version,
             skip_pg_catalog_updates: conf.skip_pg_catalog_updates,
+            drop_subscriptions_before_start: conf.drop_subscriptions_before_start,
             features: conf.features,
         })
     }
@@ -625,6 +631,7 @@ impl Endpoint {
             shard_stripe_size: Some(shard_stripe_size),
             local_proxy_config: None,
             reconfigure_concurrency: 1,
+            drop_subscriptions_before_start: self.drop_subscriptions_before_start,
         };
         let spec_path = self.endpoint_path().join("spec.json");
         std::fs::write(spec_path, serde_json::to_string_pretty(&spec)?)?;

control_plane/src/pageserver.rs

@@ -347,11 +347,31 @@ impl PageServerNode {
                 .map(|x| x.parse::<usize>())
                 .transpose()
                 .context("Failed to parse 'compaction_threshold' as an integer")?,
+            compaction_upper_limit: settings
+                .remove("compaction_upper_limit")
+                .map(|x| x.parse::<usize>())
+                .transpose()
+                .context("Failed to parse 'compaction_upper_limit' as an integer")?,
             compaction_algorithm: settings
                 .remove("compaction_algorithm")
                 .map(serde_json::from_str)
                 .transpose()
                 .context("Failed to parse 'compaction_algorithm' json")?,
+            l0_flush_delay_threshold: settings
+                .remove("l0_flush_delay_threshold")
+                .map(|x| x.parse::<usize>())
+                .transpose()
+                .context("Failed to parse 'l0_flush_delay_threshold' as an integer")?,
+            l0_flush_wait_upload: settings
+                .remove("l0_flush_wait_upload")
+                .map(|x| x.parse::<bool>())
+                .transpose()
+                .context("Failed to parse 'l0_flush_wait_upload' as a boolean")?,
+            l0_flush_stall_threshold: settings
+                .remove("l0_flush_stall_threshold")
+                .map(|x| x.parse::<usize>())
+                .transpose()
+                .context("Failed to parse 'l0_flush_stall_threshold' as an integer")?,
             gc_horizon: settings
                 .remove("gc_horizon")
                 .map(|x| x.parse::<u64>())
@@ -368,6 +388,11 @@ impl PageServerNode {
                 .map(|x| x.parse::<u8>())
                 .transpose()
                 .context("Failed to parse 'image_creation_check_threshold' as integer")?,
+            image_creation_preempt_threshold: settings
+                .remove("image_creation_preempt_threshold")
+                .map(|x| x.parse::<usize>())
+                .transpose()
+                .context("Failed to parse 'image_creation_preempt_threshold' as integer")?,
             pitr_interval: settings.remove("pitr_interval").map(|x| x.to_string()),
             walreceiver_connect_timeout: settings
                 .remove("walreceiver_connect_timeout")
@@ -418,6 +443,26 @@ impl PageServerNode {
                 .map(serde_json::from_str)
                 .transpose()
                 .context("parse `wal_receiver_protocol_override` from json")?,
+            rel_size_v2_enabled: settings
+                .remove("rel_size_v2_enabled")
+                .map(|x| x.parse::<bool>())
+                .transpose()
+                .context("Failed to parse 'rel_size_v2_enabled' as bool")?,
+            gc_compaction_enabled: settings
+                .remove("gc_compaction_enabled")
+                .map(|x| x.parse::<bool>())
+                .transpose()
+                .context("Failed to parse 'gc_compaction_enabled' as bool")?,
+            gc_compaction_initial_threshold_kb: settings
+                .remove("gc_compaction_initial_threshold_kb")
+                .map(|x| x.parse::<u64>())
+                .transpose()
+                .context("Failed to parse 'gc_compaction_initial_threshold_kb' as integer")?,
+            gc_compaction_ratio_percent: settings
+                .remove("gc_compaction_ratio_percent")
+                .map(|x| x.parse::<u64>())
+                .transpose()
+                .context("Failed to parse 'gc_compaction_ratio_percent' as integer")?,
         };
         if !settings.is_empty() {
             bail!("Unrecognized tenant settings: {settings:?}")

control_plane/storcon_cli/src/main.rs

@@ -10,8 +10,8 @@ use pageserver_api::{
     controller_api::{
         AvailabilityZone, NodeAvailabilityWrapper, NodeDescribeResponse, NodeShardResponse,
         SafekeeperDescribeResponse, SafekeeperSchedulingPolicyRequest, ShardSchedulingPolicy,
-        ShardsPreferredAzsRequest, SkSchedulingPolicy, TenantCreateRequest, TenantDescribeResponse,
-        TenantPolicyRequest,
+        ShardsPreferredAzsRequest, ShardsPreferredAzsResponse, SkSchedulingPolicy,
+        TenantCreateRequest, TenantDescribeResponse, TenantPolicyRequest,
     },
     models::{
         EvictionPolicy, EvictionPolicyLayerAccessThreshold, LocationConfigSecondary,
@@ -800,7 +800,7 @@ async fn main() -> anyhow::Result<()> {
                     .collect(),
             };
             storcon_client
-                .dispatch::<ShardsPreferredAzsRequest, ()>(
+                .dispatch::<ShardsPreferredAzsRequest, ShardsPreferredAzsResponse>(
                     Method::PUT,
                     "control/v1/preferred_azs".to_string(),
                     Some(req),

deny.toml

@@ -32,6 +32,7 @@ reason = "the marvin attack only affects private key decryption, not public key
 # https://embarkstudios.github.io/cargo-deny/checks/licenses/cfg.html
 [licenses]
 allow = [
+    "0BSD",
     "Apache-2.0",
     "BSD-2-Clause",
     "BSD-3-Clause",
@@ -41,8 +42,8 @@ allow = [
     "MIT",
     "MPL-2.0",
     "OpenSSL",
-    "Unicode-DFS-2016",
     "Unicode-3.0",
+    "Zlib",
 ]
 confidence-threshold = 0.8
 exceptions = [

docker-compose/compute_wrapper/Dockerfile

@@ -7,14 +7,12 @@ FROM $REPOSITORY/${COMPUTE_IMAGE}:$TAG
 ARG COMPUTE_IMAGE
 
 USER root
-RUN apt-get update &&       \
+RUN echo 'Acquire::Retries "5";' > /etc/apt/apt.conf.d/80-retries && \
+    apt-get update &&       \
     apt-get install -y curl \
                        jq   \
-                       python3-pip \
                        netcat-openbsd
-#Faker is required for the pg_anon test
-RUN case $COMPUTE_IMAGE in compute-node-v17) OPT="--break-system-packages";; *) OPT= ;; esac && pip3 install $OPT Faker
 #This is required for the pg_hintplan test
-RUN mkdir -p /ext-src/pg_hint_plan-src && chown postgres /ext-src/pg_hint_plan-src 
+RUN mkdir -p /ext-src/pg_hint_plan-src /postgres/contrib/file_fdw && chown postgres /ext-src/pg_hint_plan-src /postgres/contrib/file_fdw
 
-USER postgres
+USER postgres

docker-compose/compute_wrapper/shell/compute.sh

@@ -20,30 +20,55 @@ while ! nc -z pageserver 6400; do
 done
 echo "Page server is ready."
 
-echo "Create a tenant and timeline"
-generate_id tenant_id
-PARAMS=(
-     -X PUT
-     -H "Content-Type: application/json"
-     -d "{\"mode\": \"AttachedSingle\", \"generation\": 1, \"tenant_conf\": {}}"
-     "http://pageserver:9898/v1/tenant/${tenant_id}/location_config"
-)
-result=$(curl "${PARAMS[@]}")
-echo $result | jq .
+cp ${SPEC_FILE_ORG} ${SPEC_FILE}
 
-generate_id timeline_id
-PARAMS=(
-     -sbf
-     -X POST
-     -H "Content-Type: application/json"
-     -d "{\"new_timeline_id\": \"${timeline_id}\", \"pg_version\": ${PG_VERSION}}"
-     "http://pageserver:9898/v1/tenant/${tenant_id}/timeline/"
-)
-result=$(curl "${PARAMS[@]}")
-echo $result | jq .
+ if [ -n "${TENANT_ID:-}" ] && [ -n "${TIMELINE_ID:-}" ]; then
+   tenant_id=${TENANT_ID}
+   timeline_id=${TIMELINE_ID}
+else
+  echo "Check if a tenant present"
+  PARAMS=(
+       -X GET
+       -H "Content-Type: application/json"
+       "http://pageserver:9898/v1/tenant"
+  )
+  tenant_id=$(curl "${PARAMS[@]}" | jq -r .[0].id)
+  if [ -z "${tenant_id}" ] || [ "${tenant_id}" = null ]; then
+    echo "Create a tenant"
+    generate_id tenant_id
+    PARAMS=(
+         -X PUT
+         -H "Content-Type: application/json"
+         -d "{\"mode\": \"AttachedSingle\", \"generation\": 1, \"tenant_conf\": {}}"
+        "http://pageserver:9898/v1/tenant/${tenant_id}/location_config"
+    )
+    result=$(curl "${PARAMS[@]}")
+    echo $result | jq .
+  fi
+
+  echo "Check if a timeline present"
+  PARAMS=(
+       -X GET
+       -H "Content-Type: application/json"
+       "http://pageserver:9898/v1/tenant/${tenant_id}/timeline"
+  )
+  timeline_id=$(curl "${PARAMS[@]}" | jq -r .[0].timeline_id)
+  if [ -z "${timeline_id}" ] || [ "${timeline_id}" = null ]; then
+    generate_id timeline_id
+    PARAMS=(
+        -sbf
+        -X POST
+        -H "Content-Type: application/json"
+        -d "{\"new_timeline_id\": \"${timeline_id}\", \"pg_version\": ${PG_VERSION}}"
+        "http://pageserver:9898/v1/tenant/${tenant_id}/timeline/"
+    )
+    result=$(curl "${PARAMS[@]}")
+    echo $result | jq .
+  fi
+fi
 
 echo "Overwrite tenant id and timeline id in spec file"
-sed "s/TENANT_ID/${tenant_id}/" ${SPEC_FILE_ORG} > ${SPEC_FILE}
+sed -i "s/TENANT_ID/${tenant_id}/" ${SPEC_FILE}
 sed -i "s/TIMELINE_ID/${timeline_id}/" ${SPEC_FILE}
 
 cat ${SPEC_FILE}

docker-compose/docker-compose.yml

@@ -149,11 +149,13 @@ services:
       args:
         - REPOSITORY=${REPOSITORY:-neondatabase}
         - COMPUTE_IMAGE=compute-node-v${PG_VERSION:-16}
-        - TAG=${TAG:-latest}
-        - http_proxy=$http_proxy
-        - https_proxy=$https_proxy
+        - TAG=${COMPUTE_TAG:-${TAG:-latest}}
+        - http_proxy=${http_proxy:-}
+        - https_proxy=${https_proxy:-}
     environment:
       - PG_VERSION=${PG_VERSION:-16}
+      - TENANT_ID=${TENANT_ID:-}
+      - TIMELINE_ID=${TIMELINE_ID:-}
       #- RUST_BACKTRACE=1
     # Mount the test files directly, for faster editing cycle.
     volumes:
@@ -185,6 +187,8 @@ services:
   neon-test-extensions:
     profiles: ["test-extensions"]
     image: ${REPOSITORY:-neondatabase}/neon-test-extensions-v${PG_TEST_VERSION:-16}:${TAG:-latest}
+    environment:
+      - PGPASSWORD=cloud_admin
     entrypoint:
       - "/bin/bash"
       - "-c"

docker-compose/docker_compose_test.sh

@@ -18,14 +18,10 @@ cd $(dirname $0)
 COMPUTE_CONTAINER_NAME=docker-compose-compute-1
 TEST_CONTAINER_NAME=docker-compose-neon-test-extensions-1
 PSQL_OPTION="-h localhost -U cloud_admin -p 55433 -d postgres"
-: ${http_proxy:=}
-: ${https_proxy:=}
-export http_proxy https_proxy
 
 cleanup() {
     echo "show container information"
     docker ps
-    docker compose --profile test-extensions -f $COMPOSE_FILE logs
     echo "stop containers..."
     docker compose --profile test-extensions -f $COMPOSE_FILE down
 }
@@ -35,13 +31,7 @@ for pg_version in ${TEST_VERSION_ONLY-14 15 16 17}; do
     echo "clean up containers if exists"
     cleanup
     PG_TEST_VERSION=$((pg_version < 16 ? 16 : pg_version))
-    # The support of pg_anon not yet added to PG17, so we have to add the corresponding option for other PG versions
-    if [ "${pg_version}" -ne 17 ]; then
-      SPEC_PATH="compute_wrapper/var/db/postgres/specs"
-      mv $SPEC_PATH/spec.json $SPEC_PATH/spec.bak
-      jq '.cluster.settings += [{"name": "session_preload_libraries","value": "anon","vartype": "string"}]' "${SPEC_PATH}/spec.bak" > "${SPEC_PATH}/spec.json"
-    fi
-    PG_VERSION=$pg_version PG_TEST_VERSION=$PG_TEST_VERSION docker compose --profile test-extensions -f $COMPOSE_FILE up --build -d
+    PG_VERSION=$pg_version PG_TEST_VERSION=$PG_TEST_VERSION docker compose --profile test-extensions -f $COMPOSE_FILE up --quiet-pull --build -d
 
     echo "wait until the compute is ready. timeout after 60s. "
     cnt=0
@@ -50,7 +40,6 @@ for pg_version in ${TEST_VERSION_ONLY-14 15 16 17}; do
         cnt=`expr $cnt + 3`
         if [ $cnt -gt 60 ]; then
             echo "timeout before the compute is ready."
-            cleanup
             exit 1
         fi
         if docker compose --profile test-extensions -f $COMPOSE_FILE logs "compute_is_ready" | grep -q "accepting connections"; then
@@ -62,52 +51,46 @@ for pg_version in ${TEST_VERSION_ONLY-14 15 16 17}; do
     done
 
     if [ $pg_version -ge 16 ]; then
-        echo Enabling trust connection
-        docker exec $COMPUTE_CONTAINER_NAME bash -c "sed -i '\$d' /var/db/postgres/compute/pg_hba.conf && echo -e 'host\t all\t all\t all\t trust' >> /var/db/postgres/compute/pg_hba.conf && psql $PSQL_OPTION -c 'select pg_reload_conf()' "
-        echo Adding postgres role
-        docker exec $COMPUTE_CONTAINER_NAME psql $PSQL_OPTION -c "CREATE ROLE postgres SUPERUSER LOGIN"
+        docker cp ext-src $TEST_CONTAINER_NAME:/
+        docker exec $TEST_CONTAINER_NAME bash -c "apt update && apt install -y libtap-parser-sourcehandler-pgtap-perl"
         # This is required for the pg_hint_plan test, to prevent flaky log message causing the test to fail
         # It cannot be moved to Dockerfile now because the database directory is created after the start of the container
         echo Adding dummy config
         docker exec $COMPUTE_CONTAINER_NAME touch /var/db/postgres/compute/compute_ctl_temp_override.conf
-        # This block is required for the pg_anon extension test.
-        # The test assumes that it is running on the same host with the postgres engine.
-        # In our case it's not true, that's why we are copying files to the compute node
+        # The following block copies the files for the pg_hintplan test to the compute node for the extension test in an isolated docker-compose environment
         TMPDIR=$(mktemp -d)
-        # Add support for pg_anon for pg_v16
-        if [ $pg_version -ne 17 ]; then
-          docker cp $TEST_CONTAINER_NAME:/ext-src/pg_anon-src/data $TMPDIR/data
-          echo -e '1\t too \t many \t tabs' > $TMPDIR/data/bad.csv
-          docker cp $TMPDIR/data $COMPUTE_CONTAINER_NAME:/tmp/tmp_anon_alternate_data
-        rm -rf $TMPDIR
-        fi
-        TMPDIR=$(mktemp -d)
-        # The following block does the same for the pg_hintplan test
         docker cp $TEST_CONTAINER_NAME:/ext-src/pg_hint_plan-src/data $TMPDIR/data
         docker cp $TMPDIR/data $COMPUTE_CONTAINER_NAME:/ext-src/pg_hint_plan-src/
         rm -rf $TMPDIR
+        # The following block does the same for the contrib/file_fdw test
+        TMPDIR=$(mktemp -d)
+        docker cp $TEST_CONTAINER_NAME:/postgres/contrib/file_fdw/data $TMPDIR/data
+        docker cp $TMPDIR/data $COMPUTE_CONTAINER_NAME:/postgres/contrib/file_fdw/data
+        rm -rf $TMPDIR
+        # Apply patches
+        cat ../compute/patches/contrib_pg${pg_version}.patch | docker exec -i $TEST_CONTAINER_NAME bash -c "(cd /postgres && patch -p1)"
         # We are running tests now
-        if docker exec -e SKIP=timescaledb-src,rdkit-src,postgis-src,pgx_ulid-src,pgtap-src,pg_tiktoken-src,pg_jsonschema-src,pg_graphql-src,kq_imcx-src,wal2json_2_5-src \
-            $TEST_CONTAINER_NAME /run-tests.sh | tee testout.txt
-        then
-            cleanup
-        else
-            FAILED=$(tail -1 testout.txt)
-            for d in $FAILED
-            do
-                mkdir $d
-                docker cp $TEST_CONTAINER_NAME:/ext-src/$d/regression.diffs $d || true
-                docker cp $TEST_CONTAINER_NAME:/ext-src/$d/regression.out $d || true
-                cat $d/regression.out $d/regression.diffs || true
+        rm -f testout.txt testout_contrib.txt
+        docker exec -e USE_PGXS=1 -e SKIP=timescaledb-src,rdkit-src,postgis-src,pgx_ulid-src,pgtap-src,pg_tiktoken-src,pg_jsonschema-src,kq_imcx-src,wal2json_2_5-src \
+        $TEST_CONTAINER_NAME /run-tests.sh /ext-src | tee testout.txt && EXT_SUCCESS=1 || EXT_SUCCESS=0
+        docker exec -e SKIP=start-scripts,postgres_fdw,ltree_plpython,jsonb_plpython,jsonb_plperl,hstore_plpython,hstore_plperl,dblink,bool_plperl \
+        $TEST_CONTAINER_NAME /run-tests.sh /postgres/contrib | tee testout_contrib.txt && CONTRIB_SUCCESS=1 || CONTRIB_SUCCESS=0
+        if [ $EXT_SUCCESS -eq 0 ] || [ $CONTRIB_SUCCESS -eq 0 ]; then
+            CONTRIB_FAILED=
+            FAILED=
+            [ $EXT_SUCCESS -eq 0 ] && FAILED=$(tail -1 testout.txt | awk '{for(i=1;i<=NF;i++){print "/ext-src/"$i;}}')
+            [ $CONTRIB_SUCCESS -eq 0 ] && CONTRIB_FAILED=$(tail -1 testout_contrib.txt | awk '{for(i=0;i<=NF;i++){print "/postgres/contrib/"$i;}}')
+            for d in $FAILED $CONTRIB_FAILED; do
+                dn="$(basename $d)"
+                rm -rf $dn
+                mkdir $dn
+                docker cp $TEST_CONTAINER_NAME:$d/regression.diffs $dn || [ $? -eq 1 ]
+                docker cp $TEST_CONTAINER_NAME:$d/regression.out $dn || [ $? -eq 1 ]
+                cat $dn/regression.out $dn/regression.diffs || true
+                rm -rf $dn
             done
         rm -rf $FAILED
-        cleanup
         exit 1
         fi
     fi
-    cleanup
-    # Restore the original spec.json
-    if [ "$pg_version" -ne 17 ]; then
-      mv "$SPEC_PATH/spec.bak" "$SPEC_PATH/spec.json"
-    fi
 done

docker-compose/ext-src/hll-src/test-upgrade.sh

@@ -0,0 +1,5 @@
+#!/bin/sh
+set -ex
+cd "$(dirname ${0})"
+PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
+${PG_REGRESS} --use-existing --inputdir=./ --bindir='/usr/local/pgsql/bin'    --dbname=contrib_regression add_agg agg_oob auto_sparse card_op cast_shape copy_binary cumulative_add_cardinality_correction cumulative_add_comprehensive_promotion cumulative_add_sparse_edge cumulative_add_sparse_random cumulative_add_sparse_step cumulative_union_comprehensive cumulative_union_explicit_explicit cumulative_union_explicit_promotion cumulative_union_probabilistic_probabilistic cumulative_union_sparse_full_representation cumulative_union_sparse_promotion cumulative_union_sparse_sparse disable_hashagg equal explicit_thresh hash hash_any meta_func murmur_bigint murmur_bytea nosparse notequal scalar_oob storedproc transaction typmod typmod_insert union_op

docker-compose/ext-src/hypopg-src/test-upgrade.patch

@@ -0,0 +1,27 @@
+diff --git a/expected/hypopg.out b/expected/hypopg.out
+index 90121d0..859260b 100644
+--- a/expected/hypopg.out
++++ b/expected/hypopg.out
+@@ -11,7 +11,8 @@ BEGIN
+ END;
+ $_$
+ LANGUAGE plpgsql;
+-CREATE EXTENSION hypopg;
++CREATE EXTENSION IF NOT EXISTS hypopg;
++NOTICE:  extension "hypopg" already exists, skipping
+ CREATE TABLE hypo (id integer, val text, "Id2" bigint);
+ INSERT INTO hypo SELECT i, 'line ' || i
+ FROM generate_series(1,100000) f(i);
+diff --git a/test/sql/hypopg.sql b/test/sql/hypopg.sql
+index 99722b0..8d6bacb 100644
+--- a/test/sql/hypopg.sql
++++ b/test/sql/hypopg.sql
+@@ -12,7 +12,7 @@ END;
+ $_$
+ LANGUAGE plpgsql;
+
+-CREATE EXTENSION hypopg;
++CREATE EXTENSION IF NOT EXISTS hypopg;
+
+ CREATE TABLE hypo (id integer, val text, "Id2" bigint);
+

docker-compose/ext-src/hypopg-src/test-upgrade.sh

@@ -0,0 +1,6 @@
+#!/bin/sh
+set -ex
+cd "$(dirname ${0})"
+patch -p1 <test-upgrade.patch
+PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
+${PG_REGRESS} --inputdir=./ --bindir='/usr/local/pgsql/bin' --use-existing --inputdir=test --dbname=contrib_regression hypopg hypo_brin hypo_index_part hypo_include hypo_hash hypo_hide_index

docker-compose/ext-src/ip4r-src/test-upgrade.patch

@@ -0,0 +1,23 @@
+diff --git a/expected/ip4r.out b/expected/ip4r.out
+index 7527af3..b38ed29 100644
+--- a/expected/ip4r.out
++++ b/expected/ip4r.out
+@@ -1,6 +1,5 @@
+ --
+ /*CUT-HERE*/
+-CREATE EXTENSION ip4r;
+ -- Check whether any of our opclasses fail amvalidate
+ DO $d$
+   DECLARE
+diff --git a/sql/ip4r.sql b/sql/ip4r.sql
+index 65c49ec..24ade09 100644
+--- a/sql/ip4r.sql
++++ b/sql/ip4r.sql
+@@ -1,7 +1,6 @@
+ --
+
+ /*CUT-HERE*/
+-CREATE EXTENSION ip4r;
+
+ -- Check whether any of our opclasses fail amvalidate
+

docker-compose/ext-src/ip4r-src/test-upgrade.sh

@@ -0,0 +1,6 @@
+#!/bin/sh
+set -ex
+cd "$(dirname ${0})"
+patch -p1 <test-upgrade.patch
+PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
+${PG_REGRESS} --use-existing --inputdir=./ --bindir='/usr/local/pgsql/bin'    --dbname=contrib_regression ip4r ip4r-softerr ip4r-v11

docker-compose/ext-src/pg_cron-src/test-upgrade.patch

@@ -0,0 +1,75 @@
+diff --git a/expected/pg_cron-test.out b/expected/pg_cron-test.out
+index d79d542..1663886 100644
+--- a/expected/pg_cron-test.out
++++ b/expected/pg_cron-test.out
+@@ -1,30 +1,3 @@
+-CREATE EXTENSION pg_cron VERSION '1.0';
+-SELECT extversion FROM pg_extension WHERE extname='pg_cron';
+- extversion 
+-------------
+- 1.0
+-(1 row)
+-
+--- Test binary compatibility with v1.4 function signature.
+-ALTER EXTENSION pg_cron UPDATE TO '1.4';
+-SELECT cron.unschedule(job_name := 'no_such_job');
+-ERROR:  could not find valid entry for job 'no_such_job'
+-SELECT cron.schedule('testjob', '* * * * *', 'SELECT 1');
+- schedule 
+-----------
+-        1
+-(1 row)
+-
+-SELECT cron.unschedule('testjob');
+- unschedule 
+-------------
+- t
+-(1 row)
+-
+--- Test cache invalidation
+-DROP EXTENSION pg_cron;
+-CREATE EXTENSION pg_cron VERSION '1.4';
+-ALTER EXTENSION pg_cron UPDATE;
+ -- Vacuum every day at 10:00am (GMT)
+ SELECT cron.schedule('0 10 * * *', 'VACUUM');
+  schedule 
+@@ -300,8 +273,3 @@ SELECT jobid, jobname, schedule, command FROM cron.job ORDER BY jobid;
+ SELECT cron.schedule('bad-last-dom-job1', '0 11 $foo * *', 'VACUUM FULL');
+ ERROR:  invalid schedule: 0 11 $foo * *
+ HINT:  Use cron format (e.g. 5 4 * * *), or interval format '[1-59] seconds'
+--- cleaning
+-DROP EXTENSION pg_cron;
+-drop user pgcron_cront;
+-drop database pgcron_dbno;
+-drop database pgcron_dbyes;
+diff --git a/sql/pg_cron-test.sql b/sql/pg_cron-test.sql
+index 45f94d9..241cf73 100644
+--- a/sql/pg_cron-test.sql
++++ b/sql/pg_cron-test.sql
+@@ -1,17 +1,3 @@
+-CREATE EXTENSION pg_cron VERSION '1.0';
+-SELECT extversion FROM pg_extension WHERE extname='pg_cron';
+--- Test binary compatibility with v1.4 function signature.
+-ALTER EXTENSION pg_cron UPDATE TO '1.4';
+-SELECT cron.unschedule(job_name := 'no_such_job');
+-SELECT cron.schedule('testjob', '* * * * *', 'SELECT 1');
+-SELECT cron.unschedule('testjob');
+-
+--- Test cache invalidation
+-DROP EXTENSION pg_cron;
+-CREATE EXTENSION pg_cron VERSION '1.4';
+-
+-ALTER EXTENSION pg_cron UPDATE;
+-
+ -- Vacuum every day at 10:00am (GMT)
+ SELECT cron.schedule('0 10 * * *', 'VACUUM');
+ 
+@@ -156,8 +142,3 @@ SELECT jobid, jobname, schedule, command FROM cron.job ORDER BY jobid;
+ -- invalid last of day job
+ SELECT cron.schedule('bad-last-dom-job1', '0 11 $foo * *', 'VACUUM FULL');
+ 
+--- cleaning
+-DROP EXTENSION pg_cron;
+-drop user pgcron_cront;
+-drop database pgcron_dbno;
+-drop database pgcron_dbyes;

docker-compose/ext-src/pg_cron-src/test-upgrade.sh

@@ -0,0 +1,6 @@
+#!/bin/sh
+set -ex
+cd "$(dirname ${0})"
+patch -p1 <test-upgrade.patch
+PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
+${PG_REGRESS} --use-existing --inputdir=./ --bindir='/usr/local/pgsql/bin'    --dbname=contrib_regression pg_cron-test

docker-compose/ext-src/pg_graphql-src/neon-test.sh

@@ -0,0 +1,13 @@
+#!/bin/bash
+set -ex
+cd "$(dirname "${0}")"
+dropdb --if-exists contrib_regression
+createdb contrib_regression
+PGXS="$(dirname "$(pg_config --pgxs)" )"
+REGRESS="${PGXS}/../test/regress/pg_regress"
+TESTDIR="test"
+TESTS=$(ls "${TESTDIR}/sql" | sort )
+TESTS=${TESTS//\.sql/}
+psql -v ON_ERROR_STOP=1 -f test/fixtures.sql -d contrib_regression
+${REGRESS} --use-existing --dbname=contrib_regression --inputdir=${TESTDIR} ${TESTS}
+

docker-compose/ext-src/pg_ivm-src/test-upgrade.patch

@@ -0,0 +1,18 @@
+diff --git a/expected/pg_ivm.out b/expected/pg_ivm.out
+index e8798ee..cca58d0 100644
+--- a/expected/pg_ivm.out
++++ b/expected/pg_ivm.out
+@@ -1,4 +1,3 @@
+-CREATE EXTENSION pg_ivm;
+ GRANT ALL ON SCHEMA public TO public;
+ -- create a table to use as a basis for views and materialized views in various combinations
+ CREATE TABLE mv_base_a (i int, j int);
+diff --git a/sql/pg_ivm.sql b/sql/pg_ivm.sql
+index d3c1a01..9382d7f 100644
+--- a/sql/pg_ivm.sql
++++ b/sql/pg_ivm.sql
+@@ -1,4 +1,3 @@
+-CREATE EXTENSION pg_ivm;
+ GRANT ALL ON SCHEMA public TO public;
+ 
+ -- create a table to use as a basis for views and materialized views in various combinations

docker-compose/ext-src/pg_ivm-src/test-upgrade.sh

@@ -0,0 +1,6 @@
+#!/bin/sh
+set -ex
+cd "$(dirname ${0})"
+patch -p1 <test-upgrade.patch
+PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
+${PG_REGRESS} --use-existing --inputdir=./ --bindir='/usr/local/pgsql/bin' --dbname=contrib_regression pg_ivm create_immv refresh_immv

docker-compose/ext-src/pg_roaringbitmap-src/test-upgrade.patch

@@ -0,0 +1,25 @@
+diff --git a/expected/roaringbitmap.out b/expected/roaringbitmap.out
+index de70531..a5f7c15 100644
+--- a/expected/roaringbitmap.out
++++ b/expected/roaringbitmap.out
+@@ -1,7 +1,6 @@
+ --
+ --  Test roaringbitmap extension
+ --
+-CREATE EXTENSION if not exists roaringbitmap;
+ -- Test input and output
+ set roaringbitmap.output_format='array';
+ set extra_float_digits = 0;
+diff --git a/sql/roaringbitmap.sql b/sql/roaringbitmap.sql
+index a0e9c74..84bc966 100644
+--- a/sql/roaringbitmap.sql
++++ b/sql/roaringbitmap.sql
+@@ -2,8 +2,6 @@
+ --  Test roaringbitmap extension
+ --
+ 
+-CREATE EXTENSION if not exists roaringbitmap;
+-
+ -- Test input and output
+ 
+ set roaringbitmap.output_format='array';

docker-compose/ext-src/pg_roaringbitmap-src/test-upgrade.sh

@@ -0,0 +1,6 @@
+#!/bin/sh
+set -ex
+cd "$(dirname ${0})"
+patch -p1 <test-upgrade.patch
+PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
+${PG_REGRESS} --use-existing --inputdir=./ --bindir='/usr/local/pgsql/bin'    --dbname=contrib_regression roaringbitmap

docker-compose/ext-src/pg_semver-src/test-upgrade.patch

@@ -0,0 +1,24 @@
+diff --git a/test/sql/base.sql b/test/sql/base.sql
+index af599d8..2eed91b 100644
+--- a/test/sql/base.sql
++++ b/test/sql/base.sql
+@@ -2,7 +2,6 @@
+ BEGIN;
+ 
+ \i test/pgtap-core.sql
+-\i sql/semver.sql
+ 
+ SELECT plan(334);
+ --SELECT * FROM no_plan();
+diff --git a/test/sql/corpus.sql b/test/sql/corpus.sql
+index 1f5f637..a519905 100644
+--- a/test/sql/corpus.sql
++++ b/test/sql/corpus.sql
+@@ -4,7 +4,6 @@ BEGIN;
+ -- Test the SemVer corpus from https://regex101.com/r/Ly7O1x/3/.
+ 
+ \i test/pgtap-core.sql
+-\i sql/semver.sql
+ 
+ SELECT plan(71);
+ --SELECT * FROM no_plan();

docker-compose/ext-src/pg_semver-src/test-upgrade.sh

@@ -0,0 +1,6 @@
+#!/bin/sh
+set -ex
+cd "$(dirname ${0})"
+patch -p1 <test-upgrade.patch
+PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
+${PG_REGRESS} --use-existing --inputdir=./ --bindir='/usr/local/pgsql/bin'    --inputdir=test --dbname=contrib_regression base corpus

docker-compose/ext-src/pg_uuidv7-src/test-upgrade.sh

@@ -0,0 +1,5 @@
+#!/bin/sh
+set -ex
+cd "$(dirname ${0})"
+PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
+${PG_REGRESS} --use-existing --inputdir=./ --bindir='/usr/local/pgsql/bin'    --inputdir=test --dbname=contrib_regression  002_uuid_generate_v7 003_uuid_v7_to_timestamptz 004_uuid_timestamptz_to_v7 005_uuid_v7_to_timestamp 006_uuid_timestamp_to_v7

docker-compose/ext-src/pgjwt-src/neon-test.sh

@@ -0,0 +1,4 @@
+#!/bin/bash
+set -ex
+cd "$(dirname "${0}")"
+pg_prove test.sql

docker-compose/ext-src/pgjwt-src/test-upgrade.patch

@@ -0,0 +1,15 @@
+diff --git a/test.sql b/test.sql
+index d7a0ca8..f15bc76 100644
+--- a/test.sql
++++ b/test.sql
+@@ -9,9 +9,7 @@
+ \set ON_ERROR_STOP true
+ \set QUIET 1
+ 
+-CREATE EXTENSION pgcrypto;
+-CREATE EXTENSION pgtap;
+-CREATE EXTENSION pgjwt;
++CREATE EXTENSION IF NOT EXISTS pgtap;
+ 
+ BEGIN;
+ SELECT plan(23);

docker-compose/ext-src/pgjwt-src/test-upgrade.sh

@@ -0,0 +1,5 @@
+#!/bin/sh
+set -ex
+cd "$(dirname ${0})"
+patch -p1 <test-upgrade.patch
+pg_prove test.sql

docker-compose/ext-src/pgvector-src/test-upgrade.sh

@@ -0,0 +1,5 @@
+#!/bin/sh
+set -ex
+cd "$(dirname ${0})"
+PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
+${PG_REGRESS} --inputdir=./ --bindir='/usr/local/pgsql/bin' --inputdir=test --use-existing --dbname=contrib_regression bit btree cast copy halfvec hnsw_bit hnsw_halfvec hnsw_sparsevec hnsw_vector ivfflat_bit ivfflat_halfvec ivfflat_vector sparsevec vector_type

docker-compose/ext-src/plv8-src/test-upgrade.sh

@@ -0,0 +1,5 @@
+#!/bin/sh
+set -ex
+cd "$(dirname ${0})"
+PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
+${PG_REGRESS} --inputdir=./ --bindir='/usr/local/pgsql/bin'  --use-existing --dbname=contrib_regression plv8 plv8-errors scalar_args inline json startup_pre startup varparam json_conv jsonb_conv window guc es6 arraybuffer composites currentresource startup_perms bytea find_function_perms memory_limits reset show array_spread regression dialect bigint procedure

docker-compose/ext-src/postgresql-unit-src/test-upgrade.sh

@@ -0,0 +1,5 @@
+#!/bin/sh
+set -ex
+cd "$(dirname ${0})"
+PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
+${PG_REGRESS} --inputdir=./ --bindir='/usr/local/pgsql/bin' --use-existing --dbname=contrib_regression extension tables unit binary unicode prefix units time temperature functions language_functions round derived compare aggregate iec custom crosstab convert

docker-compose/ext-src/prefix-src/test-upgrade.sh

@@ -0,0 +1,5 @@
+#!/bin/sh
+set -ex
+cd "$(dirname ${0})"
+PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
+${PG_REGRESS} --use-existing --inputdir=./ --bindir='/usr/local/pgsql/bin'    --dbname=contrib_regression prefix falcon explain queries

docker-compose/ext-src/rum-src/test-upgrade.patch

@@ -0,0 +1,19 @@
+diff --git a/expected/rum.out b/expected/rum.out
+index 5966d19..8860b79 100644
+--- a/expected/rum.out
++++ b/expected/rum.out
+@@ -1,4 +1,3 @@
+-CREATE EXTENSION rum;
+ CREATE TABLE test_rum( t text, a tsvector );
+ CREATE TRIGGER tsvectorupdate
+ BEFORE UPDATE OR INSERT ON test_rum
+diff --git a/sql/rum.sql b/sql/rum.sql
+index 8414bb9..898e6ab 100644
+--- a/sql/rum.sql
++++ b/sql/rum.sql
+@@ -1,5 +1,3 @@
+-CREATE EXTENSION rum;
+-
+ CREATE TABLE test_rum( t text, a tsvector );
+
+ CREATE TRIGGER tsvectorupdate

docker-compose/ext-src/rum-src/test-upgrade.sh

@@ -0,0 +1,6 @@
+#!/bin/sh
+set -ex
+cd "$(dirname ${0})"
+patch -p1 <test-upgrade.patch
+PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
+${PG_REGRESS} --inputdir=./ --bindir='/usr/local/pgsql/bin' --use-existing --dbname=contrib_regression rum rum_validate rum_hash ruminv timestamp orderby orderby_hash altorder altorder_hash limits int2 int4 int8 float4 float8 money oid time timetz date interval macaddr inet cidr text varchar char bytea bit varbit numeric rum_weight expr array

docker-compose/run-tests.sh

@@ -1,14 +1,22 @@
 #!/bin/bash
 set -x
 
-cd /ext-src || exit 2
+extdir=${1}
+
+cd "${extdir}" || exit 2
 FAILED=
-LIST=$( (echo -e "${SKIP//","/"\n"}"; ls -d -- *-src) | sort | uniq -u)
-for d in ${LIST}
-do
-       [ -d "${d}" ] || continue
-    psql -c "select 1" >/dev/null || break
+LIST=$( (echo -e "${SKIP//","/"\n"}"; ls) | sort | uniq -u)
+for d in ${LIST}; do
+    [ -d "${d}" ] || continue
+    if ! psql -w -c "select 1" >/dev/null; then
+      FAILED="${d} ${FAILED}"
+      break
+    fi
+    if [ -f "${d}/neon-test.sh" ]; then
+       "${d}/neon-test.sh" || FAILED="${d} ${FAILED}"
+    else
        USE_PGXS=1 make -C "${d}" installcheck || FAILED="${d} ${FAILED}"
+    fi
 done
 [ -z "${FAILED}" ] && exit 0
 echo "${FAILED}"

docker-compose/test_extensions_upgrade.sh

@@ -0,0 +1,94 @@
+#!/bin/bash
+set -eux -o pipefail
+cd "$(dirname "${0}")"
+# Takes a variable name as argument. The result is stored in that variable.
+generate_id() {
+    local -n resvar=$1
+    printf -v resvar '%08x%08x%08x%08x' $SRANDOM $SRANDOM $SRANDOM $SRANDOM
+}
+if [ -z ${OLDTAG+x} ] || [ -z ${NEWTAG+x} ] || [ -z "${OLDTAG}" ] || [ -z "${NEWTAG}" ]; then
+  echo OLDTAG and NEWTAG must be defined
+  exit 1
+fi
+export PG_VERSION=${PG_VERSION:-16}
+function wait_for_ready {
+  TIME=0
+  while ! docker compose logs compute_is_ready | grep -q "accepting connections" && [ ${TIME} -le 300 ] ; do
+    ((TIME += 1 ))
+    sleep 1
+  done
+  if [ ${TIME} -gt 300 ]; then
+    echo Time is out.
+    exit 2
+  fi
+}
+function create_extensions() {
+  for ext in ${1}; do
+    docker compose exec neon-test-extensions psql -X -v ON_ERROR_STOP=1 -d contrib_regression -c "CREATE EXTENSION IF NOT EXISTS ${ext} CASCADE"
+  done
+}
+EXTENSIONS='[
+{"extname": "plv8", "extdir": "plv8-src"},
+{"extname": "vector", "extdir": "pgvector-src"},
+{"extname": "unit", "extdir": "postgresql-unit-src"},
+{"extname": "hypopg", "extdir": "hypopg-src"},
+{"extname": "rum", "extdir": "rum-src"},
+{"extname": "ip4r", "extdir": "ip4r-src"},
+{"extname": "prefix", "extdir": "prefix-src"},
+{"extname": "hll", "extdir": "hll-src"},
+{"extname": "pg_cron", "extdir": "pg_cron-src"},
+{"extname": "pg_uuidv7", "extdir": "pg_uuidv7-src"},
+{"extname": "roaringbitmap", "extdir": "pg_roaringbitmap-src"},
+{"extname": "semver", "extdir": "pg_semver-src"},
+{"extname": "pg_ivm", "extdir": "pg_ivm-src"},
+{"extname": "pgjwt", "extdir": "pgjwt-src"}
+]'
+EXTNAMES=$(echo ${EXTENSIONS} | jq -r '.[].extname' | paste -sd ' ' -)
+TAG=${NEWTAG} docker compose --profile test-extensions up --quiet-pull --build -d
+wait_for_ready
+docker compose exec neon-test-extensions psql -c "DROP DATABASE IF EXISTS contrib_regression"
+docker compose exec neon-test-extensions psql -c "CREATE DATABASE contrib_regression"
+create_extensions "${EXTNAMES}"
+query="select json_object_agg(extname,extversion) from pg_extension where extname in ('${EXTNAMES// /\',\'}')"
+new_vers=$(docker compose exec neon-test-extensions psql -Aqt -d contrib_regression -c "$query")
+docker compose --profile test-extensions down
+TAG=${OLDTAG} docker compose --profile test-extensions up --quiet-pull --build -d --force-recreate
+wait_for_ready
+docker compose cp  ext-src neon-test-extensions:/
+docker compose exec neon-test-extensions psql -c "DROP DATABASE IF EXISTS contrib_regression"
+docker compose exec neon-test-extensions psql -c "CREATE DATABASE contrib_regression"
+create_extensions "${EXTNAMES}"
+query="select pge.extname from pg_extension pge join (select key as extname, value as extversion from json_each_text('${new_vers}')) x on pge.extname=x.extname and pge.extversion <> x.extversion"
+exts=$(docker compose exec neon-test-extensions psql -Aqt -d contrib_regression -c "$query")
+if [ -z "${exts}" ]; then
+  echo "No extensions were upgraded"
+else
+  tenant_id=$(docker compose exec neon-test-extensions psql -Aqt -c "SHOW neon.tenant_id")
+  timeline_id=$(docker compose exec neon-test-extensions psql -Aqt -c "SHOW neon.timeline_id")
+  for ext in ${exts}; do
+    echo Testing ${ext}...
+    EXTDIR=$(echo ${EXTENSIONS} | jq -r '.[] | select(.extname=="'${ext}'") | .extdir')
+    generate_id new_timeline_id
+    PARAMS=(
+        -sbf
+        -X POST
+        -H "Content-Type: application/json"
+        -d "{\"new_timeline_id\": \"${new_timeline_id}\", \"pg_version\": ${PG_VERSION}, \"ancestor_timeline_id\": \"${timeline_id}\"}"
+        "http://127.0.0.1:9898/v1/tenant/${tenant_id}/timeline/"
+    )
+    result=$(curl "${PARAMS[@]}")
+    echo $result | jq .
+    TENANT_ID=${tenant_id} TIMELINE_ID=${new_timeline_id} TAG=${OLDTAG} docker compose down compute compute_is_ready
+    COMPUTE_TAG=${NEWTAG} TAG=${OLDTAG} TENANT_ID=${tenant_id} TIMELINE_ID=${new_timeline_id} docker compose up --quiet-pull -d --build compute compute_is_ready
+    wait_for_ready
+    TID=$(docker compose exec neon-test-extensions psql -Aqt -c "SHOW neon.timeline_id")
+    if [ ${TID} != ${new_timeline_id} ]; then
+      echo Timeline mismatch
+      exit 1
+    fi
+    docker compose exec neon-test-extensions psql -d contrib_regression -c "\dx ${ext}"
+    docker compose exec neon-test-extensions sh -c /ext-src/${EXTDIR}/test-upgrade.sh
+    docker compose exec neon-test-extensions psql -d contrib_regression -c "alter extension ${ext} update"
+    docker compose exec neon-test-extensions psql -d contrib_regression -c "\dx ${ext}"
+  done
+fi

docs/rfcs/041-sharded-ingest.md

@@ -0,0 +1,255 @@
+# 
+Created on Aug 2024
+Implemented on Jan 2025
+
+## Summary
+
+Data in large tenants is split up between multiple pageservers according to key hashes, as
+introduced in the [sharding RFC](031-sharding-static.md) and [shard splitting RFC](032-shard-splitting.md).
+
+Whereas currently we send all WAL to all pageserver shards, and each shard filters out the data that it needs,
+in this RFC we add a mechanism to filter the WAL on the safekeeper, so that each shard receives
+only the data it needs.
+
+This will place some extra CPU load on the safekeepers, in exchange for reducing the network bandwidth
+for ingesting WAL back to scaling as O(1) with shard count, rather than O(N_shards).
+
+## Motivation
+
+1. Large databases require higher shard counts.  Whereas currently we run with up to 8 shards for tenants
+with a few TB of storage, the next order of magnitude capacity increase will require tens of shards, such
+that sending all WAL to all shards is impractical in terms of bandwidth.
+2. For contemporary database sizes (~2TB), the pageserver is the bottleneck for ingest: since each
+   shard has to decode and process the whole WAL, sharding doesn't fully relieve this bottleneck.  To achieve significantly higher ingest speeds, we need to filter the WAL earlier so that each pageserver
+   only has to process relevant parts.
+
+## Non Goals (if relevant)
+
+We do not seek to introduce multiple WALs per timeline, or to share the work of handling a timeline's
+WAL across safekeepers (beyond simple 3x replication).  This RFC may be thought of as an incremental
+move of the ingestion bottleneck up the stack: instead of high write rates bottlenecking on the
+pageserver, they will bottleneck on the safekeeper.
+
+## Impacted components (e.g. pageserver, safekeeper, console, etc)
+
+Safekeeper, pageserver.
+
+There will be no control plane or storage controller coordination needed, as pageservers will directly
+indicate their sharding parameters to the safekeeper when subscribing for WAL.
+
+## Proposed implementation
+
+Terminology:
+- "Data pages" refers to postgres relation blocks, and SLRU blocks.
+- "Metadata pages" refers to everything else the pageserver stores, such as relation sizes and
+  directories of relations.
+
+### Phase 1: Refactor ingest
+
+Currently, pageserver ingest code is structured approximately as follows:
+1. `handle_walreceiver_connection` reads a stream of binary WAL records off a network
+   socket
+2. `WalIngest::ingest_record` to translate the record into a series of page-level modifications
+3. `DatadirModification` accumulates page updates from several `ingest_record` calls, and when
+   its `commit()` method is called, flushes these into a Timeline's open `InMemoryLayer`.
+
+This process currently assumes access to a pageserver `Timeline` throughout `ingest_record` and
+from `DatadirModification`, which is used to do read-modify-write cycles on metadata pages
+such as relation sizes and the master DBDIR page.  It also assumes that records are ingested
+strictly one after the other: they cannot be ingested in parallel because each record assumes
+that earlier records' changes have already been applied to `Timeline`.
+
+This code will be refactored to disentangle the simple, fast decode of relation page writes
+from the more complex logic for updating internal metadata.  An intermediate representation
+called `InterpretedWalRecords` will be introduced.  This is similar to the internal state of
+a `DatadirModification`, but does not require access to a Timeline.  Instead of storing
+metadata updates as materialized writes to pages, it will accumulate these as abstract operations,
+for example rather than including a write to a relation size key, this structure will include
+an operation that indicates "Update relation _foo_'s size to the max of its current value and
+_bar_", such that these may be applied later to a real Timeline.
+
+The `DatadirModification` will be aware of the `EphemeralFile` format, so that as it accumulates
+simple page writes of relation blocks, it can write them directly into a buffer in the serialized
+format.  This will avoid the need to later deserialize/reserialize this data when passing the
+structure between safekeeper and pageserver.
+
+The new pipeline will be:
+1. `handle_walreceiver_connection` reads a stream of binary WAL records off a network
+2. A `InterpretedWalRecords` is generated from the incoming WAL records.  This does not
+   require a reference to a Timeline.
+3. The logic that is current spread between `WalIngest` and `DatadirModification` for updating
+   metadata will be refactored to consume the metadata operations from the `InterpretedWalRecords`
+   and turn them into literal writes to metadata pages.  This part must be done sequentially.
+4. The resulting buffer of metadata page writes is combined with the buffer of relation block
+   writes, and written into the `InMemoryLayer`.
+
+Implemented in:
+1. https://github.com/neondatabase/neon/pull/9472
+2. https://github.com/neondatabase/neon/pull/9504
+3. https://github.com/neondatabase/neon/pull/9524
+
+### Phase 2: Decode & filter on safekeeper
+
+In the previous phase, the ingest code was modified to be able to do most of its work without access to
+a Timeline: this first stage of ingest simply converts a series of binary wal records into
+a buffer of relation/SLRU page writes, and a buffer of abstract metadata writes.
+
+The modified ingest code may be transplanted from pageserver to safekeeper (probably via a
+shared crate).  The safekeeper->pageserver network protocol is modified to:
+ - in subscription requests, send the `ShardIdentity` from the pageserver to the safekeeper
+ - in responses, transmit a `InterpretedWalRecords` instead of a raw `WalRecord`.
+ - use the `ShardIdentity` to filter the `ProcessedWalIngest` to relevant content for
+   the subscribing shard before transmitting it.
+
+The overall behavior of the pageserver->safekeeper interaction remains the same, in terms of
+consistent LSN feedback, and connection management.  Only the payload of the subscriptions
+changes, to express an LSN range of WAL as a filtered `ProcessedWalIngest` instead of the
+raw data.
+
+The ingest code on the pageserver can now skip the part where it does the first phase of
+processing, as it will receive pre-processed, compressed data off the wire.
+
+Note that `InterpretedWalRecord` batches multiple `InterpretedWalRecord(s)` in the same network
+message. Safekeeper reads WAL in chunks of 16 blocks and then decodes as many Postgres WAL records
+as possible. Each Postgres WAL record maps to one `InterpretedWalRecord` for potentially multiple shards.
+Hence, the size of the batch is given by the number of Postgres WAL records that fit in 16 blocks.
+
+The protocol needs to support evolution. Protobuf was chosen here with the view that, in the future,
+we may migrate it to GRPC altogether
+
+Implemented in:
+1. https://github.com/neondatabase/neon/pull/9746
+2. https://github.com/neondatabase/neon/pull/9821
+
+### Phase 3: Fan out interpreted WAL
+
+In the previous phase, the initial processing of WAL was moved to the safekeeper, but it is still
+done once for each shard: this will generate O(N_shards) CPU work on the safekeeper (especially
+when considering converting to Protobuf format and compression).
+
+To avoid this, we fan-out WAL from one (tenant, timeline, shard) to all other shards subscribed on
+the same safekeeper. Under normal operation, the WAL will be read from disk, decoded and interpreted
+_only_ once per (safekeeper, timeline).
+
+When the first shard of a sharded timeline subscribes to a given safekeeper a task is spawned
+for the WAL reader (`InterpretedWalReader`). This task reads WAL, decodes, interprets it and sends
+it to the sender (`InterpretedWalSender`). The sender is a future that is polled from the connection
+task. When further shards subscribe on the safekeeper they will attach themselves to the existing WAL reader.
+There's two cases to consider:
+1. The shard's requested `start_lsn` is ahead of the current position of the WAL reader. In this case, the shard
+will start receiving data when the reader reaches that LSN. The intuition here is that there's little to gain
+by letting shards "front-run" since compute backpressure is based on the laggard LSN.
+2. The shard's requested `start_lsn` is below the current position of the WAL reader. In this case, the WAL reader
+gets reset to this requested position (same intuition). Special care is taken such that advanced shards do not receive
+interpreted WAL records below their current position.
+
+The approach above implies that there is at most one WAL reader per (tenant, timeline) on a given safekeeper at any point in time.
+If this turns out to be operationally problematic, there's a trick we can deploy: `--max-delta-for-fanout` is an optional safekeeper
+argument that controls the max absolute delta between a new shard and the current WAL position of the WAL reader. If the absolute
+delta is above that value, a new reader is spawned. Note that there's currently no concurrency control on the number of WAL readers,
+so it's recommended to use large values to avoid pushing CPU utilisation too high.
+
+Unsharded tenants do not spawn a separate task for the interpreted WAL reader since there's no benefit to it. Instead they poll
+the reader and sender concurrently from the connection task.
+
+Shard splits are interesting here because it is the only case when the same shard might have two subscriptions at the same time.
+This is handled by giving readers a unique identifier. Both shards will receive the same data while respecting their requested start
+position.
+
+Implemented in:
+1. https://github.com/neondatabase/neon/pull/10190
+
+## Deployment
+
+Each phase shall be deployed independently. Special care should be taken around protocol changes.
+
+## Observability Tips
+
+* The safekeeper logs the protocol requested by the pageserver
+along with the pageserver ID, tenant, timeline and shard: `starting streaming from`.
+* There's metrics for the number of wal readers:
+  * `safekeeper_wal_readers{kind="task", target=~"pageserver.*"}` gives the number of wal reader tasks for each SK
+  * `safekeeper_wal_readers{kind="future", target=~"pageserver.*"}` gives the numer of wal readers polled inline by each SK
+  * `safekeeper_interpreted_wal_reader_tasks` gives the number of wal reader tasks per tenant, timeline
+* Interesting log lines for the fan-out reader:
+  * `Spawning interpreted`: first shard creates the interpreted wal reader
+  * `Fanning out`: a subsequent shard attaches itself to an interpreted wal reader
+  * `Aborting interpreted`: all senders have finished and the reader task is being aborted
+
+## Future Optimizations
+
+This sections describes some improvement areas which may be revisited in the future.
+
+### Buffering of Interpreted WAL
+
+The interpreted WAL reader may buffer interpreted WAL records in user space to help with serving
+subscriptions that are lagging behind the current position of the reader.
+
+Counterpoints:
+* Safekeepers serve many thousands of timelines and allocating a buffer for each might be wasteful,
+especially given that it would go unused on the happy path.
+* WAL is buffered in the kernel page cache. Usually we'd only pay the CPU cost of decoding and interpreting.
+
+### Tweaking the Pagserver Safekeeper Selection Algorithm
+
+We could make the pageserver aware of which safekeeper's already host shards for the timeline along
+with their current WAL positions. The pageserver should then prefer safkeepers that are in the same
+AZ _and_ already have a shard with a position close to the desired start position.
+
+We currently run one safekeeper per AZ, so the point is mute until that changes.
+
+### Pipelining first ingest phase
+
+The first ingest phase is a stateless transformation of a binary WAL record into a pre-processed
+output per shard.  To put multiple CPUs to work, we may pipeline this processing up to some defined buffer
+depth.
+
+## Alternatives considered
+
+### Give safekeepers enough state to fully decode WAL
+
+In this RFC, we only do the first phase of ingest on the safekeeper, because this is
+the phase that is stateless.  Subsequent changes then happen on the pageserver, with
+access to the `Timeline` state.
+
+We could do more work on the safekeeper if we transmitted metadata state to the safekeeper
+when subscribing to the WAL: for example, by telling the safekeeper all the relation sizes,
+so that it could then generate all the metadata writes for relation sizes.
+
+We avoid doing this for several reasons:
+1. Complexity: it's a more invasive protocol change
+2. Decoupling: having the safekeeper understand the `ProcessedWalIngest` already somewhat
+   infects it with knowledge of the pageserver, but this is mainly an abstract structure
+   that describes postgres writes.  However, if we taught the safekeeper about the exact
+   way that pageserver deals with metadata keys, this would be a much tighter coupling.
+3. Load: once the WAL has been processed to the point that it can be split between shards,
+   it is preferable to share out work on the remaining shards rather than adding extra CPU
+   load to the safekeeper.
+
+### Do pre-processing on the compute instead of the safekeeper
+
+Since our first stage of ingest is stateless, it could be done at any stage in the pipeline,
+all the way up to the compute.
+
+We choose not to do this, because it is useful for the safekeeper to store the raw WAL rather
+than just the preprocessed WAL:
+- The safekeeper still needs to be able to serve raw WAL back to postgres for e.g. physical replication
+- It simplifies our paxos implementation to have the offset in the write log be literally
+  the same as the LSN
+- Raw WAL must have a stable protocol since we might have to re-ingest it at arbitrary points in the future.
+  Storing raw WAL give us more flexibility to evolve the pageserver, safekeeper protocol.
+
+### Do wal pre-processing on shard 0 or a separate service, send it to other shards from there
+
+If we wanted to keep the safekeepers as entirely pure stores of raw WAL bytes, then
+we could do the initial decode and shard-splitting in some other location:
+- Shard 0 could subscribe to the full WAL and then send writes to other shards
+- A new intermediate service between the safekeeper and pageserver could do the splitting.
+
+So why not?
+- Extra network hop from shard 0 to the final destination shard
+- Clearly there is more infrastructure involved here compared with doing it inline on the safekeeper.
+- Safekeepers already have very light CPU load: typical cloud instances shapes with appropriate
+  disks for the safekeepers effectively have "free" CPU resources.
+- Doing extra work on shard 0 would complicate scheduling of shards on pageservers, because
+  shard 0 would have significantly higher CPU load under write workloads than other shards.

libs/compute_api/src/responses.rs

@@ -15,11 +15,6 @@ pub struct GenericAPIError {
     pub error: String,
 }
 
-#[derive(Debug, Clone, Serialize)]
-pub struct InfoResponse {
-    pub num_cpus: usize,
-}
-
 #[derive(Debug, Clone, Serialize)]
 pub struct ExtensionInstallResponse {
     pub extension: PgIdent,

libs/compute_api/src/spec.rs

@@ -138,6 +138,13 @@ pub struct ComputeSpec {
     /// enough spare connections for reconfiguration process to succeed.
     #[serde(default = "default_reconfigure_concurrency")]
     pub reconfigure_concurrency: usize,
+
+    /// If set to true, the compute_ctl will drop all subscriptions before starting the
+    /// compute. This is needed when we start an endpoint on a branch, so that child
+    /// would not compete with parent branch subscriptions
+    /// over the same replication content from publisher.
+    #[serde(default)] // Default false
+    pub drop_subscriptions_before_start: bool,
 }
 
 /// Feature flag to signal `compute_ctl` to enable certain experimental functionality.
@@ -197,14 +204,16 @@ impl RemoteExtSpec {
 
         // Check if extension is present in public or custom.
         // If not, then it is not allowed to be used by this compute.
-        if let Some(public_extensions) = &self.public_extensions {
-            if !public_extensions.contains(&real_ext_name.to_string()) {
-                if let Some(custom_extensions) = &self.custom_extensions {
-                    if !custom_extensions.contains(&real_ext_name.to_string()) {
-                        return Err(anyhow::anyhow!("extension {} is not found", real_ext_name));
-                    }
-                }
-            }
+        if !self
+            .public_extensions
+            .as_ref()
+            .is_some_and(|exts| exts.iter().any(|e| e == ext_name))
+            && !self
+                .custom_extensions
+                .as_ref()
+                .is_some_and(|exts| exts.iter().any(|e| e == ext_name))
+        {
+            return Err(anyhow::anyhow!("extension {} is not found", real_ext_name));
         }
 
         match self.extension_data.get(real_ext_name) {
@@ -333,6 +342,96 @@ mod tests {
     use super::*;
     use std::fs::File;
 
+    #[test]
+    fn allow_installing_remote_extensions() {
+        let rspec: RemoteExtSpec = serde_json::from_value(serde_json::json!({
+            "public_extensions": null,
+            "custom_extensions": null,
+            "library_index": {},
+            "extension_data": {},
+        }))
+        .unwrap();
+
+        rspec
+            .get_ext("ext", false, "latest", "v17")
+            .expect_err("Extension should not be found");
+
+        let rspec: RemoteExtSpec = serde_json::from_value(serde_json::json!({
+            "public_extensions": [],
+            "custom_extensions": null,
+            "library_index": {},
+            "extension_data": {},
+        }))
+        .unwrap();
+
+        rspec
+            .get_ext("ext", false, "latest", "v17")
+            .expect_err("Extension should not be found");
+
+        let rspec: RemoteExtSpec = serde_json::from_value(serde_json::json!({
+            "public_extensions": [],
+            "custom_extensions": [],
+            "library_index": {
+                "ext": "ext"
+            },
+            "extension_data": {
+                "ext": {
+                    "control_data": {
+                        "ext.control": ""
+                    },
+                    "archive_path": ""
+                }
+            },
+        }))
+        .unwrap();
+
+        rspec
+            .get_ext("ext", false, "latest", "v17")
+            .expect_err("Extension should not be found");
+
+        let rspec: RemoteExtSpec = serde_json::from_value(serde_json::json!({
+            "public_extensions": [],
+            "custom_extensions": ["ext"],
+            "library_index": {
+                "ext": "ext"
+            },
+            "extension_data": {
+                "ext": {
+                    "control_data": {
+                        "ext.control": ""
+                    },
+                    "archive_path": ""
+                }
+            },
+        }))
+        .unwrap();
+
+        rspec
+            .get_ext("ext", false, "latest", "v17")
+            .expect("Extension should be found");
+
+        let rspec: RemoteExtSpec = serde_json::from_value(serde_json::json!({
+            "public_extensions": ["ext"],
+            "custom_extensions": [],
+            "library_index": {
+                "ext": "ext"
+            },
+            "extension_data": {
+                "ext": {
+                    "control_data": {
+                        "ext.control": ""
+                    },
+                    "archive_path": ""
+                }
+            },
+        }))
+        .unwrap();
+
+        rspec
+            .get_ext("ext", false, "latest", "v17")
+            .expect("Extension should be found");
+    }
+
     #[test]
     fn parse_spec_file() {
         let file = File::open("tests/cluster_spec.json").unwrap();

libs/pageserver_api/src/config.rs

@@ -120,6 +120,7 @@ pub struct ConfigToml {
     pub no_sync: Option<bool>,
     pub wal_receiver_protocol: PostgresClientProtocol,
     pub page_service_pipelining: PageServicePipeliningConfig,
+    pub get_vectored_concurrent_io: GetVectoredConcurrentIo,
 }
 
 #[derive(Debug, Clone, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
@@ -158,6 +159,25 @@ pub enum PageServiceProtocolPipelinedExecutionStrategy {
     Tasks,
 }
 
+#[derive(Debug, Clone, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
+#[serde(tag = "mode", rename_all = "kebab-case")]
+#[serde(deny_unknown_fields)]
+pub enum GetVectoredConcurrentIo {
+    /// The read path is fully sequential: layers are visited
+    /// one after the other and IOs are issued and waited upon
+    /// from the same task that traverses the layers.
+    Sequential,
+    /// The read path still traverses layers sequentially, and
+    /// index blocks will be read into the PS PageCache from
+    /// that task, with waiting.
+    /// But data IOs are dispatched and waited upon from a sidecar
+    /// task so that the traversing task can continue to traverse
+    /// layers while the IOs are in flight.
+    /// If the PS PageCache miss rate is low, this improves
+    /// throughput dramatically.
+    SidecarTask,
+}
+
 pub mod statvfs {
     pub mod mock {
         #[derive(Debug, Clone, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
@@ -234,9 +254,26 @@ pub struct TenantConfigToml {
     // Duration::ZERO means automatic compaction is disabled.
     #[serde(with = "humantime_serde")]
     pub compaction_period: Duration,
-    // Level0 delta layer threshold for compaction.
+    /// Level0 delta layer threshold for compaction.
     pub compaction_threshold: usize,
+    /// Controls the amount of L0 included in a single compaction iteration.
+    /// The unit is `checkpoint_distance`, i.e., a size.
+    /// We add L0s to the set of layers to compact until their cumulative
+    /// size exceeds `compaction_upper_limit * checkpoint_distance`.
+    pub compaction_upper_limit: usize,
     pub compaction_algorithm: crate::models::CompactionAlgorithmSettings,
+    /// Level0 delta layer threshold at which to delay layer flushes for compaction backpressure,
+    /// such that they take 2x as long, and start waiting for layer flushes during ephemeral layer
+    /// rolls. This helps compaction keep up with WAL ingestion, and avoids read amplification
+    /// blowing up. Should be >compaction_threshold. 0 to disable. Disabled by default.
+    pub l0_flush_delay_threshold: Option<usize>,
+    /// Level0 delta layer threshold at which to stall layer flushes. Must be >compaction_threshold
+    /// to avoid deadlock. 0 to disable. Disabled by default.
+    pub l0_flush_stall_threshold: Option<usize>,
+    /// If true, Level0 delta layer flushes will wait for S3 upload before flushing the next
+    /// layer. This is a temporary backpressure mechanism which should be removed once
+    /// l0_flush_{delay,stall}_threshold is fully enabled.
+    pub l0_flush_wait_upload: bool,
     // Determines how much history is retained, to allow
     // branching and read replicas at an older point in time.
     // The unit is #of bytes of WAL.
@@ -286,6 +323,10 @@ pub struct TenantConfigToml {
     // Expresed in multiples of checkpoint distance.
     pub image_layer_creation_check_threshold: u8,
 
+    // How many multiples of L0 `compaction_threshold` will preempt image layer creation and do L0 compaction.
+    // Set to 0 to disable preemption.
+    pub image_creation_preempt_threshold: usize,
+
     /// The length for an explicit LSN lease request.
     /// Layers needed to reconstruct pages at LSN will not be GC-ed during this interval.
     #[serde(with = "humantime_serde")]
@@ -301,6 +342,20 @@ pub struct TenantConfigToml {
     pub timeline_offloading: bool,
 
     pub wal_receiver_protocol_override: Option<PostgresClientProtocol>,
+
+    /// Enable rel_size_v2 for this tenant. Once enabled, the tenant will persist this information into
+    /// `index_part.json`, and it cannot be reversed.
+    pub rel_size_v2_enabled: Option<bool>,
+
+    // gc-compaction related configs
+    /// Enable automatic gc-compaction trigger on this tenant.
+    pub gc_compaction_enabled: bool,
+    /// The initial threshold for gc-compaction in KB. Once the total size of layers below the gc-horizon is above this threshold,
+    /// gc-compaction will be triggered.
+    pub gc_compaction_initial_threshold_kb: u64,
+    /// The ratio that triggers the auto gc-compaction. If (the total size of layers between L2 LSN and gc-horizon) / (size below the L2 LSN)
+    /// is above this ratio, gc-compaction will be triggered.
+    pub gc_compaction_ratio_percent: u64,
 }
 
 pub mod defaults {
@@ -450,6 +505,11 @@ impl Default for ConfigToml {
                     execution: PageServiceProtocolPipelinedExecutionStrategy::ConcurrentFutures,
                 })
             },
+            get_vectored_concurrent_io: if !cfg!(test) {
+                GetVectoredConcurrentIo::Sequential
+            } else {
+                GetVectoredConcurrentIo::SidecarTask
+            },
         }
     }
 }
@@ -472,9 +532,17 @@ pub mod tenant_conf_defaults {
 
     pub const DEFAULT_COMPACTION_PERIOD: &str = "20 s";
     pub const DEFAULT_COMPACTION_THRESHOLD: usize = 10;
+
+    // This value needs to be tuned to avoid OOM. We have 3/4 of the total CPU threads to do background works, that's 16*3/4=9 on
+    // most of our pageservers. Compaction ~50 layers requires about 2GB memory (could be reduced later by optimizing L0 hole
+    // calculation to avoid loading all keys into the memory). So with this config, we can get a maximum peak compaction usage of 18GB.
+    pub const DEFAULT_COMPACTION_UPPER_LIMIT: usize = 50;
+
     pub const DEFAULT_COMPACTION_ALGORITHM: crate::models::CompactionAlgorithm =
         crate::models::CompactionAlgorithm::Legacy;
 
+    pub const DEFAULT_L0_FLUSH_WAIT_UPLOAD: bool = true;
+
     pub const DEFAULT_GC_HORIZON: u64 = 64 * 1024 * 1024;
 
     // Large DEFAULT_GC_PERIOD is fine as long as PITR_INTERVAL is larger.
@@ -483,6 +551,10 @@ pub mod tenant_conf_defaults {
     // Relevant: https://github.com/neondatabase/neon/issues/3394
     pub const DEFAULT_GC_PERIOD: &str = "1 hr";
     pub const DEFAULT_IMAGE_CREATION_THRESHOLD: usize = 3;
+    // If there are more than threshold * compaction_threshold (that is 3 * 10 in the default config) L0 layers, image
+    // layer creation will end immediately. Set to 0 to disable. The target default will be 3 once we
+    // want to enable this feature.
+    pub const DEFAULT_IMAGE_CREATION_PREEMPT_THRESHOLD: usize = 0;
     pub const DEFAULT_PITR_INTERVAL: &str = "7 days";
     pub const DEFAULT_WALRECEIVER_CONNECT_TIMEOUT: &str = "10 seconds";
     pub const DEFAULT_WALRECEIVER_LAGGING_WAL_TIMEOUT: &str = "10 seconds";
@@ -494,6 +566,9 @@ pub mod tenant_conf_defaults {
     // By default ingest enough WAL for two new L0 layers before checking if new image
     // image layers should be created.
     pub const DEFAULT_IMAGE_LAYER_CREATION_CHECK_THRESHOLD: u8 = 2;
+    pub const DEFAULT_GC_COMPACTION_ENABLED: bool = false;
+    pub const DEFAULT_GC_COMPACTION_INITIAL_THRESHOLD_KB: u64 = 10240000;
+    pub const DEFAULT_GC_COMPACTION_RATIO_PERCENT: u64 = 100;
 }
 
 impl Default for TenantConfigToml {
@@ -507,9 +582,13 @@ impl Default for TenantConfigToml {
             compaction_period: humantime::parse_duration(DEFAULT_COMPACTION_PERIOD)
                 .expect("cannot parse default compaction period"),
             compaction_threshold: DEFAULT_COMPACTION_THRESHOLD,
+            compaction_upper_limit: DEFAULT_COMPACTION_UPPER_LIMIT,
             compaction_algorithm: crate::models::CompactionAlgorithmSettings {
                 kind: DEFAULT_COMPACTION_ALGORITHM,
             },
+            l0_flush_delay_threshold: None,
+            l0_flush_stall_threshold: None,
+            l0_flush_wait_upload: DEFAULT_L0_FLUSH_WAIT_UPLOAD,
             gc_horizon: DEFAULT_GC_HORIZON,
             gc_period: humantime::parse_duration(DEFAULT_GC_PERIOD)
                 .expect("cannot parse default gc period"),
@@ -534,10 +613,15 @@ impl Default for TenantConfigToml {
             lazy_slru_download: false,
             timeline_get_throttle: crate::models::ThrottleConfig::disabled(),
             image_layer_creation_check_threshold: DEFAULT_IMAGE_LAYER_CREATION_CHECK_THRESHOLD,
+            image_creation_preempt_threshold: DEFAULT_IMAGE_CREATION_PREEMPT_THRESHOLD,
             lsn_lease_length: LsnLease::DEFAULT_LENGTH,
             lsn_lease_length_for_ts: LsnLease::DEFAULT_LENGTH_FOR_TS,
             timeline_offloading: false,
             wal_receiver_protocol_override: None,
+            rel_size_v2_enabled: None,
+            gc_compaction_enabled: DEFAULT_GC_COMPACTION_ENABLED,
+            gc_compaction_initial_threshold_kb: DEFAULT_GC_COMPACTION_INITIAL_THRESHOLD_KB,
+            gc_compaction_ratio_percent: DEFAULT_GC_COMPACTION_RATIO_PERCENT,
         }
     }
 }

libs/pageserver_api/src/models.rs

@@ -458,10 +458,18 @@ pub struct TenantConfigPatch {
     pub compaction_period: FieldPatch<String>,
     #[serde(skip_serializing_if = "FieldPatch::is_noop")]
     pub compaction_threshold: FieldPatch<usize>,
+    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
+    pub compaction_upper_limit: FieldPatch<usize>,
     // defer parsing compaction_algorithm, like eviction_policy
     #[serde(skip_serializing_if = "FieldPatch::is_noop")]
     pub compaction_algorithm: FieldPatch<CompactionAlgorithmSettings>,
     #[serde(skip_serializing_if = "FieldPatch::is_noop")]
+    pub l0_flush_delay_threshold: FieldPatch<usize>,
+    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
+    pub l0_flush_stall_threshold: FieldPatch<usize>,
+    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
+    pub l0_flush_wait_upload: FieldPatch<bool>,
+    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
     pub gc_horizon: FieldPatch<u64>,
     #[serde(skip_serializing_if = "FieldPatch::is_noop")]
     pub gc_period: FieldPatch<String>,
@@ -490,6 +498,8 @@ pub struct TenantConfigPatch {
     #[serde(skip_serializing_if = "FieldPatch::is_noop")]
     pub image_layer_creation_check_threshold: FieldPatch<u8>,
     #[serde(skip_serializing_if = "FieldPatch::is_noop")]
+    pub image_creation_preempt_threshold: FieldPatch<usize>,
+    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
     pub lsn_lease_length: FieldPatch<String>,
     #[serde(skip_serializing_if = "FieldPatch::is_noop")]
     pub lsn_lease_length_for_ts: FieldPatch<String>,
@@ -497,6 +507,14 @@ pub struct TenantConfigPatch {
     pub timeline_offloading: FieldPatch<bool>,
     #[serde(skip_serializing_if = "FieldPatch::is_noop")]
     pub wal_receiver_protocol_override: FieldPatch<PostgresClientProtocol>,
+    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
+    pub rel_size_v2_enabled: FieldPatch<bool>,
+    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
+    pub gc_compaction_enabled: FieldPatch<bool>,
+    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
+    pub gc_compaction_initial_threshold_kb: FieldPatch<u64>,
+    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
+    pub gc_compaction_ratio_percent: FieldPatch<u64>,
 }
 
 /// An alternative representation of `pageserver::tenant::TenantConf` with
@@ -508,8 +526,12 @@ pub struct TenantConfig {
     pub compaction_target_size: Option<u64>,
     pub compaction_period: Option<String>,
     pub compaction_threshold: Option<usize>,
+    pub compaction_upper_limit: Option<usize>,
     // defer parsing compaction_algorithm, like eviction_policy
     pub compaction_algorithm: Option<CompactionAlgorithmSettings>,
+    pub l0_flush_delay_threshold: Option<usize>,
+    pub l0_flush_stall_threshold: Option<usize>,
+    pub l0_flush_wait_upload: Option<bool>,
     pub gc_horizon: Option<u64>,
     pub gc_period: Option<String>,
     pub image_creation_threshold: Option<usize>,
@@ -524,10 +546,15 @@ pub struct TenantConfig {
     pub lazy_slru_download: Option<bool>,
     pub timeline_get_throttle: Option<ThrottleConfig>,
     pub image_layer_creation_check_threshold: Option<u8>,
+    pub image_creation_preempt_threshold: Option<usize>,
     pub lsn_lease_length: Option<String>,
     pub lsn_lease_length_for_ts: Option<String>,
     pub timeline_offloading: Option<bool>,
     pub wal_receiver_protocol_override: Option<PostgresClientProtocol>,
+    pub rel_size_v2_enabled: Option<bool>,
+    pub gc_compaction_enabled: Option<bool>,
+    pub gc_compaction_initial_threshold_kb: Option<u64>,
+    pub gc_compaction_ratio_percent: Option<u64>,
 }
 
 impl TenantConfig {
@@ -538,7 +565,11 @@ impl TenantConfig {
             mut compaction_target_size,
             mut compaction_period,
             mut compaction_threshold,
+            mut compaction_upper_limit,
             mut compaction_algorithm,
+            mut l0_flush_delay_threshold,
+            mut l0_flush_stall_threshold,
+            mut l0_flush_wait_upload,
             mut gc_horizon,
             mut gc_period,
             mut image_creation_threshold,
@@ -553,10 +584,15 @@ impl TenantConfig {
             mut lazy_slru_download,
             mut timeline_get_throttle,
             mut image_layer_creation_check_threshold,
+            mut image_creation_preempt_threshold,
             mut lsn_lease_length,
             mut lsn_lease_length_for_ts,
             mut timeline_offloading,
             mut wal_receiver_protocol_override,
+            mut rel_size_v2_enabled,
+            mut gc_compaction_enabled,
+            mut gc_compaction_initial_threshold_kb,
+            mut gc_compaction_ratio_percent,
         } = self;
 
         patch.checkpoint_distance.apply(&mut checkpoint_distance);
@@ -566,7 +602,17 @@ impl TenantConfig {
             .apply(&mut compaction_target_size);
         patch.compaction_period.apply(&mut compaction_period);
         patch.compaction_threshold.apply(&mut compaction_threshold);
+        patch
+            .compaction_upper_limit
+            .apply(&mut compaction_upper_limit);
         patch.compaction_algorithm.apply(&mut compaction_algorithm);
+        patch
+            .l0_flush_delay_threshold
+            .apply(&mut l0_flush_delay_threshold);
+        patch
+            .l0_flush_stall_threshold
+            .apply(&mut l0_flush_stall_threshold);
+        patch.l0_flush_wait_upload.apply(&mut l0_flush_wait_upload);
         patch.gc_horizon.apply(&mut gc_horizon);
         patch.gc_period.apply(&mut gc_period);
         patch
@@ -593,6 +639,9 @@ impl TenantConfig {
         patch
             .image_layer_creation_check_threshold
             .apply(&mut image_layer_creation_check_threshold);
+        patch
+            .image_creation_preempt_threshold
+            .apply(&mut image_creation_preempt_threshold);
         patch.lsn_lease_length.apply(&mut lsn_lease_length);
         patch
             .lsn_lease_length_for_ts
@@ -601,6 +650,16 @@ impl TenantConfig {
         patch
             .wal_receiver_protocol_override
             .apply(&mut wal_receiver_protocol_override);
+        patch.rel_size_v2_enabled.apply(&mut rel_size_v2_enabled);
+        patch
+            .gc_compaction_enabled
+            .apply(&mut gc_compaction_enabled);
+        patch
+            .gc_compaction_initial_threshold_kb
+            .apply(&mut gc_compaction_initial_threshold_kb);
+        patch
+            .gc_compaction_ratio_percent
+            .apply(&mut gc_compaction_ratio_percent);
 
         Self {
             checkpoint_distance,
@@ -608,7 +667,11 @@ impl TenantConfig {
             compaction_target_size,
             compaction_period,
             compaction_threshold,
+            compaction_upper_limit,
             compaction_algorithm,
+            l0_flush_delay_threshold,
+            l0_flush_stall_threshold,
+            l0_flush_wait_upload,
             gc_horizon,
             gc_period,
             image_creation_threshold,
@@ -623,10 +686,15 @@ impl TenantConfig {
             lazy_slru_download,
             timeline_get_throttle,
             image_layer_creation_check_threshold,
+            image_creation_preempt_threshold,
             lsn_lease_length,
             lsn_lease_length_for_ts,
             timeline_offloading,
             wal_receiver_protocol_override,
+            rel_size_v2_enabled,
+            gc_compaction_enabled,
+            gc_compaction_initial_threshold_kb,
+            gc_compaction_ratio_percent,
         }
     }
 }
@@ -969,6 +1037,13 @@ pub struct TenantConfigPatchRequest {
     pub config: TenantConfigPatch, // as we have a flattened field, we should reject all unknown fields in it
 }
 
+#[derive(Serialize, Deserialize, Debug)]
+pub struct TenantWaitLsnRequest {
+    #[serde(flatten)]
+    pub timelines: HashMap<TimelineId, Lsn>,
+    pub timeout: Duration,
+}
+
 /// See [`TenantState::attachment_status`] and the OpenAPI docs for context.
 #[derive(Serialize, Deserialize, Clone)]
 #[serde(tag = "slug", content = "data", rename_all = "snake_case")]

libs/pq_proto/src/lib.rs

@@ -182,6 +182,13 @@ pub struct CancelKeyData {
     pub cancel_key: i32,
 }
 
+pub fn id_to_cancel_key(id: u64) -> CancelKeyData {
+    CancelKeyData {
+        backend_pid: (id >> 32) as i32,
+        cancel_key: (id & 0xffffffff) as i32,
+    }
+}
+
 impl fmt::Display for CancelKeyData {
     fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
         let hi = (self.backend_pid as u64) << 32;

libs/proxy/tokio-postgres2/Cargo.toml

@@ -19,3 +19,4 @@ postgres-protocol2 = { path = "../postgres-protocol2" }
 postgres-types2 = { path = "../postgres-types2" }
 tokio = { workspace = true, features = ["io-util", "time", "net"] }
 tokio-util = { workspace = true, features = ["codec"] }
+serde = { workspace = true, features = ["derive"] }

libs/proxy/tokio-postgres2/src/cancel_token.rs

@@ -3,12 +3,13 @@ use crate::tls::TlsConnect;
 
 use crate::{cancel_query, client::SocketConfig, tls::MakeTlsConnect};
 use crate::{cancel_query_raw, Error};
+use serde::{Deserialize, Serialize};
 use tokio::io::{AsyncRead, AsyncWrite};
 use tokio::net::TcpStream;
 
 /// The capability to request cancellation of in-progress queries on a
 /// connection.
-#[derive(Clone)]
+#[derive(Clone, Serialize, Deserialize)]
 pub struct CancelToken {
     pub socket_config: Option<SocketConfig>,
     pub ssl_mode: SslMode,

libs/proxy/tokio-postgres2/src/client.rs

@@ -18,6 +18,7 @@ use fallible_iterator::FallibleIterator;
 use futures_util::{future, ready, TryStreamExt};
 use parking_lot::Mutex;
 use postgres_protocol2::message::{backend::Message, frontend};
+use serde::{Deserialize, Serialize};
 use std::collections::HashMap;
 use std::fmt;
 use std::sync::Arc;
@@ -137,7 +138,7 @@ impl InnerClient {
     }
 }
 
-#[derive(Clone)]
+#[derive(Clone, Serialize, Deserialize)]
 pub struct SocketConfig {
     pub host: Host,
     pub port: u16,

libs/proxy/tokio-postgres2/src/config.rs

@@ -7,6 +7,7 @@ use crate::tls::MakeTlsConnect;
 use crate::tls::TlsConnect;
 use crate::{Client, Connection, Error};
 use postgres_protocol2::message::frontend::StartupMessageParams;
+use serde::{Deserialize, Serialize};
 use std::fmt;
 use std::str;
 use std::time::Duration;
@@ -16,7 +17,7 @@ pub use postgres_protocol2::authentication::sasl::ScramKeys;
 use tokio::net::TcpStream;
 
 /// TLS configuration.
-#[derive(Debug, Copy, Clone, PartialEq, Eq)]
+#[derive(Debug, Copy, Clone, PartialEq, Eq, Serialize, Deserialize)]
 #[non_exhaustive]
 pub enum SslMode {
     /// Do not use TLS.
@@ -50,7 +51,7 @@ pub enum ReplicationMode {
 }
 
 /// A host specification.
-#[derive(Debug, Clone, PartialEq, Eq)]
+#[derive(Debug, Clone, PartialEq, Eq, Serialize, Deserialize)]
 pub enum Host {
     /// A TCP hostname.
     Tcp(String),