Merge pull request #8374 from neondatabase/rc/2024-07-15

Storage & Compute release 2024-07-15

.gitattributes

@@ -1,2 +0,0 @@
-# allows for nicer hunk headers with git show
-*.rs diff=rust

.github/actionlint.yml

@@ -8,9 +8,6 @@ self-hosted-runner:
     - small-arm64
     - us-east-2
 config-variables:
-  - BENCHMARK_PROJECT_ID_PUB
-  - BENCHMARK_PROJECT_ID_SUB
   - REMOTE_STORAGE_AZURE_CONTAINER
   - REMOTE_STORAGE_AZURE_REGION
   - SLACK_UPCOMING_RELEASE_CHANNEL_ID
-  - DEV_AWS_OIDC_ROLE_ARN

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

@@ -9,13 +9,16 @@ inputs:
     description: 'Region ID, if not set the project will be created in the default region'
     default: aws-us-east-2
   postgres_version:
-    description: 'Postgres version; default is 16'
-    default: '16'
+    description: 'Postgres version; default is 15'
+    default: '15'
   api_host:
     description: 'Neon API host'
     default: console-stage.neon.build
+  provisioner:
+    description: 'k8s-pod or k8s-neonvm'
+    default: 'k8s-pod'
   compute_units:
-    description: '[Min, Max] compute units'
+    description: '[Min, Max] compute units; Min and Max are used for k8s-neonvm with autoscaling, for k8s-pod values Min and Max should be equal'
     default: '[1, 1]'
 
 outputs:
@@ -34,6 +37,10 @@ runs:
       # A shell without `set -x` to not to expose password/dsn in logs
       shell: bash -euo pipefail {0}
       run: |
+        if [ "${PROVISIONER}" == "k8s-pod" ] && [ "${MIN_CU}" != "${MAX_CU}" ]; then
+          echo >&2 "For k8s-pod provisioner MIN_CU should be equal to MAX_CU"
+        fi
+
         project=$(curl \
           "https://${API_HOST}/api/v2/projects" \
           --fail \
@@ -45,7 +52,7 @@ runs:
               \"name\": \"Created by actions/neon-project-create; GITHUB_RUN_ID=${GITHUB_RUN_ID}\",
               \"pg_version\": ${POSTGRES_VERSION},
               \"region_id\": \"${REGION_ID}\",
-              \"provisioner\": \"k8s-neonvm\",
+              \"provisioner\": \"${PROVISIONER}\",
               \"autoscaling_limit_min_cu\": ${MIN_CU},
               \"autoscaling_limit_max_cu\": ${MAX_CU},
               \"settings\": { }
@@ -68,5 +75,6 @@ runs:
         API_KEY: ${{ inputs.api_key }}
         REGION_ID: ${{ inputs.region_id }}
         POSTGRES_VERSION: ${{ inputs.postgres_version }}
+        PROVISIONER: ${{ inputs.provisioner }}
         MIN_CU: ${{ fromJSON(inputs.compute_units)[0] }}
         MAX_CU: ${{ fromJSON(inputs.compute_units)[1] }}

.github/actions/run-python-test-set/action.yml

@@ -131,8 +131,8 @@ runs:
           exit 1
         fi
         if [[ "${{ inputs.run_in_parallel }}" == "true" ]]; then
-          # -n sets the number of parallel processes that pytest-xdist will run
-          EXTRA_PARAMS="-n12 $EXTRA_PARAMS"
+          # -n16 uses sixteen processes to run tests via pytest-xdist
+          EXTRA_PARAMS="-n16 $EXTRA_PARAMS"
 
           # --dist=loadgroup points tests marked with @pytest.mark.xdist_group
           # to the same worker to make @pytest.mark.order work with xdist

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

@@ -1,292 +0,0 @@
-name: Build and Test Locally
-
-on:
-  workflow_call:
-    inputs:
-      arch:
-        description: 'x64 or arm64'
-        required: true
-        type: string
-      build-tag:
-        description: 'build tag'
-        required: true
-        type: string
-      build-tools-image:
-        description: 'build-tools image'
-        required: true
-        type: string
-      build-type:
-        description: 'debug or release'
-        required: true
-        type: string
-      pg-versions:
-        description: 'a json array of postgres versions to run regression tests on'
-        required: true
-        type: string
-
-defaults:
-  run:
-    shell: bash -euxo pipefail {0}
-
-env:
-  RUST_BACKTRACE: 1
-  COPT: '-Werror'
-  AWS_ACCESS_KEY_ID: ${{ secrets.AWS_ACCESS_KEY_DEV }}
-  AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_KEY_DEV }}
-
-jobs:
-  build-neon:
-    runs-on: ${{ fromJson(format('["self-hosted", "{0}"]', inputs.arch == 'arm64' && 'large-arm64' || 'large')) }}
-    container:
-      image: ${{ inputs.build-tools-image }}
-      credentials:
-        username: ${{ secrets.NEON_DOCKERHUB_USERNAME }}
-        password: ${{ secrets.NEON_DOCKERHUB_PASSWORD }}
-      # Raise locked memory limit for tokio-epoll-uring.
-      # On 5.10 LTS kernels < 5.10.162 (and generally mainline kernels < 5.12),
-      # io_uring will account the memory of the CQ and SQ as locked.
-      # More details: https://github.com/neondatabase/neon/issues/6373#issuecomment-1905814391
-      options: --init --shm-size=512mb --ulimit memlock=67108864:67108864
-    env:
-      BUILD_TYPE: ${{ inputs.build-type }}
-      GIT_VERSION: ${{ github.event.pull_request.head.sha || github.sha }}
-      BUILD_TAG: ${{ inputs.build-tag }}
-
-    steps:
-      - name: Fix git ownership
-        run: |
-          # Workaround for `fatal: detected dubious ownership in repository at ...`
-          #
-          # Use both ${{ github.workspace }} and ${GITHUB_WORKSPACE} because they're different on host and in containers
-          #   Ref https://github.com/actions/checkout/issues/785
-          #
-          git config --global --add safe.directory ${{ github.workspace }}
-          git config --global --add safe.directory ${GITHUB_WORKSPACE}
-          for r in 14 15 16; do
-            git config --global --add safe.directory "${{ github.workspace }}/vendor/postgres-v$r"
-            git config --global --add safe.directory "${GITHUB_WORKSPACE}/vendor/postgres-v$r"
-          done
-
-      - uses: actions/checkout@v4
-        with:
-          submodules: true
-          fetch-depth: 1
-
-      - name: Set pg 14 revision for caching
-        id: pg_v14_rev
-        run: echo pg_rev=$(git rev-parse HEAD:vendor/postgres-v14) >> $GITHUB_OUTPUT
-
-      - name: Set pg 15 revision for caching
-        id: pg_v15_rev
-        run: echo pg_rev=$(git rev-parse HEAD:vendor/postgres-v15) >> $GITHUB_OUTPUT
-
-      - name: Set pg 16 revision for caching
-        id: pg_v16_rev
-        run: echo pg_rev=$(git rev-parse HEAD:vendor/postgres-v16) >> $GITHUB_OUTPUT
-
-      # Set some environment variables used by all the steps.
-      #
-      # CARGO_FLAGS is extra options to pass to "cargo build", "cargo test" etc.
-      #   It also includes --features, if any
-      #
-      # CARGO_FEATURES is passed to "cargo metadata". It is separate from CARGO_FLAGS,
-      #   because "cargo metadata" doesn't accept --release or --debug options
-      #
-      # We run tests with addtional features, that are turned off by default (e.g. in release builds), see
-      # corresponding Cargo.toml files for their descriptions.
-      - name: Set env variables
-        run: |
-          CARGO_FEATURES="--features testing"
-          if [[ $BUILD_TYPE == "debug" ]]; then
-            cov_prefix="scripts/coverage --profraw-prefix=$GITHUB_JOB --dir=/tmp/coverage run"
-            CARGO_FLAGS="--locked"
-          elif [[ $BUILD_TYPE == "release" ]]; then
-            cov_prefix=""
-            CARGO_FLAGS="--locked --release"
-          fi
-          {
-            echo "cov_prefix=${cov_prefix}"
-            echo "CARGO_FEATURES=${CARGO_FEATURES}"
-            echo "CARGO_FLAGS=${CARGO_FLAGS}"
-            echo "CARGO_HOME=${GITHUB_WORKSPACE}/.cargo"
-          } >> $GITHUB_ENV
-
-      - name: Cache postgres v14 build
-        id: cache_pg_14
-        uses: actions/cache@v4
-        with:
-          path: pg_install/v14
-          key: v1-${{ runner.os }}-${{ runner.arch }}-${{ inputs.build-type }}-pg-${{ steps.pg_v14_rev.outputs.pg_rev }}-${{ hashFiles('Makefile', 'Dockerfile.build-tools') }}
-
-      - name: Cache postgres v15 build
-        id: cache_pg_15
-        uses: actions/cache@v4
-        with:
-          path: pg_install/v15
-          key: v1-${{ runner.os }}-${{ runner.arch }}-${{ inputs.build-type }}-pg-${{ steps.pg_v15_rev.outputs.pg_rev }}-${{ hashFiles('Makefile', 'Dockerfile.build-tools') }}
-
-      - name: Cache postgres v16 build
-        id: cache_pg_16
-        uses: actions/cache@v4
-        with:
-          path: pg_install/v16
-          key: v1-${{ runner.os }}-${{ runner.arch }}-${{ inputs.build-type }}-pg-${{ steps.pg_v16_rev.outputs.pg_rev }}-${{ hashFiles('Makefile', 'Dockerfile.build-tools') }}
-
-      - name: Build postgres v14
-        if: steps.cache_pg_14.outputs.cache-hit != 'true'
-        run: mold -run make postgres-v14 -j$(nproc)
-
-      - name: Build postgres v15
-        if: steps.cache_pg_15.outputs.cache-hit != 'true'
-        run: mold -run make postgres-v15 -j$(nproc)
-
-      - name: Build postgres v16
-        if: steps.cache_pg_16.outputs.cache-hit != 'true'
-        run: mold -run make postgres-v16 -j$(nproc)
-
-      - name: Build neon extensions
-        run: mold -run make neon-pg-ext -j$(nproc)
-
-      - name: Build walproposer-lib
-        run: mold -run make walproposer-lib -j$(nproc)
-
-      - name: Run cargo build
-        run: |
-          PQ_LIB_DIR=$(pwd)/pg_install/v16/lib
-          export PQ_LIB_DIR
-          ${cov_prefix} mold -run cargo build $CARGO_FLAGS $CARGO_FEATURES --bins --tests
-
-      # Do install *before* running rust tests because they might recompile the
-      # binaries with different features/flags.
-      - name: Install rust binaries
-        run: |
-          # Install target binaries
-          mkdir -p /tmp/neon/bin/
-          binaries=$(
-            ${cov_prefix} cargo metadata $CARGO_FEATURES --format-version=1 --no-deps |
-            jq -r '.packages[].targets[] | select(.kind | index("bin")) | .name'
-          )
-          for bin in $binaries; do
-            SRC=target/$BUILD_TYPE/$bin
-            DST=/tmp/neon/bin/$bin
-            cp "$SRC" "$DST"
-          done
-
-          # Install test executables and write list of all binaries (for code coverage)
-          if [[ $BUILD_TYPE == "debug" ]]; then
-            # Keep bloated coverage data files away from the rest of the artifact
-            mkdir -p /tmp/coverage/
-
-            mkdir -p /tmp/neon/test_bin/
-
-            test_exe_paths=$(
-              ${cov_prefix} cargo test $CARGO_FLAGS $CARGO_FEATURES --message-format=json --no-run |
-              jq -r '.executable | select(. != null)'
-            )
-            for bin in $test_exe_paths; do
-              SRC=$bin
-              DST=/tmp/neon/test_bin/$(basename $bin)
-
-              # We don't need debug symbols for code coverage, so strip them out to make
-              # the artifact smaller.
-              strip "$SRC" -o "$DST"
-              echo "$DST" >> /tmp/coverage/binaries.list
-            done
-
-            for bin in $binaries; do
-              echo "/tmp/neon/bin/$bin" >> /tmp/coverage/binaries.list
-            done
-          fi
-
-      - name: Run rust tests
-        env:
-          NEXTEST_RETRIES: 3
-        run: |
-          PQ_LIB_DIR=$(pwd)/pg_install/v16/lib
-          export PQ_LIB_DIR
-          LD_LIBRARY_PATH=$(pwd)/pg_install/v16/lib
-          export LD_LIBRARY_PATH
-
-          #nextest does not yet support running doctests
-          cargo test --doc $CARGO_FLAGS $CARGO_FEATURES
-
-          for io_engine in std-fs tokio-epoll-uring ; do
-            NEON_PAGESERVER_UNIT_TEST_VIRTUAL_FILE_IOENGINE=$io_engine ${cov_prefix} cargo nextest run $CARGO_FLAGS $CARGO_FEATURES
-          done
-
-          # Run separate tests for real S3
-          export ENABLE_REAL_S3_REMOTE_STORAGE=nonempty
-          export REMOTE_STORAGE_S3_BUCKET=neon-github-ci-tests
-          export REMOTE_STORAGE_S3_REGION=eu-central-1
-          ${cov_prefix} cargo nextest run $CARGO_FLAGS $CARGO_FEATURES -E 'package(remote_storage)' -E 'test(test_real_s3)'
-
-          # Run separate tests for real Azure Blob Storage
-          # XXX: replace region with `eu-central-1`-like region
-          export ENABLE_REAL_AZURE_REMOTE_STORAGE=y
-          export AZURE_STORAGE_ACCOUNT="${{ secrets.AZURE_STORAGE_ACCOUNT_DEV }}"
-          export AZURE_STORAGE_ACCESS_KEY="${{ secrets.AZURE_STORAGE_ACCESS_KEY_DEV }}"
-          export REMOTE_STORAGE_AZURE_CONTAINER="${{ vars.REMOTE_STORAGE_AZURE_CONTAINER }}"
-          export REMOTE_STORAGE_AZURE_REGION="${{ vars.REMOTE_STORAGE_AZURE_REGION }}"
-          ${cov_prefix} cargo nextest run $CARGO_FLAGS $CARGO_FEATURES -E 'package(remote_storage)' -E 'test(test_real_azure)'
-
-      - name: Install postgres binaries
-        run: cp -a pg_install /tmp/neon/pg_install
-
-      - name: Upload Neon artifact
-        uses: ./.github/actions/upload
-        with:
-          name: neon-${{ runner.os }}-${{ runner.arch }}-${{ inputs.build-type }}-artifact
-          path: /tmp/neon
-
-      # 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'
-        uses: ./.github/actions/save-coverage-data
-
-  regress-tests:
-    # Run test on x64 only
-    if: inputs.arch == 'x64'
-    needs: [ build-neon ]
-    runs-on: ${{ fromJson(format('["self-hosted", "{0}"]', inputs.arch == 'arm64' && 'large-arm64' || 'large')) }}
-    container:
-      image: ${{ inputs.build-tools-image }}
-      credentials:
-        username: ${{ secrets.NEON_DOCKERHUB_USERNAME }}
-        password: ${{ secrets.NEON_DOCKERHUB_PASSWORD }}
-      # for changed limits, see comments on `options:` earlier in this file
-      options: --init --shm-size=512mb --ulimit memlock=67108864:67108864
-    strategy:
-      fail-fast: false
-      matrix:
-        pg_version: ${{ fromJson(inputs.pg-versions) }}
-    steps:
-      - uses: actions/checkout@v4
-        with:
-          submodules: true
-          fetch-depth: 1
-
-      - name: Pytest regression tests
-        uses: ./.github/actions/run-python-test-set
-        timeout-minutes: 60
-        with:
-          build_type: ${{ inputs.build-type }}
-          test_selection: regress
-          needs_postgres_source: true
-          run_with_real_s3: true
-          real_s3_bucket: neon-github-ci-tests
-          real_s3_region: eu-central-1
-          rerun_flaky: true
-          pg_version: ${{ matrix.pg_version }}
-        env:
-          TEST_RESULT_CONNSTR: ${{ secrets.REGRESS_TEST_RESULT_CONNSTR_NEW }}
-          CHECK_ONDISK_DATA_COMPATIBILITY: nonempty
-          BUILD_TAG: ${{ inputs.build-tag }}
-          PAGESERVER_VIRTUAL_FILE_IO_ENGINE: tokio-epoll-uring
-
-      # Temporary disable this step until we figure out why it's so flaky
-      # Ref https://github.com/neondatabase/neon/issues/4540
-      - name: Merge and upload coverage data
-        if: |
-          false &&
-          inputs.build-type == 'debug' && matrix.pg_version == 'v16'
-        uses: ./.github/actions/save-coverage-data

.github/workflows/benchmarking.yml

@@ -56,24 +56,17 @@ concurrency:
 jobs:
   bench:
     if: ${{ github.event.inputs.run_only_pgvector_tests == 'false' || github.event.inputs.run_only_pgvector_tests == null }}
-    permissions:
-      contents: write
-      statuses: write
-      id-token: write # Required for OIDC authentication in azure runners
     strategy:
-      fail-fast: false
       matrix:
         include:
-          - DEFAULT_PG_VERSION: 16
+          - DEFAULT_PG_VERSION: 14
             PLATFORM: "neon-staging"
             region_id: ${{ github.event.inputs.region_id || 'aws-us-east-2' }}
-            RUNNER: [ self-hosted, us-east-2, x64 ]
-            IMAGE: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/build-tools:pinned
+            provisioner: 'k8s-pod' 
           - DEFAULT_PG_VERSION: 16
             PLATFORM: "azure-staging"
             region_id: 'azure-eastus2'
-            RUNNER: [ self-hosted, eastus2, x64 ]
-            IMAGE: neondatabase/build-tools:pinned
+            provisioner: 'k8s-neonvm'
     env:
       TEST_PG_BENCH_DURATIONS_MATRIX: "300"
       TEST_PG_BENCH_SCALES_MATRIX: "10,100"
@@ -84,21 +77,14 @@ jobs:
       SAVE_PERF_REPORT: ${{ github.event.inputs.save_perf_report || ( github.ref_name == 'main' ) }}
       PLATFORM: ${{ matrix.PLATFORM }}
 
-    runs-on: ${{ matrix.RUNNER }}
+    runs-on: [ self-hosted, us-east-2, x64 ]
     container:
-      image: ${{ matrix.IMAGE }}
+      image: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/build-tools:pinned
       options: --init
 
     steps:
     - uses: actions/checkout@v4
 
-    - name: Configure AWS credentials # necessary on Azure runners
-      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: Download Neon artifact
       uses: ./.github/actions/download
       with:
@@ -113,6 +99,7 @@ jobs:
         region_id: ${{ matrix.region_id }}
         postgres_version: ${{ env.DEFAULT_PG_VERSION }}
         api_key: ${{ secrets.NEON_STAGING_API_KEY }}
+        provisioner: ${{ matrix.provisioner }}
 
     - name: Run benchmark
       uses: ./.github/actions/run-python-test-set
@@ -159,10 +146,9 @@ jobs:
         SLACK_BOT_TOKEN: ${{ secrets.SLACK_BOT_TOKEN }}
 
   replication-tests:
-    if: ${{ github.event.inputs.run_only_pgvector_tests == 'false' || github.event.inputs.run_only_pgvector_tests == null }}
     env:
       POSTGRES_DISTRIB_DIR: /tmp/neon/pg_install
-      DEFAULT_PG_VERSION: 16
+      DEFAULT_PG_VERSION: 14
       TEST_OUTPUT: /tmp/test_output
       BUILD_TYPE: remote
       SAVE_PERF_REPORT: ${{ github.event.inputs.save_perf_report || ( github.ref_name == 'main' ) }}
@@ -176,7 +162,6 @@ jobs:
     steps:
     - uses: actions/checkout@v4
 
-    
     - name: Download Neon artifact
       uses: ./.github/actions/download
       with:
@@ -184,7 +169,7 @@ jobs:
         path: /tmp/neon/
         prefix: latest
 
-    - name: Run Logical Replication benchmarks
+    - name: Run benchmark
       uses: ./.github/actions/run-python-test-set
       with:
         build_type: ${{ env.BUILD_TYPE }}
@@ -192,15 +177,12 @@ jobs:
         run_in_parallel: false
         save_perf_report: ${{ env.SAVE_PERF_REPORT }}
         extra_params: -m remote_cluster --timeout 5400
-        pg_version: ${{ env.DEFAULT_PG_VERSION }}
       env:
         VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
         PERF_TEST_RESULT_CONNSTR: "${{ secrets.PERF_TEST_RESULT_CONNSTR }}"
         NEON_API_KEY: ${{ secrets.NEON_STAGING_API_KEY }}
-        BENCHMARK_PROJECT_ID_PUB: ${{ vars.BENCHMARK_PROJECT_ID_PUB }}
-        BENCHMARK_PROJECT_ID_SUB: ${{ vars.BENCHMARK_PROJECT_ID_SUB }}
 
-    - name: Run Physical Replication benchmarks
+    - name: Run benchmark
       uses: ./.github/actions/run-python-test-set
       with:
         build_type: ${{ env.BUILD_TYPE }}
@@ -208,7 +190,6 @@ jobs:
         run_in_parallel: false
         save_perf_report: ${{ env.SAVE_PERF_REPORT }}
         extra_params: -m remote_cluster --timeout 5400
-        pg_version: ${{ env.DEFAULT_PG_VERSION }}
       env:
         VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
         PERF_TEST_RESULT_CONNSTR: "${{ secrets.PERF_TEST_RESULT_CONNSTR }}"
@@ -232,16 +213,13 @@ jobs:
     # Create matrices for the benchmarking jobs, so we run benchmarks on rds only once a week (on Saturday)
     #
     # Available platforms:
-    # - neonvm-captest-new: Freshly created project (1 CU)
-    # - neonvm-captest-freetier: Use freetier-sized compute (0.25 CU)
-    # - neonvm-captest-azure-new: Freshly created project (1 CU) in azure region
-    # - neonvm-captest-azure-freetier: Use freetier-sized compute (0.25 CU) in azure region
-    # - neonvm-captest-reuse: Reusing existing project
+    # - neon-captest-new: Freshly created project (1 CU)
+    # - neon-captest-freetier: Use freetier-sized compute (0.25 CU)
+    # - neon-captest-reuse: Reusing existing project
     # - rds-aurora: Aurora Postgres Serverless v2 with autoscaling from 0.5 to 2 ACUs
     # - rds-postgres: RDS Postgres db.m5.large instance (2 vCPU, 8 GiB) with gp3 EBS storage
     env:
       RUN_AWS_RDS_AND_AURORA: ${{ github.event.inputs.run_AWS_RDS_AND_AURORA || 'false' }}
-      DEFAULT_REGION_ID: ${{ github.event.inputs.region_id || 'aws-us-east-2' }}
     runs-on: ubuntu-22.04
     outputs:
       pgbench-compare-matrix: ${{ steps.pgbench-compare-matrix.outputs.matrix }}
@@ -252,36 +230,23 @@ jobs:
     - name: Generate matrix for pgbench benchmark
       id: pgbench-compare-matrix
       run: |
-        region_id_default=${{ env.DEFAULT_REGION_ID }}
-        runner_default='["self-hosted", "us-east-2", "x64"]'
-        runner_azure='["self-hosted", "eastus2", "x64"]'
-        image_default="369495373322.dkr.ecr.eu-central-1.amazonaws.com/build-tools:pinned"
         matrix='{
-          "pg_version" : [
-            16
-          ],
-          "region_id" : [
-            "'"$region_id_default"'"
-            ],
           "platform": [
-            "neonvm-captest-new",
-            "neonvm-captest-reuse",
+            "neon-captest-new",
+            "neon-captest-reuse",
             "neonvm-captest-new"
           ],
           "db_size": [ "10gb" ],
-          "runner": ['"$runner_default"'],
-          "image": [ "'"$image_default"'" ],
-          "include": [{ "pg_version": 16, "region_id": "'"$region_id_default"'", "platform": "neonvm-captest-freetier",       "db_size": "3gb" ,"runner": '"$runner_default"', "image": "'"$image_default"'" },
-                      { "pg_version": 16, "region_id": "'"$region_id_default"'", "platform": "neonvm-captest-new",            "db_size": "10gb","runner": '"$runner_default"', "image": "'"$image_default"'" },
-                      { "pg_version": 16, "region_id": "'"$region_id_default"'", "platform": "neonvm-captest-new",            "db_size": "50gb","runner": '"$runner_default"', "image": "'"$image_default"'" },
-                      { "pg_version": 16, "region_id": "azure-eastus2",          "platform": "neonvm-azure-captest-freetier", "db_size": "3gb" ,"runner": '"$runner_azure"',   "image": "neondatabase/build-tools:pinned" },
-                      { "pg_version": 16, "region_id": "azure-eastus2",          "platform": "neonvm-azure-captest-new",      "db_size": "10gb","runner": '"$runner_azure"',   "image": "neondatabase/build-tools:pinned" },
-                      { "pg_version": 16, "region_id": "azure-eastus2",          "platform": "neonvm-azure-captest-new",      "db_size": "50gb","runner": '"$runner_azure"',   "image": "neondatabase/build-tools:pinned" },
-                      { "pg_version": 16, "region_id": "'"$region_id_default"'", "platform": "neonvm-captest-sharding-reuse", "db_size": "50gb","runner": '"$runner_default"', "image": "'"$image_default"'" }]
+          "include": [{ "platform": "neon-captest-freetier",         "db_size": "3gb"  },
+                      { "platform": "neon-captest-new",              "db_size": "50gb" },
+                      { "platform": "neonvm-captest-freetier",       "db_size": "3gb"  },
+                      { "platform": "neonvm-captest-new",            "db_size": "50gb" },
+                      { "platform": "neonvm-captest-sharding-reuse", "db_size": "50gb" }]
         }'
 
         if [ "$(date +%A)" = "Saturday" ]; then
-          matrix=$(echo "$matrix" | jq '.include += [{ "pg_version": 14, "region_id": "'"$region_id_default"'", "platform": "rds-postgres", "db_size": "10gb","runner": '"$runner_default"', "image": "'"$image_default"'" }]')
+          matrix=$(echo "$matrix" | jq '.include += [{ "platform": "rds-postgres", "db_size": "10gb"},
+                                                     { "platform": "rds-aurora",   "db_size": "50gb"}]')
         fi
 
         echo "matrix=$(echo "$matrix" | jq --compact-output '.')" >> $GITHUB_OUTPUT
@@ -291,7 +256,7 @@ jobs:
       run: |
         matrix='{
           "platform": [
-            "neonvm-captest-reuse"
+            "neon-captest-reuse"
           ]
         }'
 
@@ -307,7 +272,7 @@ jobs:
       run: |
         matrix='{
           "platform": [
-            "neonvm-captest-reuse"
+            "neon-captest-reuse"
           ],
           "scale": [
             "10"
@@ -324,10 +289,6 @@ jobs:
   pgbench-compare:
     if: ${{ github.event.inputs.run_only_pgvector_tests == 'false' || github.event.inputs.run_only_pgvector_tests == null }}
     needs: [ generate-matrices ]
-    permissions:
-      contents: write
-      statuses: write
-      id-token: write # Required for OIDC authentication in azure runners
 
     strategy:
       fail-fast: false
@@ -337,15 +298,15 @@ 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 }}
+      DEFAULT_PG_VERSION: 14
       TEST_OUTPUT: /tmp/test_output
       BUILD_TYPE: remote
       SAVE_PERF_REPORT: ${{ github.event.inputs.save_perf_report || ( github.ref_name == 'main' ) }}
       PLATFORM: ${{ matrix.platform }}
 
-    runs-on: ${{ matrix.runner }}
+    runs-on: [ self-hosted, us-east-2, x64 ]
     container:
-      image: ${{ matrix.image }}
+      image: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/build-tools:pinned
       options: --init
 
     # Increase timeout to 8h, default timeout is 6h
@@ -354,13 +315,6 @@ jobs:
     steps:
     - uses: actions/checkout@v4
 
-    - name: Configure AWS credentials # necessary on Azure runners
-      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: Download Neon artifact
       uses: ./.github/actions/download
       with:
@@ -369,26 +323,27 @@ jobs:
         prefix: latest
 
     - name: Create Neon Project
-      if: contains(fromJson('["neonvm-captest-new", "neonvm-captest-freetier", "neonvm-azure-captest-freetier", "neonvm-azure-captest-new"]'), matrix.platform)
+      if: contains(fromJson('["neon-captest-new", "neon-captest-freetier", "neonvm-captest-new", "neonvm-captest-freetier"]'), matrix.platform)
       id: create-neon-project
       uses: ./.github/actions/neon-project-create
       with:
-        region_id: ${{ matrix.region_id }}
+        region_id: ${{ github.event.inputs.region_id || 'aws-us-east-2' }}
         postgres_version: ${{ env.DEFAULT_PG_VERSION }}
         api_key: ${{ secrets.NEON_STAGING_API_KEY }}
-        compute_units: ${{ (contains(matrix.platform, 'captest-freetier') && '[0.25, 0.25]') || '[1, 1]' }}
+        compute_units: ${{ (matrix.platform == 'neon-captest-freetier' && '[0.25, 0.25]') || '[1, 1]' }}
+        provisioner: ${{ (contains(matrix.platform, 'neonvm-') && 'k8s-neonvm') || 'k8s-pod' }}
 
     - name: Set up Connection String
       id: set-up-connstr
       run: |
         case "${PLATFORM}" in
-          neonvm-captest-reuse)
+          neon-captest-reuse)
             CONNSTR=${{ secrets.BENCHMARK_CAPTEST_CONNSTR }}
             ;;
           neonvm-captest-sharding-reuse)
             CONNSTR=${{ secrets.BENCHMARK_CAPTEST_SHARDING_CONNSTR }}
             ;;
-          neonvm-captest-new | neonvm-captest-freetier | neonvm-azure-captest-new | neonvm-azure-captest-freetier)
+          neon-captest-new | neon-captest-freetier | neonvm-captest-new | neonvm-captest-freetier)
             CONNSTR=${{ steps.create-neon-project.outputs.dsn }}
             ;;
           rds-aurora)
@@ -413,7 +368,6 @@ 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 }}
       env:
         BENCHMARK_CONNSTR: ${{ steps.set-up-connstr.outputs.connstr }}
         VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
@@ -427,7 +381,6 @@ 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 }}
       env:
         BENCHMARK_CONNSTR: ${{ steps.set-up-connstr.outputs.connstr }}
         VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
@@ -441,7 +394,6 @@ 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 }}
       env:
         BENCHMARK_CONNSTR: ${{ steps.set-up-connstr.outputs.connstr }}
         VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
@@ -468,21 +420,6 @@ jobs:
         SLACK_BOT_TOKEN: ${{ secrets.SLACK_BOT_TOKEN }}
 
   pgbench-pgvector:
-    permissions:
-      contents: write
-      statuses: write
-      id-token: write # Required for OIDC authentication in azure runners
-    strategy:
-      fail-fast: false
-      matrix:
-        include:
-          - PLATFORM: "neonvm-captest-pgvector"
-            RUNNER: [ self-hosted, us-east-2, x64 ]
-            IMAGE: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/build-tools:pinned
-          - PLATFORM: "azure-captest-pgvector"
-            RUNNER: [ self-hosted, eastus2, x64 ]
-            IMAGE: neondatabase/build-tools:pinned
-
     env:
       TEST_PG_BENCH_DURATIONS_MATRIX: "15m"
       TEST_PG_BENCH_SCALES_MATRIX: "1"
@@ -490,60 +427,30 @@ jobs:
       DEFAULT_PG_VERSION: 16
       TEST_OUTPUT: /tmp/test_output
       BUILD_TYPE: remote
-      LD_LIBRARY_PATH: /home/nonroot/pg/usr/lib/x86_64-linux-gnu
       SAVE_PERF_REPORT: ${{ github.event.inputs.save_perf_report || ( github.ref_name == 'main' ) }}
-      PLATFORM: ${{ matrix.PLATFORM }}
+      PLATFORM: "neon-captest-pgvector"
 
-    runs-on: ${{ matrix.RUNNER }}
+    runs-on: [ self-hosted, us-east-2, x64 ]
     container:
-      image: ${{ matrix.IMAGE }}
+      image: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/build-tools:pinned
       options: --init
 
     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: |
-        cd /home/nonroot
-        wget -q https://apt.postgresql.org/pub/repos/apt/pool/main/p/postgresql-16/libpq5_16.4-1.pgdg110%2B1_amd64.deb
-        wget -q https://apt.postgresql.org/pub/repos/apt/pool/main/p/postgresql-16/postgresql-client-16_16.4-1.pgdg110%2B1_amd64.deb
-        wget -q https://apt.postgresql.org/pub/repos/apt/pool/main/p/postgresql-16/postgresql-16_16.4-1.pgdg110%2B1_amd64.deb 
-        dpkg -x libpq5_16.4-1.pgdg110+1_amd64.deb pg
-        dpkg -x postgresql-client-16_16.4-1.pgdg110+1_amd64.deb pg
-        dpkg -x postgresql-16_16.4-1.pgdg110+1_amd64.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/x86_64-linux-gnu /tmp/neon/pg_install/v16/lib 
-        /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
 
     - name: Set up Connection String
       id: set-up-connstr
       run: |
-        case "${PLATFORM}" in
-          neonvm-captest-pgvector)
-            CONNSTR=${{ secrets.BENCHMARK_PGVECTOR_CONNSTR }}
-            ;;
-          azure-captest-pgvector)
-            CONNSTR=${{ secrets.BENCHMARK_PGVECTOR_CONNSTR_AZURE }}
-            ;;
-          *)
-            echo >&2 "Unknown PLATFORM=${PLATFORM}"
-            exit 1
-            ;;
-        esac
+        CONNSTR=${{ secrets.BENCHMARK_PGVECTOR_CONNSTR }}
 
         echo "connstr=${CONNSTR}" >> $GITHUB_OUTPUT
-        
-    - name: Configure AWS credentials # necessary on Azure runners to read/write from/to S3
-      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
@@ -553,7 +460,6 @@ 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: ${{ env.DEFAULT_PG_VERSION }}
       env:
         VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
         PERF_TEST_RESULT_CONNSTR: "${{ secrets.PERF_TEST_RESULT_CONNSTR }}"
@@ -567,12 +473,11 @@ 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 }}
       env:
         BENCHMARK_CONNSTR: ${{ steps.set-up-connstr.outputs.connstr }}
         VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
         PERF_TEST_RESULT_CONNSTR: "${{ secrets.PERF_TEST_RESULT_CONNSTR }}"
-    
+
     - name: Create Allure report
       if: ${{ !cancelled() }}
       uses: ./.github/actions/allure-report-generate
@@ -582,10 +487,11 @@ jobs:
       uses: slackapi/slack-github-action@v1
       with:
         channel-id: "C033QLM5P7D" # dev-staging-stream
-        slack-message: "Periodic perf testing ${PLATFORM}: ${{ job.status }}\n${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}"
+        slack-message: "Periodic perf testing neon-captest-pgvector: ${{ job.status }}\n${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}"
       env:
         SLACK_BOT_TOKEN: ${{ secrets.SLACK_BOT_TOKEN }}
 
+
   clickbench-compare:
     # ClichBench DB for rds-aurora and rds-Postgres deployed to the same clusters
     # we use for performance testing in pgbench-compare.
@@ -630,7 +536,7 @@ jobs:
       id: set-up-connstr
       run: |
         case "${PLATFORM}" in
-          neonvm-captest-reuse)
+          neon-captest-reuse)
             CONNSTR=${{ secrets.BENCHMARK_CAPTEST_CLICKBENCH_10M_CONNSTR }}
             ;;
           rds-aurora)
@@ -640,7 +546,7 @@ jobs:
             CONNSTR=${{ secrets.BENCHMARK_RDS_POSTGRES_CLICKBENCH_10M_CONNSTR }}
             ;;
           *)
-            echo >&2 "Unknown PLATFORM=${PLATFORM}. Allowed only 'neonvm-captest-reuse', 'rds-aurora', or 'rds-postgres'"
+            echo >&2 "Unknown PLATFORM=${PLATFORM}. Allowed only 'neon-captest-reuse', 'rds-aurora', or 'rds-postgres'"
             exit 1
             ;;
         esac
@@ -717,7 +623,7 @@ jobs:
     - name: Get Connstring Secret Name
       run: |
         case "${PLATFORM}" in
-          neonvm-captest-reuse)
+          neon-captest-reuse)
             ENV_PLATFORM=CAPTEST_TPCH
             ;;
           rds-aurora)
@@ -727,7 +633,7 @@ jobs:
             ENV_PLATFORM=RDS_AURORA_TPCH
             ;;
           *)
-            echo >&2 "Unknown PLATFORM=${PLATFORM}. Allowed only 'neonvm-captest-reuse', 'rds-aurora', or 'rds-postgres'"
+            echo >&2 "Unknown PLATFORM=${PLATFORM}. Allowed only 'neon-captest-reuse', 'rds-aurora', or 'rds-postgres'"
             exit 1
             ;;
         esac
@@ -804,7 +710,7 @@ jobs:
       id: set-up-connstr
       run: |
         case "${PLATFORM}" in
-          neonvm-captest-reuse)
+          neon-captest-reuse)
             CONNSTR=${{ secrets.BENCHMARK_USER_EXAMPLE_CAPTEST_CONNSTR }}
             ;;
           rds-aurora)
@@ -814,7 +720,7 @@ jobs:
             CONNSTR=${{ secrets.BENCHMARK_USER_EXAMPLE_RDS_POSTGRES_CONNSTR }}
             ;;
           *)
-            echo >&2 "Unknown PLATFORM=${PLATFORM}. Allowed only 'neonvm-captest-reuse', 'rds-aurora', or 'rds-postgres'"
+            echo >&2 "Unknown PLATFORM=${PLATFORM}. Allowed only 'neon-captest-reuse', 'rds-aurora', or 'rds-postgres'"
             exit 1
             ;;
         esac
@@ -829,7 +735,6 @@ 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: ${{ env.DEFAULT_PG_VERSION }}
       env:
         VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
         PERF_TEST_RESULT_CONNSTR: "${{ secrets.PERF_TEST_RESULT_CONNSTR }}"

.github/workflows/build-build-tools-image.yml

@@ -72,12 +72,6 @@ jobs:
           username: ${{ secrets.NEON_DOCKERHUB_USERNAME }}
           password: ${{ secrets.NEON_DOCKERHUB_PASSWORD }}
 
-      - uses: docker/login-action@v3
-        with:
-          registry: cache.neon.build
-          username: ${{ secrets.NEON_CI_DOCKERCACHE_USERNAME }}
-          password: ${{ secrets.NEON_CI_DOCKERCACHE_PASSWORD }}
-
       - uses: docker/build-push-action@v6
         with:
           context: .
@@ -85,8 +79,8 @@ jobs:
           push: true
           pull: true
           file: Dockerfile.build-tools
-          cache-from: type=registry,ref=cache.neon.build/build-tools:cache-${{ matrix.arch }}
-          cache-to: ${{ github.ref_name == 'main' && format('type=registry,ref=cache.neon.build/build-tools:cache-{0},mode=max', matrix.arch) || '' }}
+          cache-from: type=registry,ref=neondatabase/build-tools:cache-${{ matrix.arch }}
+          cache-to: ${{ github.ref_name == 'main' && format('type=registry,ref=neondatabase/build-tools:cache-{0},mode=max', matrix.arch) || '' }}
           tags: neondatabase/build-tools:${{ inputs.image-tag }}-${{ matrix.arch }}
 
       - name: Remove custom docker config directory

.github/workflows/build_and_test.yml

@@ -125,11 +125,7 @@ jobs:
 
   check-codestyle-rust:
     needs: [ check-permissions, build-build-tools-image ]
-    strategy:
-      matrix:
-        arch: [ x64, arm64 ]
-    runs-on: ${{ fromJson(format('["self-hosted", "{0}"]', matrix.arch == 'arm64' && 'small-arm64' || 'small')) }}
-
+    runs-on: [ self-hosted, gen3, small ]
     container:
       image: ${{ needs.build-build-tools-image.outputs.image }}
       credentials:
@@ -197,30 +193,291 @@ jobs:
         if: ${{ !cancelled() }}
         run: cargo deny check --hide-inclusion-graph
 
-  build-and-test-locally:
-    needs: [ tag, build-build-tools-image ]
+  build-neon:
+    needs: [ check-permissions, tag, build-build-tools-image ]
+    runs-on: [ self-hosted, gen3, large ]
+    container:
+      image: ${{ needs.build-build-tools-image.outputs.image }}
+      credentials:
+        username: ${{ secrets.NEON_DOCKERHUB_USERNAME }}
+        password: ${{ secrets.NEON_DOCKERHUB_PASSWORD }}
+      # Raise locked memory limit for tokio-epoll-uring.
+      # On 5.10 LTS kernels < 5.10.162 (and generally mainline kernels < 5.12),
+      # io_uring will account the memory of the CQ and SQ as locked.
+      # More details: https://github.com/neondatabase/neon/issues/6373#issuecomment-1905814391
+      options: --init --shm-size=512mb --ulimit memlock=67108864:67108864
     strategy:
       fail-fast: false
       matrix:
-        arch: [ x64 ]
-        # Do not build or run tests in debug for release branches
-        build-type: ${{ fromJson((startsWith(github.ref_name, 'release' && github.event_name == 'push')) && '["release"]' || '["debug", "release"]') }}
-        include:
-          - build-type: release
-            arch: arm64
-    uses: ./.github/workflows/_build-and-test-locally.yml
-    with:
-      arch: ${{ matrix.arch }}
-      build-tools-image: ${{ needs.build-build-tools-image.outputs.image }}
-      build-tag: ${{ needs.tag.outputs.build-tag }}
-      build-type: ${{ matrix.build-type }}
-      # Run tests on all Postgres versions in release builds and only on the latest version in debug builds
-      pg-versions: ${{ matrix.build-type == 'release' && '["v14", "v15", "v16"]' || '["v16"]' }}
-    secrets: inherit
+        build_type: [ debug, release ]
+    env:
+      BUILD_TYPE: ${{ matrix.build_type }}
+      GIT_VERSION: ${{ github.event.pull_request.head.sha || github.sha }}
+      BUILD_TAG: ${{ needs.tag.outputs.build-tag }}
+
+    steps:
+      - name: Fix git ownership
+        run: |
+          # Workaround for `fatal: detected dubious ownership in repository at ...`
+          #
+          # Use both ${{ github.workspace }} and ${GITHUB_WORKSPACE} because they're different on host and in containers
+          #   Ref https://github.com/actions/checkout/issues/785
+          #
+          git config --global --add safe.directory ${{ github.workspace }}
+          git config --global --add safe.directory ${GITHUB_WORKSPACE}
+          for r in 14 15 16; do
+            git config --global --add safe.directory "${{ github.workspace }}/vendor/postgres-v$r"
+            git config --global --add safe.directory "${GITHUB_WORKSPACE}/vendor/postgres-v$r"
+          done
+
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          submodules: true
+          fetch-depth: 1
+
+      - name: Set pg 14 revision for caching
+        id: pg_v14_rev
+        run: echo pg_rev=$(git rev-parse HEAD:vendor/postgres-v14) >> $GITHUB_OUTPUT
+
+      - name: Set pg 15 revision for caching
+        id: pg_v15_rev
+        run: echo pg_rev=$(git rev-parse HEAD:vendor/postgres-v15) >> $GITHUB_OUTPUT
+
+      - name: Set pg 16 revision for caching
+        id: pg_v16_rev
+        run: echo pg_rev=$(git rev-parse HEAD:vendor/postgres-v16) >> $GITHUB_OUTPUT
+
+      # Set some environment variables used by all the steps.
+      #
+      # CARGO_FLAGS is extra options to pass to "cargo build", "cargo test" etc.
+      #   It also includes --features, if any
+      #
+      # CARGO_FEATURES is passed to "cargo metadata". It is separate from CARGO_FLAGS,
+      #   because "cargo metadata" doesn't accept --release or --debug options
+      #
+      # We run tests with addtional features, that are turned off by default (e.g. in release builds), see
+      # corresponding Cargo.toml files for their descriptions.
+      - name: Set env variables
+        run: |
+          CARGO_FEATURES="--features testing"
+          if [[ $BUILD_TYPE == "debug" ]]; then
+            cov_prefix="scripts/coverage --profraw-prefix=$GITHUB_JOB --dir=/tmp/coverage run"
+            CARGO_FLAGS="--locked"
+          elif [[ $BUILD_TYPE == "release" ]]; then
+            cov_prefix=""
+            CARGO_FLAGS="--locked --release"
+          fi
+          {
+            echo "cov_prefix=${cov_prefix}"
+            echo "CARGO_FEATURES=${CARGO_FEATURES}"
+            echo "CARGO_FLAGS=${CARGO_FLAGS}"
+            echo "CARGO_HOME=${GITHUB_WORKSPACE}/.cargo"
+          } >> $GITHUB_ENV
+
+      # Disabled for now
+      # Don't include the ~/.cargo/registry/src directory. It contains just
+      # uncompressed versions of the crates in ~/.cargo/registry/cache
+      # directory, and it's faster to let 'cargo' to rebuild it from the
+      # compressed crates.
+#      - name: Cache cargo deps
+#        id: cache_cargo
+#        uses: actions/cache@v4
+#        with:
+#          path: |
+#            ~/.cargo/registry/
+#            !~/.cargo/registry/src
+#            ~/.cargo/git/
+#            target/
+#          # Fall back to older versions of the key, if no cache for current Cargo.lock was found
+#          key: |
+#            v1-${{ runner.os }}-${{ runner.arch }}-${{ matrix.build_type }}-cargo-${{ hashFiles('rust-toolchain.toml') }}-${{ hashFiles('Cargo.lock') }}
+#            v1-${{ runner.os }}-${{ runner.arch }}-${{ matrix.build_type }}-cargo-${{ hashFiles('rust-toolchain.toml') }}-
+
+      - name: Cache postgres v14 build
+        id: cache_pg_14
+        uses: actions/cache@v4
+        with:
+          path: pg_install/v14
+          key: v1-${{ runner.os }}-${{ runner.arch }}-${{ matrix.build_type }}-pg-${{ steps.pg_v14_rev.outputs.pg_rev }}-${{ hashFiles('Makefile', 'Dockerfile.build-tools') }}
+
+      - name: Cache postgres v15 build
+        id: cache_pg_15
+        uses: actions/cache@v4
+        with:
+          path: pg_install/v15
+          key: v1-${{ runner.os }}-${{ runner.arch }}-${{ matrix.build_type }}-pg-${{ steps.pg_v15_rev.outputs.pg_rev }}-${{ hashFiles('Makefile', 'Dockerfile.build-tools') }}
+
+      - name: Cache postgres v16 build
+        id: cache_pg_16
+        uses: actions/cache@v4
+        with:
+          path: pg_install/v16
+          key: v1-${{ runner.os }}-${{ runner.arch }}-${{ matrix.build_type }}-pg-${{ steps.pg_v16_rev.outputs.pg_rev }}-${{ hashFiles('Makefile', 'Dockerfile.build-tools') }}
+
+      - name: Build postgres v14
+        if: steps.cache_pg_14.outputs.cache-hit != 'true'
+        run: mold -run make postgres-v14 -j$(nproc)
+
+      - name: Build postgres v15
+        if: steps.cache_pg_15.outputs.cache-hit != 'true'
+        run: mold -run make postgres-v15 -j$(nproc)
+
+      - name: Build postgres v16
+        if: steps.cache_pg_16.outputs.cache-hit != 'true'
+        run: mold -run make postgres-v16 -j$(nproc)
+
+      - name: Build neon extensions
+        run: mold -run make neon-pg-ext -j$(nproc)
+
+      - name: Build walproposer-lib
+        run: mold -run make walproposer-lib -j$(nproc)
+
+      - 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 $CARGO_FEATURES --bins --tests
+
+      # Do install *before* running rust tests because they might recompile the
+      # binaries with different features/flags.
+      - name: Install rust binaries
+        run: |
+          # Install target binaries
+          mkdir -p /tmp/neon/bin/
+          binaries=$(
+            ${cov_prefix} cargo metadata $CARGO_FEATURES --format-version=1 --no-deps |
+            jq -r '.packages[].targets[] | select(.kind | index("bin")) | .name'
+          )
+          for bin in $binaries; do
+            SRC=target/$BUILD_TYPE/$bin
+            DST=/tmp/neon/bin/$bin
+            cp "$SRC" "$DST"
+          done
+
+          # Install test executables and write list of all binaries (for code coverage)
+          if [[ $BUILD_TYPE == "debug" ]]; then
+            # Keep bloated coverage data files away from the rest of the artifact
+            mkdir -p /tmp/coverage/
+
+            mkdir -p /tmp/neon/test_bin/
+
+            test_exe_paths=$(
+              ${cov_prefix} cargo test $CARGO_FLAGS $CARGO_FEATURES --message-format=json --no-run |
+              jq -r '.executable | select(. != null)'
+            )
+            for bin in $test_exe_paths; do
+              SRC=$bin
+              DST=/tmp/neon/test_bin/$(basename $bin)
+
+              # We don't need debug symbols for code coverage, so strip them out to make
+              # the artifact smaller.
+              strip "$SRC" -o "$DST"
+              echo "$DST" >> /tmp/coverage/binaries.list
+            done
+
+            for bin in $binaries; do
+              echo "/tmp/neon/bin/$bin" >> /tmp/coverage/binaries.list
+            done
+          fi
+
+      - name: Run rust tests
+        env:
+          NEXTEST_RETRIES: 3
+        run: |
+          PQ_LIB_DIR=$(pwd)/pg_install/v16/lib
+          export PQ_LIB_DIR
+          LD_LIBRARY_PATH=$(pwd)/pg_install/v16/lib
+          export LD_LIBRARY_PATH
+
+          #nextest does not yet support running doctests
+          cargo test --doc $CARGO_FLAGS $CARGO_FEATURES
+
+          for io_engine in std-fs tokio-epoll-uring ; do
+            NEON_PAGESERVER_UNIT_TEST_VIRTUAL_FILE_IOENGINE=$io_engine ${cov_prefix} cargo nextest run $CARGO_FLAGS $CARGO_FEATURES
+          done
+
+          # Run separate tests for real S3
+          export ENABLE_REAL_S3_REMOTE_STORAGE=nonempty
+          export REMOTE_STORAGE_S3_BUCKET=neon-github-ci-tests
+          export REMOTE_STORAGE_S3_REGION=eu-central-1
+          ${cov_prefix} cargo nextest run $CARGO_FLAGS $CARGO_FEATURES -E 'package(remote_storage)' -E 'test(test_real_s3)'
+
+          # Run separate tests for real Azure Blob Storage
+          # XXX: replace region with `eu-central-1`-like region
+          export ENABLE_REAL_AZURE_REMOTE_STORAGE=y
+          export AZURE_STORAGE_ACCOUNT="${{ secrets.AZURE_STORAGE_ACCOUNT_DEV }}"
+          export AZURE_STORAGE_ACCESS_KEY="${{ secrets.AZURE_STORAGE_ACCESS_KEY_DEV }}"
+          export REMOTE_STORAGE_AZURE_CONTAINER="${{ vars.REMOTE_STORAGE_AZURE_CONTAINER }}"
+          export REMOTE_STORAGE_AZURE_REGION="${{ vars.REMOTE_STORAGE_AZURE_REGION }}"
+          ${cov_prefix} cargo nextest run $CARGO_FLAGS $CARGO_FEATURES -E 'package(remote_storage)' -E 'test(test_real_azure)'
+
+      - name: Install postgres binaries
+        run: cp -a pg_install /tmp/neon/pg_install
+
+      - name: Upload Neon artifact
+        uses: ./.github/actions/upload
+        with:
+          name: neon-${{ runner.os }}-${{ runner.arch }}-${{ matrix.build_type }}-artifact
+          path: /tmp/neon
+
+      # XXX: keep this after the binaries.list is formed, so the coverage can properly work later
+      - name: Merge and upload coverage data
+        if: matrix.build_type == 'debug'
+        uses: ./.github/actions/save-coverage-data
+
+  regress-tests:
+    needs: [ check-permissions, build-neon, build-build-tools-image, tag ]
+    runs-on: [ self-hosted, gen3, large ]
+    container:
+      image: ${{ needs.build-build-tools-image.outputs.image }}
+      credentials:
+        username: ${{ secrets.NEON_DOCKERHUB_USERNAME }}
+        password: ${{ secrets.NEON_DOCKERHUB_PASSWORD }}
+      # for changed limits, see comments on `options:` earlier in this file
+      options: --init --shm-size=512mb --ulimit memlock=67108864:67108864
+    strategy:
+      fail-fast: false
+      matrix:
+        build_type: [ debug, release ]
+        pg_version: [ v14, v15, v16 ]
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          submodules: true
+          fetch-depth: 1
+
+      - name: Pytest regression tests
+        uses: ./.github/actions/run-python-test-set
+        timeout-minutes: 60
+        with:
+          build_type: ${{ matrix.build_type }}
+          test_selection: regress
+          needs_postgres_source: true
+          run_with_real_s3: true
+          real_s3_bucket: neon-github-ci-tests
+          real_s3_region: eu-central-1
+          rerun_flaky: true
+          pg_version: ${{ matrix.pg_version }}
+        env:
+          TEST_RESULT_CONNSTR: ${{ secrets.REGRESS_TEST_RESULT_CONNSTR_NEW }}
+          CHECK_ONDISK_DATA_COMPATIBILITY: nonempty
+          BUILD_TAG: ${{ needs.tag.outputs.build-tag }}
+          PAGESERVER_VIRTUAL_FILE_IO_ENGINE: tokio-epoll-uring
+          PAGESERVER_GET_VECTORED_IMPL: vectored
+          PAGESERVER_GET_IMPL: vectored
+          PAGESERVER_VALIDATE_VEC_GET: true
+
+      # Temporary disable this step until we figure out why it's so flaky
+      # Ref https://github.com/neondatabase/neon/issues/4540
+      - name: Merge and upload coverage data
+        if: |
+          false &&
+          matrix.build_type == 'debug' && matrix.pg_version == 'v14'
+        uses: ./.github/actions/save-coverage-data
 
-  # Keep `benchmarks` job outside of `build-and-test-locally` workflow to make job failures non-blocking
   get-benchmarks-durations:
-    if: github.ref_name == 'main' || contains(github.event.pull_request.labels.*.name, 'run-benchmarks')
     outputs:
       json: ${{ steps.get-benchmark-durations.outputs.json }}
     needs: [ check-permissions, build-build-tools-image ]
@@ -231,6 +488,7 @@ jobs:
         username: ${{ secrets.NEON_DOCKERHUB_USERNAME }}
         password: ${{ secrets.NEON_DOCKERHUB_PASSWORD }}
       options: --init
+    if: github.ref_name == 'main' || contains(github.event.pull_request.labels.*.name, 'run-benchmarks')
     steps:
       - name: Checkout
         uses: actions/checkout@v4
@@ -255,8 +513,7 @@ jobs:
           echo "json=$(jq --compact-output '.' /tmp/benchmark_durations.json)" >> $GITHUB_OUTPUT
 
   benchmarks:
-    if: github.ref_name == 'main' || contains(github.event.pull_request.labels.*.name, 'run-benchmarks')
-    needs: [ check-permissions, build-and-test-locally, build-build-tools-image, get-benchmarks-durations ]
+    needs: [ check-permissions, build-neon, build-build-tools-image, get-benchmarks-durations ]
     runs-on: [ self-hosted, gen3, small ]
     container:
       image: ${{ needs.build-build-tools-image.outputs.image }}
@@ -265,6 +522,7 @@ jobs:
         password: ${{ secrets.NEON_DOCKERHUB_PASSWORD }}
       # for changed limits, see comments on `options:` earlier in this file
       options: --init --shm-size=512mb --ulimit memlock=67108864:67108864
+    if: github.ref_name == 'main' || contains(github.event.pull_request.labels.*.name, 'run-benchmarks')
     strategy:
       fail-fast: false
       matrix:
@@ -289,6 +547,9 @@ jobs:
           PERF_TEST_RESULT_CONNSTR: "${{ secrets.PERF_TEST_RESULT_CONNSTR }}"
           TEST_RESULT_CONNSTR: "${{ secrets.REGRESS_TEST_RESULT_CONNSTR_NEW }}"
           PAGESERVER_VIRTUAL_FILE_IO_ENGINE: tokio-epoll-uring
+          PAGESERVER_GET_VECTORED_IMPL: vectored
+          PAGESERVER_GET_IMPL: vectored
+          PAGESERVER_VALIDATE_VEC_GET: false
       # XXX: no coverage data handling here, since benchmarks are run on release builds,
       # while coverage is currently collected for the debug ones
 
@@ -309,7 +570,7 @@ jobs:
         SLACK_BOT_TOKEN: ${{ secrets.SLACK_BOT_TOKEN }}
 
   create-test-report:
-    needs: [ check-permissions, build-and-test-locally, coverage-report, build-build-tools-image, benchmarks ]
+    needs: [ check-permissions, regress-tests, coverage-report, benchmarks, build-build-tools-image ]
     if: ${{ !cancelled() && contains(fromJSON('["skipped", "success"]'), needs.check-permissions.result) }}
     outputs:
       report-url: ${{ steps.create-allure-report.outputs.report-url }}
@@ -360,7 +621,7 @@ jobs:
             })
 
   coverage-report:
-    needs: [ check-permissions, build-build-tools-image, build-and-test-locally ]
+    needs: [ check-permissions, regress-tests, build-build-tools-image ]
     runs-on: [ self-hosted, gen3, small ]
     container:
       image: ${{ needs.build-build-tools-image.outputs.image }}
@@ -499,12 +760,6 @@ jobs:
           username: ${{ secrets.NEON_DOCKERHUB_USERNAME }}
           password: ${{ secrets.NEON_DOCKERHUB_PASSWORD }}
 
-      - uses: docker/login-action@v3
-        with:
-          registry: cache.neon.build
-          username: ${{ secrets.NEON_CI_DOCKERCACHE_USERNAME }}
-          password: ${{ secrets.NEON_CI_DOCKERCACHE_PASSWORD }}
-
       - uses: docker/build-push-action@v6
         with:
           context: .
@@ -516,8 +771,8 @@ jobs:
           push: true
           pull: true
           file: Dockerfile
-          cache-from: type=registry,ref=cache.neon.build/neon:cache-${{ matrix.arch }}
-          cache-to: ${{ github.ref_name == 'main' && format('type=registry,ref=cache.neon.build/neon:cache-{0},mode=max', matrix.arch) || '' }}
+          cache-from: type=registry,ref=neondatabase/neon:cache-${{ matrix.arch }}
+          cache-to: ${{ github.ref_name == 'main' && format('type=registry,ref=neondatabase/neon:cache-{0},mode=max', matrix.arch) || '' }}
           tags: |
             neondatabase/neon:${{ needs.tag.outputs.build-tag }}-${{ matrix.arch }}
 
@@ -596,12 +851,6 @@ jobs:
           username: ${{ secrets.AWS_ACCESS_KEY_DEV }}
           password: ${{ secrets.AWS_SECRET_KEY_DEV }}
 
-      - uses: docker/login-action@v3
-        with:
-          registry: cache.neon.build
-          username: ${{ secrets.NEON_CI_DOCKERCACHE_USERNAME }}
-          password: ${{ secrets.NEON_CI_DOCKERCACHE_PASSWORD }}
-
       - name: Build compute-node image
         uses: docker/build-push-action@v6
         with:
@@ -615,8 +864,8 @@ jobs:
           push: true
           pull: true
           file: Dockerfile.compute-node
-          cache-from: type=registry,ref=cache.neon.build/compute-node-${{ matrix.version }}:cache-${{ matrix.arch }}
-          cache-to: ${{ github.ref_name == 'main' && format('type=registry,ref=cache.neon.build/compute-node-{0}:cache-{1},mode=max', matrix.version, matrix.arch) || '' }}
+          cache-from: type=registry,ref=neondatabase/compute-node-${{ matrix.version }}:cache-${{ matrix.arch }}
+          cache-to: ${{ github.ref_name == 'main' && format('type=registry,ref=neondatabase/compute-node-{0}:cache-{1},mode=max', matrix.version, matrix.arch) || '' }}
           tags: |
             neondatabase/compute-node-${{ matrix.version }}:${{ needs.tag.outputs.build-tag }}-${{ matrix.arch }}
 
@@ -635,8 +884,8 @@ jobs:
           pull: true
           file: Dockerfile.compute-node
           target: neon-pg-ext-test
-          cache-from: type=registry,ref=cache.neon.build/neon-test-extensions-${{ matrix.version }}:cache-${{ matrix.arch }}
-          cache-to: ${{ github.ref_name == 'main' && format('type=registry,ref=cache.neon.build/neon-test-extensions-{0}:cache-{1},mode=max', matrix.version, matrix.arch) || '' }}
+          cache-from: type=registry,ref=neondatabase/neon-test-extensions-${{ matrix.version }}:cache-${{ matrix.arch }}
+          cache-to: ${{ github.ref_name == 'main' && format('type=registry,ref=neondatabase/neon-test-extensions-{0}:cache-{1},mode=max', matrix.version, matrix.arch) || '' }}
           tags: |
             neondatabase/neon-test-extensions-${{ matrix.version }}:${{needs.tag.outputs.build-tag}}-${{ matrix.arch }}
 
@@ -836,9 +1085,6 @@ jobs:
           rm -rf .docker-custom
 
   promote-images:
-    permissions:
-      contents: read  # This is required for actions/checkout
-      id-token: write # This is required for Azure Login to work.
     needs: [ check-permissions, tag, test-images, vm-compute-node-image ]
     runs-on: ubuntu-22.04
 
@@ -865,28 +1111,6 @@ jobs:
                                                neondatabase/vm-compute-node-${version}:${{ needs.tag.outputs.build-tag }}
           done
 
-      - name: Azure login
-        if: github.ref_name == 'main'
-        uses: azure/login@6c251865b4e6290e7b78be643ea2d005bc51f69a  # @v2.1.1
-        with:
-          client-id: ${{ secrets.AZURE_DEV_CLIENT_ID }}
-          tenant-id: ${{ secrets.AZURE_TENANT_ID }}
-          subscription-id: ${{ secrets.AZURE_DEV_SUBSCRIPTION_ID }}
-
-      - name: Login to ACR
-        if: github.ref_name == 'main'
-        run: |
-          az acr login --name=neoneastus2
-
-      - name: Copy docker images to ACR-dev
-        if: github.ref_name == 'main'
-        run: |
-          for image in neon compute-tools {vm-,}compute-node-{v14,v15,v16}; do
-            docker buildx imagetools create \
-              -t neoneastus2.azurecr.io/neondatabase/${image}:${{ needs.tag.outputs.build-tag }} \
-                                        neondatabase/${image}:${{ needs.tag.outputs.build-tag }}
-          done
-
       - name: Add latest tag to images
         if: github.ref_name == 'main'
         run: |
@@ -999,7 +1223,7 @@ jobs:
           exit 1
 
   deploy:
-    needs: [ check-permissions, promote-images, tag, build-and-test-locally, trigger-custom-extensions-build-and-wait ]
+    needs: [ check-permissions, promote-images, tag, regress-tests, trigger-custom-extensions-build-and-wait ]
     if: github.ref_name == 'main' || github.ref_name == 'release'|| github.ref_name == 'release-proxy'
 
     runs-on: [ self-hosted, gen3, small ]
@@ -1100,7 +1324,7 @@ jobs:
             })
 
   promote-compatibility-data:
-    needs: [ check-permissions, promote-images, tag, build-and-test-locally ]
+    needs: [ check-permissions, promote-images, tag, regress-tests ]
     if: github.ref_name == 'release'
 
     runs-on: [ self-hosted, gen3, small ]
@@ -1139,7 +1363,7 @@ jobs:
           done
 
   pin-build-tools-image:
-    needs: [ build-build-tools-image, promote-images, build-and-test-locally ]
+    needs: [ build-build-tools-image, promote-images, regress-tests ]
     if: github.ref_name == 'main'
     uses: ./.github/workflows/pin-build-tools-image.yml
     with:
@@ -1161,7 +1385,7 @@ jobs:
     needs:
       - check-codestyle-python
       - check-codestyle-rust
-      - build-and-test-locally
+      - regress-tests
       - test-images
     runs-on: ubuntu-22.04
     steps:

.github/workflows/neon_extra_builds.yml

@@ -133,6 +133,221 @@ jobs:
       - name: Check that no warnings are produced
         run: ./run_clippy.sh
 
+  check-linux-arm-build:
+    needs: [ check-permissions, build-build-tools-image ]
+    timeout-minutes: 90
+    runs-on: [ self-hosted, small-arm64 ]
+
+    env:
+      # Use release build only, to have less debug info around
+      # Hence keeping target/ (and general cache size) smaller
+      BUILD_TYPE: release
+      CARGO_FEATURES: --features testing
+      CARGO_FLAGS: --release
+      AWS_ACCESS_KEY_ID: ${{ secrets.AWS_ACCESS_KEY_DEV }}
+      AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_KEY_DEV }}
+
+    container:
+      image: ${{ needs.build-build-tools-image.outputs.image }}
+      credentials:
+        username: ${{ secrets.NEON_DOCKERHUB_USERNAME }}
+        password: ${{ secrets.NEON_DOCKERHUB_PASSWORD }}
+      options: --init
+
+    steps:
+      - name: Fix git ownership
+        run: |
+          # Workaround for `fatal: detected dubious ownership in repository at ...`
+          #
+          # Use both ${{ github.workspace }} and ${GITHUB_WORKSPACE} because they're different on host and in containers
+          #   Ref https://github.com/actions/checkout/issues/785
+          #
+          git config --global --add safe.directory ${{ github.workspace }}
+          git config --global --add safe.directory ${GITHUB_WORKSPACE}
+          for r in 14 15 16; do
+            git config --global --add safe.directory "${{ github.workspace }}/vendor/postgres-v$r"
+            git config --global --add safe.directory "${GITHUB_WORKSPACE}/vendor/postgres-v$r"
+          done
+
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          submodules: true
+          fetch-depth: 1
+
+      - name: Set pg 14 revision for caching
+        id: pg_v14_rev
+        run: echo pg_rev=$(git rev-parse HEAD:vendor/postgres-v14) >> $GITHUB_OUTPUT
+
+      - name: Set pg 15 revision for caching
+        id: pg_v15_rev
+        run: echo pg_rev=$(git rev-parse HEAD:vendor/postgres-v15) >> $GITHUB_OUTPUT
+
+      - name: Set pg 16 revision for caching
+        id: pg_v16_rev
+        run: echo pg_rev=$(git rev-parse HEAD:vendor/postgres-v16) >> $GITHUB_OUTPUT
+
+      - name: Set env variables
+        run: |
+          echo "CARGO_HOME=${GITHUB_WORKSPACE}/.cargo" >> $GITHUB_ENV
+
+      - name: Cache postgres v14 build
+        id: cache_pg_14
+        uses: actions/cache@v4
+        with:
+          path: pg_install/v14
+          key: v1-${{ runner.os }}-${{ runner.arch }}-${{ env.BUILD_TYPE }}-pg-${{ steps.pg_v14_rev.outputs.pg_rev }}-${{ hashFiles('Makefile') }}
+
+      - name: Cache postgres v15 build
+        id: cache_pg_15
+        uses: actions/cache@v4
+        with:
+          path: pg_install/v15
+          key: v1-${{ runner.os }}-${{ runner.arch }}-${{ env.BUILD_TYPE }}-pg-${{ steps.pg_v15_rev.outputs.pg_rev }}-${{ hashFiles('Makefile') }}
+
+      - name: Cache postgres v16 build
+        id: cache_pg_16
+        uses: actions/cache@v4
+        with:
+          path: pg_install/v16
+          key: v1-${{ runner.os }}-${{ runner.arch }}-${{ env.BUILD_TYPE }}-pg-${{ steps.pg_v16_rev.outputs.pg_rev }}-${{ hashFiles('Makefile') }}
+
+      - name: Build postgres v14
+        if: steps.cache_pg_14.outputs.cache-hit != 'true'
+        run: mold -run make postgres-v14 -j$(nproc)
+
+      - name: Build postgres v15
+        if: steps.cache_pg_15.outputs.cache-hit != 'true'
+        run: mold -run make postgres-v15 -j$(nproc)
+
+      - name: Build postgres v16
+        if: steps.cache_pg_16.outputs.cache-hit != 'true'
+        run: mold -run make postgres-v16 -j$(nproc)
+
+      - name: Build neon extensions
+        run: mold -run make neon-pg-ext -j$(nproc)
+
+      - name: Build walproposer-lib
+        run: mold -run make walproposer-lib -j$(nproc)
+
+      - name: Run cargo build
+        run: |
+          PQ_LIB_DIR=$(pwd)/pg_install/v16/lib
+          export PQ_LIB_DIR
+          mold -run cargo build --locked $CARGO_FLAGS $CARGO_FEATURES --bins --tests -j$(nproc)
+
+      - name: Run cargo test
+        env:
+          NEXTEST_RETRIES: 3
+        run: |
+          PQ_LIB_DIR=$(pwd)/pg_install/v16/lib
+          export PQ_LIB_DIR
+          LD_LIBRARY_PATH=$(pwd)/pg_install/v16/lib
+          export LD_LIBRARY_PATH
+
+          cargo nextest run $CARGO_FEATURES -j$(nproc)
+
+          # Run separate tests for real S3
+          export ENABLE_REAL_S3_REMOTE_STORAGE=nonempty
+          export REMOTE_STORAGE_S3_BUCKET=neon-github-ci-tests
+          export REMOTE_STORAGE_S3_REGION=eu-central-1
+          # Avoid `$CARGO_FEATURES` since there's no `testing` feature in the e2e tests now
+          cargo nextest run --package remote_storage --test test_real_s3 -j$(nproc)
+
+          # Run separate tests for real Azure Blob Storage
+          # XXX: replace region with `eu-central-1`-like region
+          export ENABLE_REAL_AZURE_REMOTE_STORAGE=y
+          export AZURE_STORAGE_ACCOUNT="${{ secrets.AZURE_STORAGE_ACCOUNT_DEV }}"
+          export AZURE_STORAGE_ACCESS_KEY="${{ secrets.AZURE_STORAGE_ACCESS_KEY_DEV }}"
+          export REMOTE_STORAGE_AZURE_CONTAINER="${{ vars.REMOTE_STORAGE_AZURE_CONTAINER }}"
+          export REMOTE_STORAGE_AZURE_REGION="${{ vars.REMOTE_STORAGE_AZURE_REGION }}"
+          # Avoid `$CARGO_FEATURES` since there's no `testing` feature in the e2e tests now
+          cargo nextest run --package remote_storage --test test_real_azure -j$(nproc)
+
+  check-codestyle-rust-arm:
+    needs: [ check-permissions, build-build-tools-image ]
+    timeout-minutes: 90
+    runs-on: [ self-hosted, small-arm64 ]
+
+    container:
+      image: ${{ needs.build-build-tools-image.outputs.image }}
+      credentials:
+        username: ${{ secrets.NEON_DOCKERHUB_USERNAME }}
+        password: ${{ secrets.NEON_DOCKERHUB_PASSWORD }}
+      options: --init
+
+    strategy:
+      fail-fast: false
+      matrix:
+        build_type: [ debug, release ]
+
+    steps:
+      - name: Fix git ownership
+        run: |
+          # Workaround for `fatal: detected dubious ownership in repository at ...`
+          #
+          # Use both ${{ github.workspace }} and ${GITHUB_WORKSPACE} because they're different on host and in containers
+          #   Ref https://github.com/actions/checkout/issues/785
+          #
+          git config --global --add safe.directory ${{ github.workspace }}
+          git config --global --add safe.directory ${GITHUB_WORKSPACE}
+          for r in 14 15 16; do
+            git config --global --add safe.directory "${{ github.workspace }}/vendor/postgres-v$r"
+            git config --global --add safe.directory "${GITHUB_WORKSPACE}/vendor/postgres-v$r"
+          done
+
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          submodules: true
+          fetch-depth: 1
+
+      # Some of our rust modules use FFI and need those to be checked
+      - name: Get postgres headers
+        run: make postgres-headers -j$(nproc)
+
+      # cargo hack runs the given cargo subcommand (clippy in this case) for all feature combinations.
+      # This will catch compiler & clippy warnings in all feature combinations.
+      # TODO: use cargo hack for build and test as well, but, that's quite expensive.
+      # NB: keep clippy args in sync with ./run_clippy.sh
+      - run: |
+          CLIPPY_COMMON_ARGS="$( source .neon_clippy_args; echo "$CLIPPY_COMMON_ARGS")"
+          if [ "$CLIPPY_COMMON_ARGS" = "" ]; then
+            echo "No clippy args found in .neon_clippy_args"
+            exit 1
+          fi
+          echo "CLIPPY_COMMON_ARGS=${CLIPPY_COMMON_ARGS}" >> $GITHUB_ENV
+
+      - name: Run cargo clippy (debug)
+        if: matrix.build_type == 'debug'
+        run: cargo hack --feature-powerset clippy $CLIPPY_COMMON_ARGS
+      - name: Run cargo clippy (release)
+        if: matrix.build_type == 'release'
+        run: cargo hack --feature-powerset clippy --release $CLIPPY_COMMON_ARGS
+
+      - name: Check documentation generation
+        if: matrix.build_type == 'release'
+        run: cargo doc --workspace --no-deps --document-private-items -j$(nproc)
+        env:
+            RUSTDOCFLAGS: "-Dwarnings -Arustdoc::private_intra_doc_links"
+
+      # Use `${{ !cancelled() }}` to run quck tests after the longer clippy run
+      - name: Check formatting
+        if: ${{ !cancelled() && matrix.build_type == 'release' }}
+        run: cargo fmt --all -- --check
+
+      # https://github.com/facebookincubator/cargo-guppy/tree/bec4e0eb29dcd1faac70b1b5360267fc02bf830e/tools/cargo-hakari#2-keep-the-workspace-hack-up-to-date-in-ci
+      - name: Check rust dependencies
+        if: ${{ !cancelled() && matrix.build_type == 'release' }}
+        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() && matrix.build_type == 'release' }}
+        run: cargo deny check
+
   gather-rust-build-stats:
     needs: [ check-permissions, build-build-tools-image ]
     if: |
@@ -149,6 +364,8 @@ jobs:
 
     env:
       BUILD_TYPE: release
+      # remove the cachepot wrapper and build without crate caches
+      RUSTC_WRAPPER: ""
       # build with incremental compilation produce partial results
       # so do not attempt to cache this build, also disable the incremental compilation
       CARGO_INCREMENTAL: 0

.github/workflows/pg-clients.yml

@@ -13,7 +13,6 @@ on:
     paths:
       - '.github/workflows/pg-clients.yml'
       - 'test_runner/pg_clients/**'
-      - 'test_runner/logical_repl/**'
       - 'poetry.lock'
   workflow_dispatch:
 
@@ -50,101 +49,6 @@ jobs:
       image-tag: ${{ needs.check-build-tools-image.outputs.image-tag }}
     secrets: inherit
 
-  test-logical-replication:
-    needs: [ build-build-tools-image ]
-    runs-on: ubuntu-22.04
-
-    container:
-      image: ${{ needs.build-build-tools-image.outputs.image }}
-      credentials:
-        username: ${{ secrets.NEON_DOCKERHUB_USERNAME }}
-        password: ${{ secrets.NEON_DOCKERHUB_PASSWORD }}
-      options: --init --user root
-    services:
-      clickhouse:
-        image: clickhouse/clickhouse-server:24.6.3.64
-        ports:
-          - 9000:9000
-          - 8123:8123
-      zookeeper:
-        image: quay.io/debezium/zookeeper:2.7
-        ports:
-          - 2181:2181
-      kafka:
-        image: quay.io/debezium/kafka:2.7
-        env:
-          ZOOKEEPER_CONNECT: "zookeeper:2181"
-          KAFKA_ADVERTISED_LISTENERS: PLAINTEXT://kafka:9092
-          KAFKA_BROKER_ID: 1
-          KAFKA_OFFSETS_TOPIC_REPLICATION_FACTOR: 1
-          KAFKA_JMX_PORT: 9991
-        ports:
-          - 9092:9092
-      debezium:
-        image: quay.io/debezium/connect:2.7
-        env:
-          BOOTSTRAP_SERVERS: kafka:9092
-          GROUP_ID: 1
-          CONFIG_STORAGE_TOPIC: debezium-config
-          OFFSET_STORAGE_TOPIC: debezium-offset
-          STATUS_STORAGE_TOPIC: debezium-status
-          DEBEZIUM_CONFIG_CONNECTOR_CLASS: io.debezium.connector.postgresql.PostgresConnector
-        ports:
-          - 8083:8083
-    steps:
-      - uses: actions/checkout@v4
-
-      - name: Download Neon artifact
-        uses: ./.github/actions/download
-        with:
-          name: neon-${{ runner.os }}-${{ runner.arch }}-release-artifact
-          path: /tmp/neon/
-          prefix: latest
-
-      - name: Create Neon Project
-        id: create-neon-project
-        uses: ./.github/actions/neon-project-create
-        with:
-          api_key: ${{ secrets.NEON_STAGING_API_KEY }}
-          postgres_version: ${{ env.DEFAULT_PG_VERSION }}
-
-      - name: Run tests
-        uses: ./.github/actions/run-python-test-set
-        with:
-          build_type: remote
-          test_selection: logical_repl
-          run_in_parallel: false
-          extra_params: -m remote_cluster
-          pg_version: ${{ env.DEFAULT_PG_VERSION }}
-        env:
-          BENCHMARK_CONNSTR: ${{ steps.create-neon-project.outputs.dsn }}
-
-      - name: Delete Neon Project
-        if: always()
-        uses: ./.github/actions/neon-project-delete
-        with:
-          project_id: ${{ steps.create-neon-project.outputs.project_id }}
-          api_key: ${{ secrets.NEON_STAGING_API_KEY }}
-
-      - name: Create Allure report
-        if: ${{ !cancelled() }}
-        id: create-allure-report
-        uses: ./.github/actions/allure-report-generate
-        with:
-          store-test-results-into-db: true
-        env:
-          REGRESS_TEST_RESULT_CONNSTR_NEW: ${{ secrets.REGRESS_TEST_RESULT_CONNSTR_NEW }}
-
-      - name: Post to a Slack channel
-        if: github.event.schedule && failure()
-        uses: slackapi/slack-github-action@v1
-        with:
-          channel-id: "C06KHQVQ7U3" # on-call-qa-staging-stream
-          slack-message: |
-            Testing the logical replication: <${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}|${{ job.status }}> (<${{ steps.create-allure-report.outputs.report-url }}|test report>)
-        env:
-          SLACK_BOT_TOKEN: ${{ secrets.SLACK_BOT_TOKEN }}
-
   test-postgres-client-libs:
     needs: [ build-build-tools-image ]
     runs-on: ubuntu-22.04

.github/workflows/pin-build-tools-image.yml

@@ -7,20 +7,12 @@ on:
         description: 'Source tag'
         required: true
         type: string
-      force:
-        description: 'Force the image to be pinned'
-        default: false
-        type: boolean
   workflow_call:
     inputs:
       from-tag:
         description: 'Source tag'
         required: true
         type: string
-      force:
-        description: 'Force the image to be pinned'
-        default: false
-        type: boolean
 
 defaults:
   run:
@@ -30,18 +22,15 @@ concurrency:
   group: pin-build-tools-image-${{ inputs.from-tag }}
   cancel-in-progress: false
 
-# No permission for GITHUB_TOKEN by default; the **minimal required** set of permissions should be granted in each job.
 permissions: {}
 
-env:
-  FROM_TAG: ${{ inputs.from-tag }}
-  TO_TAG: pinned
-
 jobs:
-  check-manifests:
+  tag-image:
     runs-on: ubuntu-22.04
-    outputs:
-      skip: ${{ steps.check-manifests.outputs.skip }}
+
+    env:
+      FROM_TAG: ${{ inputs.from-tag }}
+      TO_TAG: pinned
 
     steps:
       - name: Check if we really need to pin the image
@@ -58,44 +47,27 @@ jobs:
 
           echo "skip=${skip}" | tee -a $GITHUB_OUTPUT
 
-  tag-image:
-    needs: check-manifests
-
-    # use format(..) to catch both inputs.force = true AND inputs.force = 'true'
-    if: needs.check-manifests.outputs.skip == 'false' || format('{0}', inputs.force) == 'true'
-
-    runs-on: ubuntu-22.04
-
-    permissions:
-      id-token: write # for `azure/login`
-
-    steps:
       - uses: docker/login-action@v3
-
+        if: steps.check-manifests.outputs.skip == 'false'
         with:
           username: ${{ secrets.NEON_DOCKERHUB_USERNAME }}
           password: ${{ secrets.NEON_DOCKERHUB_PASSWORD }}
 
+      - name: Tag build-tools with `${{ env.TO_TAG }}` in Docker Hub
+        if: steps.check-manifests.outputs.skip == 'false'
+        run: |
+          docker buildx imagetools create -t neondatabase/build-tools:${TO_TAG} \
+                                             neondatabase/build-tools:${FROM_TAG}
+
       - uses: docker/login-action@v3
+        if: steps.check-manifests.outputs.skip == 'false'
         with:
           registry: 369495373322.dkr.ecr.eu-central-1.amazonaws.com
           username: ${{ secrets.AWS_ACCESS_KEY_DEV }}
           password: ${{ secrets.AWS_SECRET_KEY_DEV }}
 
-      - name: Azure login
-        uses: azure/login@6c251865b4e6290e7b78be643ea2d005bc51f69a  # @v2.1.1
-        with:
-          client-id: ${{ secrets.AZURE_DEV_CLIENT_ID }}
-          tenant-id: ${{ secrets.AZURE_TENANT_ID }}
-          subscription-id: ${{ secrets.AZURE_DEV_SUBSCRIPTION_ID }}
-
-      - name: Login to ACR
-        run: |
-          az acr login --name=neoneastus2
-
-      - name: Tag build-tools with `${{ env.TO_TAG }}` in Docker Hub, ECR, and ACR
+      - name: Tag build-tools with `${{ env.TO_TAG }}` in ECR
+        if: steps.check-manifests.outputs.skip == 'false'
         run: |
           docker buildx imagetools create -t 369495373322.dkr.ecr.eu-central-1.amazonaws.com/build-tools:${TO_TAG} \
-                                          -t neoneastus2.azurecr.io/neondatabase/build-tools:${TO_TAG} \
-                                          -t neondatabase/build-tools:${TO_TAG} \
                                              neondatabase/build-tools:${FROM_TAG}

.github/workflows/trigger-e2e-tests.yml

@@ -13,6 +13,8 @@ defaults:
 env:
   # A concurrency group that we use for e2e-tests runs, matches `concurrency.group` above with `github.repository` as a prefix
   E2E_CONCURRENCY_GROUP: ${{ github.repository }}-e2e-tests-${{ github.ref_name }}-${{ github.ref_name == 'main' && github.sha || 'anysha' }}
+  AWS_ACCESS_KEY_ID: ${{ secrets.AWS_ACCESS_KEY_DEV }}
+  AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_KEY_DEV }}
 
 jobs:
   cancel-previous-e2e-tests:
@@ -62,35 +64,19 @@ jobs:
     needs: [ tag ]
     runs-on: ubuntu-22.04
     env:
-      EVENT_ACTION: ${{ github.event.action }}
-      GH_TOKEN: ${{ secrets.CI_ACCESS_TOKEN }}
       TAG: ${{ needs.tag.outputs.build-tag }}
     steps:
-      - name: Wait for `promote-images` job to finish
-        # It's important to have a timeout here, the script in the step can run infinitely
-        timeout-minutes: 60
+      - name: check if ecr image are present
+        env:
+          AWS_ACCESS_KEY_ID: ${{ secrets.AWS_ACCESS_KEY_DEV }}
+          AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_KEY_DEV }}
         run: |
-          if [ "${GITHUB_EVENT_NAME}" != "pull_request" ] || [ "${EVENT_ACTION}" != "ready_for_review" ]; then
-            exit 0
-          fi
-
-          # For PRs we use the run id as the tag
-          BUILD_AND_TEST_RUN_ID=${TAG}
-          while true; do
-            conclusion=$(gh run --repo ${GITHUB_REPOSITORY} view ${BUILD_AND_TEST_RUN_ID} --json jobs --jq '.jobs[] | select(.name == "promote-images") | .conclusion')
-            case "$conclusion" in
-              success)
-                break
-                ;;
-              failure | cancelled | skipped)
-                echo "The 'promote-images' job didn't succeed: '${conclusion}'. Exiting..."
-                exit 1
-                ;;
-              *)
-                echo "The 'promote-images' hasn't succeed yet. Waiting..."
-                sleep 60
-                ;;
-            esac
+          for REPO in neon compute-tools compute-node-v14 vm-compute-node-v14 compute-node-v15 vm-compute-node-v15 compute-node-v16 vm-compute-node-v16; do
+            OUTPUT=$(aws ecr describe-images --repository-name ${REPO} --region eu-central-1 --query "imageDetails[?imageTags[?contains(@, '${TAG}')]]" --output text)
+            if [ "$OUTPUT" == "" ]; then
+              echo "$REPO with image tag $TAG not found" >> $GITHUB_OUTPUT
+              exit 1
+            fi
           done
 
       - name: Set e2e-platforms

CODEOWNERS

@@ -1,13 +1,13 @@
 /compute_tools/ @neondatabase/control-plane @neondatabase/compute
 /storage_controller @neondatabase/storage
 /libs/pageserver_api/ @neondatabase/storage
-/libs/postgres_ffi/ @neondatabase/compute @neondatabase/storage
+/libs/postgres_ffi/ @neondatabase/compute @neondatabase/safekeepers
 /libs/remote_storage/ @neondatabase/storage
-/libs/safekeeper_api/ @neondatabase/storage
+/libs/safekeeper_api/ @neondatabase/safekeepers
 /libs/vm_monitor/ @neondatabase/autoscaling
 /pageserver/ @neondatabase/storage
 /pgxn/ @neondatabase/compute
-/pgxn/neon/ @neondatabase/compute @neondatabase/storage
+/pgxn/neon/ @neondatabase/compute @neondatabase/safekeepers
 /proxy/ @neondatabase/proxy
-/safekeeper/ @neondatabase/storage
+/safekeeper/ @neondatabase/safekeepers
 /vendor/ @neondatabase/compute

Cargo.toml

@@ -9,14 +9,10 @@ members = [
     "pageserver/ctl",
     "pageserver/client",
     "pageserver/pagebench",
-    "proxy/core",
-    "proxy/sasl",
-    "proxy/proxy",
-    "proxy/pg_sni_router",
+    "proxy",
     "safekeeper",
     "storage_broker",
     "storage_controller",
-    "storage_controller/client",
     "storage_scrubber",
     "workspace_hack",
     "libs/compute_api",
@@ -129,7 +125,7 @@ parquet = { version = "51.0.0", default-features = false, features = ["zstd"] }
 parquet_derive = "51.0.0"
 pbkdf2 = { version = "0.12.1", features = ["simple", "std"] }
 pin-project-lite = "0.2"
-procfs = "0.16"
+procfs = "0.14"
 prometheus = {version = "0.13", default-features=false, features = ["process"]} # removes protobuf dependency
 prost = "0.11"
 rand = "0.8"
@@ -186,17 +182,14 @@ tower-service = "0.3.2"
 tracing = "0.1"
 tracing-error = "0.2.0"
 tracing-opentelemetry = "0.21.0"
-tracing-subscriber = { version = "0.3", default-features = false, features = ["smallvec", "fmt", "tracing-log", "std", "env-filter", "json"] }
-try-lock = "0.2.5"
+tracing-subscriber = { version = "0.3", default-features = false, features = ["smallvec", "fmt", "tracing-log", "std", "env-filter", "json", "ansi"] }
 twox-hash = { version = "1.6.3", default-features = false }
-typed-json = "0.1"
 url = "2.2"
 urlencoding = "2.1"
 uuid = { version = "1.6.1", features = ["v4", "v7", "serde"] }
 walkdir = "2.3.2"
 rustls-native-certs = "0.7"
 x509-parser = "0.15"
-whoami = "1.5.1"
 
 ## TODO replace this with tracing
 env_logger = "0.10"
@@ -208,6 +201,9 @@ postgres-protocol = { git = "https://github.com/neondatabase/rust-postgres.git",
 postgres-types = { git = "https://github.com/neondatabase/rust-postgres.git", branch="neon" }
 tokio-postgres = { git = "https://github.com/neondatabase/rust-postgres.git", branch="neon" }
 
+## Other git libraries
+heapless = { default-features=false, features=[], git = "https://github.com/japaric/heapless.git", rev = "644653bf3b831c6bb4963be2de24804acf5e5001" } # upstream release pending
+
 ## Local libraries
 compute_api = { version = "0.1", path = "./libs/compute_api/" }
 consumption_metrics = { version = "0.1", path = "./libs/consumption_metrics/" }
@@ -223,7 +219,6 @@ remote_storage = { version = "0.1", path = "./libs/remote_storage/" }
 safekeeper_api = { version = "0.1", path = "./libs/safekeeper_api" }
 desim = { version = "0.1", path = "./libs/desim" }
 storage_broker = { version = "0.1", path = "./storage_broker/" } # Note: main broker code is inside the binary crate, so linking with the library shouldn't be heavy.
-storage_controller_client = { path = "./storage_controller/client" }
 tenant_size_model = { version = "0.1", path = "./libs/tenant_size_model/" }
 tracing-utils = { version = "0.1", path = "./libs/tracing-utils/" }
 utils = { version = "0.1", path = "./libs/utils/" }

Dockerfile

@@ -17,7 +17,7 @@ COPY --chown=nonroot pgxn pgxn
 COPY --chown=nonroot Makefile Makefile
 COPY --chown=nonroot scripts/ninstall.sh scripts/ninstall.sh
 
-ENV BUILD_TYPE=release
+ENV BUILD_TYPE release
 RUN set -e \
     && mold -run make -j $(nproc) -s neon-pg-ext \
     && rm -rf pg_install/build \
@@ -29,12 +29,24 @@ WORKDIR /home/nonroot
 ARG GIT_VERSION=local
 ARG BUILD_TAG
 
+# Enable https://github.com/paritytech/cachepot to cache Rust crates' compilation results in Docker builds.
+# Set up cachepot to use an AWS S3 bucket for cache results, to reuse it between `docker build` invocations.
+# cachepot falls back to local filesystem if S3 is misconfigured, not failing the build
+ARG RUSTC_WRAPPER=cachepot
+ENV AWS_REGION=eu-central-1
+ENV CACHEPOT_S3_KEY_PREFIX=cachepot
+ARG CACHEPOT_BUCKET=neon-github-dev
+#ARG AWS_ACCESS_KEY_ID
+#ARG AWS_SECRET_ACCESS_KEY
+
 COPY --from=pg-build /home/nonroot/pg_install/v14/include/postgresql/server pg_install/v14/include/postgresql/server
 COPY --from=pg-build /home/nonroot/pg_install/v15/include/postgresql/server pg_install/v15/include/postgresql/server
 COPY --from=pg-build /home/nonroot/pg_install/v16/include/postgresql/server pg_install/v16/include/postgresql/server
 COPY --from=pg-build /home/nonroot/pg_install/v16/lib                       pg_install/v16/lib
 COPY --chown=nonroot . .
 
+# Show build caching stats to check if it was used in the end.
+# Has to be the part of the same RUN since cachepot daemon is killed in the end of this RUN, losing the compilation stats.
 RUN set -e \
     && PQ_LIB_DIR=$(pwd)/pg_install/v16/lib RUSTFLAGS="-Clinker=clang -Clink-arg=-fuse-ld=mold -Clink-arg=-Wl,--no-rosegment" cargo build \
       --bin pg_sni_router  \
@@ -46,7 +58,8 @@ RUN set -e \
       --bin proxy  \
       --bin neon_local \
       --bin storage_scrubber \
-      --locked --release
+      --locked --release \
+    && cachepot -s
 
 # Build final image
 #
@@ -80,24 +93,20 @@ COPY --from=pg-build /home/nonroot/postgres_install.tar.gz /data/
 
 # By default, pageserver uses `.neon/` working directory in WORKDIR, so create one and fill it with the dummy config.
 # Now, when `docker run ... pageserver` is run, it can start without errors, yet will have some default dummy values.
-RUN mkdir -p /data/.neon/ && \
-  echo "id=1234" > "/data/.neon/identity.toml" && \
-  echo "broker_endpoint='http://storage_broker:50051'\n" \
-       "pg_distrib_dir='/usr/local/'\n" \
-       "listen_pg_addr='0.0.0.0:6400'\n" \
-       "listen_http_addr='0.0.0.0:9898'\n" \
-  > /data/.neon/pageserver.toml && \
-  chown -R neon:neon /data/.neon
+RUN mkdir -p /data/.neon/ && chown -R neon:neon /data/.neon/ \
+    && /usr/local/bin/pageserver -D /data/.neon/ --init \
+       -c "id=1234" \
+       -c "broker_endpoint='http://storage_broker:50051'" \
+       -c "pg_distrib_dir='/usr/local/'" \
+       -c "listen_pg_addr='0.0.0.0:6400'" \
+       -c "listen_http_addr='0.0.0.0:9898'"
 
 # When running a binary that links with libpq, default to using our most recent postgres version.  Binaries
 # that want a particular postgres version will select it explicitly: this is just a default.
-ENV LD_LIBRARY_PATH=/usr/local/v16/lib
+ENV LD_LIBRARY_PATH /usr/local/v16/lib
 
 
 VOLUME ["/data"]
 USER neon
 EXPOSE 6400
 EXPOSE 9898
-
-CMD ["/usr/local/bin/pageserver", "-D", "/data/.neon"]
-

Dockerfile.build-tools

@@ -58,7 +58,7 @@ RUN set -e \
     && rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/*
 
 # protobuf-compiler (protoc)
-ENV PROTOC_VERSION=25.1
+ENV PROTOC_VERSION 25.1
 RUN curl -fsSL "https://github.com/protocolbuffers/protobuf/releases/download/v${PROTOC_VERSION}/protoc-${PROTOC_VERSION}-linux-$(uname -m | sed 's/aarch64/aarch_64/g').zip" -o "protoc.zip" \
     && unzip -q protoc.zip -d protoc \
     && mv protoc/bin/protoc /usr/local/bin/protoc \
@@ -99,7 +99,7 @@ RUN curl "https://awscli.amazonaws.com/awscli-exe-linux-$(uname -m).zip" -o "aws
     && rm awscliv2.zip
 
 # Mold: A Modern Linker
-ENV MOLD_VERSION=v2.33.0
+ENV MOLD_VERSION v2.31.0
 RUN set -e \
     && git clone https://github.com/rui314/mold.git \
     && mkdir mold/build \
@@ -168,7 +168,7 @@ USER nonroot:nonroot
 WORKDIR /home/nonroot
 
 # Python
-ENV PYTHON_VERSION=3.9.19 \
+ENV PYTHON_VERSION=3.9.18 \
     PYENV_ROOT=/home/nonroot/.pyenv \
     PATH=/home/nonroot/.pyenv/shims:/home/nonroot/.pyenv/bin:/home/nonroot/.poetry/bin:$PATH
 RUN set -e \
@@ -192,14 +192,9 @@ 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=1.80.1
+ENV RUSTC_VERSION=1.79.0
 ENV RUSTUP_HOME="/home/nonroot/.rustup"
 ENV PATH="/home/nonroot/.cargo/bin:${PATH}"
-ARG RUSTFILT_VERSION=0.2.1
-ARG CARGO_HAKARI_VERSION=0.9.30
-ARG CARGO_DENY_VERSION=0.16.1
-ARG CARGO_HACK_VERSION=0.6.31
-ARG CARGO_NEXTEST_VERSION=0.9.72
 RUN curl -sSO https://static.rust-lang.org/rustup/dist/$(uname -m)-unknown-linux-gnu/rustup-init && whoami && \
 	chmod +x rustup-init && \
 	./rustup-init -y --default-toolchain ${RUSTC_VERSION} && \
@@ -208,13 +203,15 @@ RUN curl -sSO https://static.rust-lang.org/rustup/dist/$(uname -m)-unknown-linux
     . "$HOME/.cargo/env" && \
     cargo --version && rustup --version && \
     rustup component add llvm-tools-preview rustfmt clippy && \
-    cargo install rustfilt            --version ${RUSTFILT_VERSION} && \
-    cargo install cargo-hakari        --version ${CARGO_HAKARI_VERSION} && \
-    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 --git https://github.com/paritytech/cachepot && \
+    cargo install rustfilt && \
+    cargo install cargo-hakari && \
+    cargo install cargo-deny --locked && \
+    cargo install cargo-hack && \
+    cargo install cargo-nextest && \
     rm -rf /home/nonroot/.cargo/registry && \
     rm -rf /home/nonroot/.cargo/git
+ENV RUSTC_WRAPPER=cachepot
 
 # Show versions
 RUN whoami \

Dockerfile.compute-node

@@ -94,7 +94,7 @@ RUN wget https://gitlab.com/Oslandia/SFCGAL/-/archive/v1.3.10/SFCGAL-v1.3.10.tar
     DESTDIR=/sfcgal make install -j $(getconf _NPROCESSORS_ONLN) && \
     make clean && cp -R /sfcgal/* /
 
-ENV PATH="/usr/local/pgsql/bin:$PATH"
+ENV PATH "/usr/local/pgsql/bin:$PATH"
 
 RUN wget https://download.osgeo.org/postgis/source/postgis-3.3.3.tar.gz -O postgis.tar.gz && \
     echo "74eb356e3f85f14233791013360881b6748f78081cc688ff9d6f0f673a762d13 postgis.tar.gz" | sha256sum --check && \
@@ -311,12 +311,9 @@ RUN wget https://github.com/iCyberon/pg_hashids/archive/refs/tags/v1.2.1.tar.gz
 FROM build-deps AS rum-pg-build
 COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
 
-COPY patches/rum.patch /rum.patch
-
 RUN wget https://github.com/postgrespro/rum/archive/refs/tags/1.3.13.tar.gz -O rum.tar.gz && \
     echo "6ab370532c965568df6210bd844ac6ba649f53055e48243525b0b7e5c4d69a7d rum.tar.gz" | sha256sum --check && \
     mkdir rum-src && cd rum-src && tar xzf ../rum.tar.gz --strip-components=1 -C . && \
-    patch -p1 < /rum.patch && \
     make -j $(getconf _NPROCESSORS_ONLN) PG_CONFIG=/usr/local/pgsql/bin/pg_config USE_PGXS=1 && \
     make -j $(getconf _NPROCESSORS_ONLN) install PG_CONFIG=/usr/local/pgsql/bin/pg_config USE_PGXS=1 && \
     echo 'trusted = true' >> /usr/local/pgsql/share/extension/rum.control
@@ -411,7 +408,7 @@ FROM build-deps AS timescaledb-pg-build
 COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
 
 ARG PG_VERSION
-ENV PATH="/usr/local/pgsql/bin:$PATH"
+ENV PATH "/usr/local/pgsql/bin:$PATH"
 
 RUN case "${PG_VERSION}" in \
       "v14" | "v15") \
@@ -444,7 +441,7 @@ FROM build-deps AS pg-hint-plan-pg-build
 COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
 
 ARG PG_VERSION
-ENV PATH="/usr/local/pgsql/bin:$PATH"
+ENV PATH "/usr/local/pgsql/bin:$PATH"
 
 RUN case "${PG_VERSION}" in \
       "v14") \
@@ -480,7 +477,7 @@ RUN case "${PG_VERSION}" in \
 FROM build-deps AS pg-cron-pg-build
 COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
 
-ENV PATH="/usr/local/pgsql/bin/:$PATH"
+ENV PATH "/usr/local/pgsql/bin/:$PATH"
 RUN wget https://github.com/citusdata/pg_cron/archive/refs/tags/v1.6.0.tar.gz -O pg_cron.tar.gz && \
     echo "383a627867d730222c272bfd25cd5e151c578d73f696d32910c7db8c665cc7db pg_cron.tar.gz" | sha256sum --check && \
     mkdir pg_cron-src && cd pg_cron-src && tar xzf ../pg_cron.tar.gz --strip-components=1 -C . && \
@@ -506,7 +503,7 @@ RUN apt-get update && \
         libboost-system1.74-dev \
         libeigen3-dev
 
-ENV PATH="/usr/local/pgsql/bin/:/usr/local/pgsql/:$PATH"
+ENV PATH "/usr/local/pgsql/bin/:/usr/local/pgsql/:$PATH"
 RUN wget https://github.com/rdkit/rdkit/archive/refs/tags/Release_2023_03_3.tar.gz -O rdkit.tar.gz && \
     echo "bdbf9a2e6988526bfeb8c56ce3cdfe2998d60ac289078e2215374288185e8c8d rdkit.tar.gz" | sha256sum --check && \
     mkdir rdkit-src && cd rdkit-src && tar xzf ../rdkit.tar.gz --strip-components=1 -C . && \
@@ -546,7 +543,7 @@ RUN wget https://github.com/rdkit/rdkit/archive/refs/tags/Release_2023_03_3.tar.
 FROM build-deps AS pg-uuidv7-pg-build
 COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
 
-ENV PATH="/usr/local/pgsql/bin/:$PATH"
+ENV PATH "/usr/local/pgsql/bin/:$PATH"
 RUN wget https://github.com/fboulnois/pg_uuidv7/archive/refs/tags/v1.0.1.tar.gz -O pg_uuidv7.tar.gz && \
     echo "0d0759ab01b7fb23851ecffb0bce27822e1868a4a5819bfd276101c716637a7a pg_uuidv7.tar.gz" | sha256sum --check && \
     mkdir pg_uuidv7-src && cd pg_uuidv7-src && tar xzf ../pg_uuidv7.tar.gz --strip-components=1 -C . && \
@@ -563,7 +560,7 @@ RUN wget https://github.com/fboulnois/pg_uuidv7/archive/refs/tags/v1.0.1.tar.gz
 FROM build-deps AS pg-roaringbitmap-pg-build
 COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
 
-ENV PATH="/usr/local/pgsql/bin/:$PATH"
+ENV PATH "/usr/local/pgsql/bin/:$PATH"
 RUN wget https://github.com/ChenHuajun/pg_roaringbitmap/archive/refs/tags/v0.5.4.tar.gz -O pg_roaringbitmap.tar.gz && \
     echo "b75201efcb1c2d1b014ec4ae6a22769cc7a224e6e406a587f5784a37b6b5a2aa pg_roaringbitmap.tar.gz" | sha256sum --check && \
     mkdir pg_roaringbitmap-src && cd pg_roaringbitmap-src && tar xzf ../pg_roaringbitmap.tar.gz --strip-components=1 -C . && \
@@ -580,7 +577,7 @@ RUN wget https://github.com/ChenHuajun/pg_roaringbitmap/archive/refs/tags/v0.5.4
 FROM build-deps AS pg-semver-pg-build
 COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
 
-ENV PATH="/usr/local/pgsql/bin/:$PATH"
+ENV PATH "/usr/local/pgsql/bin/:$PATH"
 RUN wget https://github.com/theory/pg-semver/archive/refs/tags/v0.32.1.tar.gz -O pg_semver.tar.gz && \
     echo "fbdaf7512026d62eec03fad8687c15ed509b6ba395bff140acd63d2e4fbe25d7 pg_semver.tar.gz" | sha256sum --check && \
     mkdir pg_semver-src && cd pg_semver-src && tar xzf ../pg_semver.tar.gz --strip-components=1 -C . && \
@@ -598,7 +595,7 @@ FROM build-deps AS pg-embedding-pg-build
 COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
 
 ARG PG_VERSION
-ENV PATH="/usr/local/pgsql/bin/:$PATH"
+ENV PATH "/usr/local/pgsql/bin/:$PATH"
 RUN case "${PG_VERSION}" in \
       "v14" | "v15") \
         export PG_EMBEDDING_VERSION=0.3.5 \
@@ -622,7 +619,7 @@ RUN case "${PG_VERSION}" in \
 FROM build-deps AS pg-anon-pg-build
 COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
 
-ENV PATH="/usr/local/pgsql/bin/:$PATH"
+ENV PATH "/usr/local/pgsql/bin/:$PATH"
 RUN wget  https://github.com/neondatabase/postgresql_anonymizer/archive/refs/tags/neon_1.1.1.tar.gz -O pg_anon.tar.gz && \
     echo "321ea8d5c1648880aafde850a2c576e4a9e7b9933a34ce272efc839328999fa9  pg_anon.tar.gz" | sha256sum --check && \
     mkdir pg_anon-src && cd pg_anon-src && tar xzf ../pg_anon.tar.gz --strip-components=1 -C . && \
@@ -657,7 +654,7 @@ RUN curl -sSO https://static.rust-lang.org/rustup/dist/$(uname -m)-unknown-linux
     chmod +x rustup-init && \
     ./rustup-init -y --no-modify-path --profile minimal --default-toolchain stable && \
     rm rustup-init && \
-    cargo install --locked --version 0.11.3 cargo-pgrx && \
+    cargo install --locked --version 0.10.2 cargo-pgrx && \
     /bin/bash -c 'cargo pgrx init --pg${PG_VERSION:1}=/usr/local/pgsql/bin/pg_config'
 
 USER root
@@ -672,15 +669,10 @@ USER root
 FROM rust-extensions-build AS pg-jsonschema-pg-build
 ARG PG_VERSION
 
-RUN wget https://github.com/supabase/pg_jsonschema/archive/refs/tags/v0.3.1.tar.gz -O pg_jsonschema.tar.gz && \
-    echo "61df3db1ed83cf24f6aa39c826f8818bfa4f0bd33b587fd6b2b1747985642297 pg_jsonschema.tar.gz" | sha256sum --check && \
+RUN wget https://github.com/supabase/pg_jsonschema/archive/refs/tags/v0.2.0.tar.gz -O pg_jsonschema.tar.gz && \
+    echo "9118fc508a6e231e7a39acaa6f066fcd79af17a5db757b47d2eefbe14f7794f0 pg_jsonschema.tar.gz" | sha256sum --check && \
     mkdir pg_jsonschema-src && cd pg_jsonschema-src && tar xzf ../pg_jsonschema.tar.gz --strip-components=1 -C . && \
-    # see commit 252b3685a27a0f4c31a0f91e983c6314838e89e8
-    # `unsafe-postgres` feature allows to build pgx extensions
-    # against postgres forks that decided to change their ABI name (like us).
-    # With that we can build extensions without forking them and using stock
-    # pgx. As this feature is new few manual version bumps were required.
-    sed -i 's/pgrx = "0.11.3"/pgrx = { version = "0.11.3", features = [ "unsafe-postgres" ] }/g' Cargo.toml && \
+    sed -i 's/pgrx = "0.10.2"/pgrx = { version = "0.10.2", features = [ "unsafe-postgres" ] }/g' Cargo.toml && \
     cargo pgrx install --release && \
     echo "trusted = true" >> /usr/local/pgsql/share/extension/pg_jsonschema.control
 
@@ -694,10 +686,10 @@ RUN wget https://github.com/supabase/pg_jsonschema/archive/refs/tags/v0.3.1.tar.
 FROM rust-extensions-build AS pg-graphql-pg-build
 ARG PG_VERSION
 
-RUN wget https://github.com/supabase/pg_graphql/archive/refs/tags/v1.5.7.tar.gz -O pg_graphql.tar.gz && \
-    echo "2b3e567a5b31019cb97ae0e33263c1bcc28580be5a444ac4c8ece5c4be2aea41 pg_graphql.tar.gz" | sha256sum --check && \
+RUN wget https://github.com/supabase/pg_graphql/archive/refs/tags/v1.4.0.tar.gz -O pg_graphql.tar.gz && \
+    echo "bd8dc7230282b3efa9ae5baf053a54151ed0e66881c7c53750e2d0c765776edc pg_graphql.tar.gz" | sha256sum --check && \
     mkdir pg_graphql-src && cd pg_graphql-src && tar xzf ../pg_graphql.tar.gz --strip-components=1 -C . && \
-    sed -i 's/pgrx = "=0.11.3"/pgrx = { version = "0.11.3", features = [ "unsafe-postgres" ] }/g' Cargo.toml && \
+    sed -i 's/pgrx = "=0.10.2"/pgrx = { version = "0.10.2", features = [ "unsafe-postgres" ] }/g' Cargo.toml && \
     cargo pgrx install --release && \
     # it's needed to enable extension because it uses untrusted C language
     sed -i 's/superuser = false/superuser = true/g' /usr/local/pgsql/share/extension/pg_graphql.control && \
@@ -717,9 +709,6 @@ ARG PG_VERSION
 RUN wget https://github.com/kelvich/pg_tiktoken/archive/26806147b17b60763039c6a6878884c41a262318.tar.gz -O pg_tiktoken.tar.gz && \
     echo "e64e55aaa38c259512d3e27c572da22c4637418cf124caba904cd50944e5004e pg_tiktoken.tar.gz" | sha256sum --check && \
     mkdir pg_tiktoken-src && cd pg_tiktoken-src && tar xzf ../pg_tiktoken.tar.gz --strip-components=1 -C . && \
-    # TODO update pgrx version in the pg_tiktoken repo and remove this line
-    sed -i 's/pgrx = { version = "=0.10.2",/pgrx = { version = "0.11.3",/g' Cargo.toml && \
-    sed -i 's/pgrx-tests = "=0.10.2"/pgrx-tests = "0.11.3"/g' Cargo.toml && \
     cargo pgrx install --release && \
     echo "trusted = true" >> /usr/local/pgsql/share/extension/pg_tiktoken.control
 
@@ -733,10 +722,14 @@ RUN wget https://github.com/kelvich/pg_tiktoken/archive/26806147b17b60763039c6a6
 FROM rust-extensions-build AS pg-pgx-ulid-build
 ARG PG_VERSION
 
-RUN wget https://github.com/pksunkara/pgx_ulid/archive/refs/tags/v0.1.5.tar.gz -O pgx_ulid.tar.gz && \
-    echo "9d1659a2da65af0133d5451c454de31b37364e3502087dadf579f790bc8bef17 pgx_ulid.tar.gz" | sha256sum --check && \
+RUN wget https://github.com/pksunkara/pgx_ulid/archive/refs/tags/v0.1.3.tar.gz -O pgx_ulid.tar.gz && \
+    echo "ee5db82945d2d9f2d15597a80cf32de9dca67b897f605beb830561705f12683c pgx_ulid.tar.gz" | sha256sum --check && \
     mkdir pgx_ulid-src && cd pgx_ulid-src && tar xzf ../pgx_ulid.tar.gz --strip-components=1 -C . && \
-    sed -i 's/pgrx       = "^0.11.2"/pgrx = { version = "=0.11.3", features = [ "unsafe-postgres" ] }/g' Cargo.toml && \
+    echo "******************* Apply a patch for Postgres 16 support; delete in the next release ******************" && \
+    wget https://github.com/pksunkara/pgx_ulid/commit/f84954cf63fc8c80d964ac970d9eceed3c791196.patch && \
+    patch -p1 < f84954cf63fc8c80d964ac970d9eceed3c791196.patch && \
+    echo "********************************************************************************************************" && \
+    sed -i 's/pgrx       = "=0.10.2"/pgrx = { version = "=0.10.2", features = [ "unsafe-postgres" ] }/g' Cargo.toml && \
     cargo pgrx install --release && \
     echo "trusted = true" >> /usr/local/pgsql/share/extension/ulid.control
 
@@ -750,7 +743,7 @@ RUN wget https://github.com/pksunkara/pgx_ulid/archive/refs/tags/v0.1.5.tar.gz -
 FROM build-deps AS wal2json-pg-build
 COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
 
-ENV PATH="/usr/local/pgsql/bin/:$PATH"
+ENV PATH "/usr/local/pgsql/bin/:$PATH"
 RUN wget https://github.com/eulerto/wal2json/archive/refs/tags/wal2json_2_5.tar.gz && \
     echo "b516653575541cf221b99cf3f8be9b6821f6dbcfc125675c85f35090f824f00e wal2json_2_5.tar.gz" | sha256sum --check && \
     mkdir wal2json-src && cd wal2json-src && tar xzf ../wal2json_2_5.tar.gz --strip-components=1 -C . && \
@@ -766,7 +759,7 @@ RUN wget https://github.com/eulerto/wal2json/archive/refs/tags/wal2json_2_5.tar.
 FROM build-deps AS pg-ivm-build
 COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
 
-ENV PATH="/usr/local/pgsql/bin/:$PATH"
+ENV PATH "/usr/local/pgsql/bin/:$PATH"
 RUN wget https://github.com/sraoss/pg_ivm/archive/refs/tags/v1.7.tar.gz -O pg_ivm.tar.gz && \
     echo "ebfde04f99203c7be4b0e873f91104090e2e83e5429c32ac242d00f334224d5e pg_ivm.tar.gz" | sha256sum --check && \
     mkdir pg_ivm-src && cd pg_ivm-src && tar xzf ../pg_ivm.tar.gz --strip-components=1 -C . && \
@@ -783,7 +776,7 @@ RUN wget https://github.com/sraoss/pg_ivm/archive/refs/tags/v1.7.tar.gz -O pg_iv
 FROM build-deps AS pg-partman-build
 COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
 
-ENV PATH="/usr/local/pgsql/bin/:$PATH"
+ENV PATH "/usr/local/pgsql/bin/:$PATH"
 RUN wget https://github.com/pgpartman/pg_partman/archive/refs/tags/v5.0.1.tar.gz -O pg_partman.tar.gz && \
     echo "75b541733a9659a6c90dbd40fccb904a630a32880a6e3044d0c4c5f4c8a65525 pg_partman.tar.gz" | sha256sum --check && \
     mkdir pg_partman-src && cd pg_partman-src && tar xzf ../pg_partman.tar.gz --strip-components=1 -C . && \
@@ -933,8 +926,7 @@ COPY --from=pgjwt-pg-build /pgjwt.tar.gz /ext-src
 #COPY --from=pg-tiktoken-pg-build /home/nonroot/pg_tiktoken.tar.gz /ext-src
 COPY --from=hypopg-pg-build /hypopg.tar.gz /ext-src
 COPY --from=pg-hashids-pg-build /pg_hashids.tar.gz /ext-src
-COPY --from=rum-pg-build /rum.tar.gz /ext-src
-COPY patches/rum.patch /ext-src
+#COPY --from=rum-pg-build /rum.tar.gz /ext-src
 #COPY --from=pgtap-pg-build /pgtap.tar.gz /ext-src
 COPY --from=ip4r-pg-build /ip4r.tar.gz /ext-src
 COPY --from=prefix-pg-build /prefix.tar.gz /ext-src
@@ -946,7 +938,7 @@ COPY patches/pg_hintplan.patch /ext-src
 COPY --from=pg-cron-pg-build /pg_cron.tar.gz /ext-src
 COPY patches/pg_cron.patch /ext-src
 #COPY --from=pg-pgx-ulid-build /home/nonroot/pgx_ulid.tar.gz /ext-src
-#COPY --from=rdkit-pg-build /rdkit.tar.gz /ext-src
+COPY --from=rdkit-pg-build /rdkit.tar.gz /ext-src
 COPY --from=pg-uuidv7-pg-build /pg_uuidv7.tar.gz /ext-src
 COPY --from=pg-roaringbitmap-pg-build /pg_roaringbitmap.tar.gz /ext-src
 COPY --from=pg-semver-pg-build /pg_semver.tar.gz /ext-src
@@ -961,7 +953,6 @@ RUN cd /ext-src/ && for f in *.tar.gz; \
     rm -rf $dname; mkdir $dname; tar xzf $f --strip-components=1 -C $dname \
     || exit 1; rm -f $f; done
 RUN cd /ext-src/pgvector-src && patch -p1 <../pgvector.patch
-RUN cd /ext-src/rum-src && patch -p1 <../rum.patch
 # cmake is required for the h3 test
 RUN apt-get update && apt-get install -y cmake
 RUN patch -p1 < /ext-src/pg_hintplan.patch
@@ -1034,6 +1025,6 @@ RUN apt update &&  \
     rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/* && \
     localedef -i en_US -c -f UTF-8 -A /usr/share/locale/locale.alias en_US.UTF-8
 
-ENV LANG=en_US.utf8
+ENV LANG en_US.utf8
 USER postgres
 ENTRYPOINT ["/usr/local/bin/compute_ctl"]

Makefile

@@ -69,8 +69,6 @@ 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"
-
 #
 # Top level Makefile to build Neon and PostgreSQL
 #
@@ -81,24 +79,15 @@ all: neon postgres neon-pg-ext
 #
 # The 'postgres_ffi' depends on the Postgres headers.
 .PHONY: neon
-neon: postgres-headers walproposer-lib cargo-target-dir
+neon: postgres-headers walproposer-lib
 	+@echo "Compiling Neon"
 	$(CARGO_CMD_PREFIX) cargo build $(CARGO_BUILD_FLAGS)
-.PHONY: cargo-target-dir
-cargo-target-dir:
-	# https://github.com/rust-lang/cargo/issues/14281
-	mkdir -p target
-	test -e target/CACHEDIR.TAG || echo "$(CACHEDIR_TAG_CONTENTS)" > target/CACHEDIR.TAG
 
 ### PostgreSQL parts
 # Some rules are duplicated for Postgres v14 and 15. We may want to refactor
 # to avoid the duplication in the future, but it's tolerable for now.
 #
 $(POSTGRES_INSTALL_DIR)/build/%/config.status:
-
-	mkdir -p $(POSTGRES_INSTALL_DIR)
-	test -e $(POSTGRES_INSTALL_DIR)/CACHEDIR.TAG || echo "$(CACHEDIR_TAG_CONTENTS)" > $(POSTGRES_INSTALL_DIR)/CACHEDIR.TAG
-
 	+@echo "Configuring Postgres $* build"
 	@test -s $(ROOT_PROJECT_DIR)/vendor/postgres-$*/configure || { \
 		echo "\nPostgres submodule not found in $(ROOT_PROJECT_DIR)/vendor/postgres-$*/, execute "; \

compute_tools/Cargo.toml

@@ -4,11 +4,6 @@ version = "0.1.0"
 edition.workspace = true
 license.workspace = true
 
-[features]
-default = []
-# Enables test specific features.
-testing = []
-
 [dependencies]
 anyhow.workspace = true
 async-compression.workspace = true

compute_tools/src/compute.rs

@@ -400,15 +400,7 @@ impl ComputeNode {
     pub fn get_basebackup(&self, compute_state: &ComputeState, lsn: Lsn) -> Result<()> {
         let mut retry_period_ms = 500.0;
         let mut attempts = 0;
-        const DEFAULT_ATTEMPTS: u16 = 10;
-        #[cfg(feature = "testing")]
-        let max_attempts = if let Ok(v) = env::var("NEON_COMPUTE_TESTING_BASEBACKUP_RETRIES") {
-            u16::from_str(&v).unwrap()
-        } else {
-            DEFAULT_ATTEMPTS
-        };
-        #[cfg(not(feature = "testing"))]
-        let max_attempts = DEFAULT_ATTEMPTS;
+        let max_attempts = 10;
         loop {
             let result = self.try_get_basebackup(compute_state, lsn);
             match result {

compute_tools/src/migration.rs

@@ -9,9 +9,6 @@ pub(crate) struct MigrationRunner<'m> {
 
 impl<'m> MigrationRunner<'m> {
     pub fn new(client: &'m mut Client, migrations: &'m [&'m str]) -> Self {
-        // The neon_migration.migration_id::id column is a bigint, which is equivalent to an i64
-        assert!(migrations.len() + 1 < i64::MAX as usize);
-
         Self { client, migrations }
     }
 
@@ -25,8 +22,11 @@ impl<'m> MigrationRunner<'m> {
         Ok(row.get::<&str, i64>("id"))
     }
 
-    fn update_migration_id(&mut self, migration_id: i64) -> Result<()> {
-        let setval = format!("UPDATE neon_migration.migration_id SET id={}", migration_id);
+    fn update_migration_id(&mut self) -> Result<()> {
+        let setval = format!(
+            "UPDATE neon_migration.migration_id SET id={}",
+            self.migrations.len()
+        );
 
         self.client
             .simple_query(&setval)
@@ -57,49 +57,44 @@ impl<'m> MigrationRunner<'m> {
     pub fn run_migrations(mut self) -> Result<()> {
         self.prepare_migrations()?;
 
-        let mut current_migration = self.get_migration_id()? as usize;
-        while current_migration < self.migrations.len() {
-            macro_rules! migration_id {
-                ($cm:expr) => {
-                    ($cm + 1) as i64
-                };
-            }
+        let mut current_migration: usize = self.get_migration_id()? as usize;
+        let starting_migration_id = current_migration;
 
+        let query = "BEGIN";
+        self.client
+            .simple_query(query)
+            .context("run_migrations begin")?;
+
+        while current_migration < self.migrations.len() {
             let migration = self.migrations[current_migration];
 
             if migration.starts_with("-- SKIP") {
-                info!("Skipping migration id={}", migration_id!(current_migration));
+                info!("Skipping migration id={}", current_migration);
             } else {
                 info!(
                     "Running migration id={}:\n{}\n",
-                    migration_id!(current_migration),
-                    migration
+                    current_migration, migration
                 );
-
-                self.client
-                    .simple_query("BEGIN")
-                    .context("begin migration")?;
-
                 self.client.simple_query(migration).with_context(|| {
-                    format!(
-                        "run_migrations migration id={}",
-                        migration_id!(current_migration)
-                    )
+                    format!("run_migration current_migration={}", current_migration)
                 })?;
-
-                // Migration IDs start at 1
-                self.update_migration_id(migration_id!(current_migration))?;
-
-                self.client
-                    .simple_query("COMMIT")
-                    .context("commit migration")?;
-
-                info!("Finished migration id={}", migration_id!(current_migration));
             }
 
             current_migration += 1;
         }
 
+        self.update_migration_id()?;
+
+        let query = "COMMIT";
+        self.client
+            .simple_query(query)
+            .context("run_migrations commit")?;
+
+        info!(
+            "Ran {} migrations",
+            (self.migrations.len() - starting_migration_id)
+        );
+
         Ok(())
     }
 }

compute_tools/src/spec.rs

@@ -777,21 +777,21 @@ pub fn handle_migrations(client: &mut Client) -> Result<()> {
 
     // Add new migrations in numerical order.
     let migrations = [
-        include_str!("./migrations/0001-neon_superuser_bypass_rls.sql"),
-        include_str!("./migrations/0002-alter_roles.sql"),
-        include_str!("./migrations/0003-grant_pg_create_subscription_to_neon_superuser.sql"),
-        include_str!("./migrations/0004-grant_pg_monitor_to_neon_superuser.sql"),
-        include_str!("./migrations/0005-grant_all_on_tables_to_neon_superuser.sql"),
-        include_str!("./migrations/0006-grant_all_on_sequences_to_neon_superuser.sql"),
+        include_str!("./migrations/0000-neon_superuser_bypass_rls.sql"),
+        include_str!("./migrations/0001-alter_roles.sql"),
+        include_str!("./migrations/0002-grant_pg_create_subscription_to_neon_superuser.sql"),
+        include_str!("./migrations/0003-grant_pg_monitor_to_neon_superuser.sql"),
+        include_str!("./migrations/0004-grant_all_on_tables_to_neon_superuser.sql"),
+        include_str!("./migrations/0005-grant_all_on_sequences_to_neon_superuser.sql"),
         include_str!(
-            "./migrations/0007-grant_all_on_tables_to_neon_superuser_with_grant_option.sql"
+            "./migrations/0006-grant_all_on_tables_to_neon_superuser_with_grant_option.sql"
         ),
         include_str!(
-            "./migrations/0008-grant_all_on_sequences_to_neon_superuser_with_grant_option.sql"
+            "./migrations/0007-grant_all_on_sequences_to_neon_superuser_with_grant_option.sql"
         ),
-        include_str!("./migrations/0009-revoke_replication_for_previously_allowed_roles.sql"),
+        include_str!("./migrations/0008-revoke_replication_for_previously_allowed_roles.sql"),
         include_str!(
-            "./migrations/0010-grant_snapshot_synchronization_funcs_to_neon_superuser.sql"
+            "./migrations/0009-grant_snapshot_synchronization_funcs_to_neon_superuser.sql"
         ),
     ];
 

control_plane/Cargo.toml

@@ -40,7 +40,6 @@ safekeeper_api.workspace = true
 postgres_connection.workspace = true
 storage_broker.workspace = true
 utils.workspace = true
-whoami.workspace = true
 
 compute_api.workspace = true
 workspace_hack.workspace = true

control_plane/src/background_process.rs

@@ -289,7 +289,7 @@ fn fill_remote_storage_secrets_vars(mut cmd: &mut Command) -> &mut Command {
 
 fn fill_env_vars_prefixed_neon(mut cmd: &mut Command) -> &mut Command {
     for (var, val) in std::env::vars() {
-        if var.starts_with("NEON_") {
+        if var.starts_with("NEON_PAGESERVER_") {
             cmd = cmd.env(var, val);
         }
     }

control_plane/src/bin/neon_local.rs

@@ -21,9 +21,7 @@ use pageserver_api::config::{
     DEFAULT_HTTP_LISTEN_PORT as DEFAULT_PAGESERVER_HTTP_PORT,
     DEFAULT_PG_LISTEN_PORT as DEFAULT_PAGESERVER_PG_PORT,
 };
-use pageserver_api::controller_api::{
-    NodeAvailabilityWrapper, PlacementPolicy, TenantCreateRequest,
-};
+use pageserver_api::controller_api::{PlacementPolicy, TenantCreateRequest};
 use pageserver_api::models::{ShardParameters, TimelineCreateRequest, TimelineInfo};
 use pageserver_api::shard::{ShardCount, ShardStripeSize, TenantShardId};
 use postgres_backend::AuthType;
@@ -1252,70 +1250,9 @@ async fn handle_start_all(
             exit(1);
         }
     }
-
-    neon_start_status_check(env, retry_timeout).await?;
-
     Ok(())
 }
 
-async fn neon_start_status_check(
-    env: &local_env::LocalEnv,
-    retry_timeout: &Duration,
-) -> anyhow::Result<()> {
-    const RETRY_INTERVAL: Duration = Duration::from_millis(100);
-    const NOTICE_AFTER_RETRIES: Duration = Duration::from_secs(5);
-
-    if env.control_plane_api.is_none() {
-        return Ok(());
-    }
-
-    let storcon = StorageController::from_env(env);
-
-    let retries = retry_timeout.as_millis() / RETRY_INTERVAL.as_millis();
-    let notice_after_retries = retry_timeout.as_millis() / NOTICE_AFTER_RETRIES.as_millis();
-
-    println!("\nRunning neon status check");
-
-    for retry in 0..retries {
-        if retry == notice_after_retries {
-            println!("\nNeon status check has not passed yet, continuing to wait")
-        }
-
-        let mut passed = true;
-        let mut nodes = storcon.node_list().await?;
-        let mut pageservers = env.pageservers.clone();
-
-        if nodes.len() != pageservers.len() {
-            continue;
-        }
-
-        nodes.sort_by_key(|ps| ps.id);
-        pageservers.sort_by_key(|ps| ps.id);
-
-        for (idx, pageserver) in pageservers.iter().enumerate() {
-            let node = &nodes[idx];
-            if node.id != pageserver.id {
-                passed = false;
-                break;
-            }
-
-            if !matches!(node.availability, NodeAvailabilityWrapper::Active) {
-                passed = false;
-                break;
-            }
-        }
-
-        if passed {
-            println!("\nNeon started and passed status check");
-            return Ok(());
-        }
-
-        tokio::time::sleep(RETRY_INTERVAL).await;
-    }
-
-    anyhow::bail!("\nNeon passed status check")
-}
-
 async fn handle_stop_all(sub_match: &ArgMatches, env: &local_env::LocalEnv) -> Result<()> {
     let immediate =
         sub_match.get_one::<String>("stop-mode").map(|s| s.as_str()) == Some("immediate");

control_plane/src/broker.rs

@@ -1,9 +1,9 @@
 //! Code to manage the storage broker
 //!
-//! In the local test environment, the storage broker stores its data directly in
+//! In the local test environment, the data for each safekeeper is stored in
 //!
 //! ```text
-//!   .neon
+//!   .neon/safekeepers/<safekeeper id>
 //! ```
 use std::time::Duration;
 

control_plane/src/local_env.rs

@@ -151,31 +151,23 @@ pub struct NeonBroker {
 pub struct NeonStorageControllerConf {
     /// Heartbeat timeout before marking a node offline
     #[serde(with = "humantime_serde")]
-    pub max_offline: Duration,
-
-    #[serde(with = "humantime_serde")]
-    pub max_warming_up: Duration,
+    pub max_unavailable: Duration,
 
     /// Threshold for auto-splitting a tenant into shards
     pub split_threshold: Option<u64>,
-
-    pub max_secondary_lag_bytes: Option<u64>,
 }
 
 impl NeonStorageControllerConf {
     // Use a shorter pageserver unavailability interval than the default to speed up tests.
-    const DEFAULT_MAX_OFFLINE_INTERVAL: std::time::Duration = std::time::Duration::from_secs(10);
-
-    const DEFAULT_MAX_WARMING_UP_INTERVAL: std::time::Duration = std::time::Duration::from_secs(30);
+    const DEFAULT_MAX_UNAVAILABLE_INTERVAL: std::time::Duration =
+        std::time::Duration::from_secs(10);
 }
 
 impl Default for NeonStorageControllerConf {
     fn default() -> Self {
         Self {
-            max_offline: Self::DEFAULT_MAX_OFFLINE_INTERVAL,
-            max_warming_up: Self::DEFAULT_MAX_WARMING_UP_INTERVAL,
+            max_unavailable: Self::DEFAULT_MAX_UNAVAILABLE_INTERVAL,
             split_threshold: None,
-            max_secondary_lag_bytes: None,
         }
     }
 }
@@ -517,6 +509,7 @@ impl LocalEnv {
                 #[derive(serde::Serialize, serde::Deserialize)]
                 // (allow unknown fields, unlike PageServerConf)
                 struct PageserverConfigTomlSubset {
+                    id: NodeId,
                     listen_pg_addr: String,
                     listen_http_addr: String,
                     pg_auth_type: AuthType,
@@ -528,30 +521,18 @@ impl LocalEnv {
                         .with_context(|| format!("read {:?}", config_toml_path))?,
                 )
                 .context("parse pageserver.toml")?;
-                let identity_toml_path = dentry.path().join("identity.toml");
-                #[derive(serde::Serialize, serde::Deserialize)]
-                struct IdentityTomlSubset {
-                    id: NodeId,
-                }
-                let identity_toml: IdentityTomlSubset = toml_edit::de::from_str(
-                    &std::fs::read_to_string(&identity_toml_path)
-                        .with_context(|| format!("read {:?}", identity_toml_path))?,
-                )
-                .context("parse identity.toml")?;
                 let PageserverConfigTomlSubset {
+                    id: config_toml_id,
                     listen_pg_addr,
                     listen_http_addr,
                     pg_auth_type,
                     http_auth_type,
                 } = config_toml;
-                let IdentityTomlSubset {
-                    id: identity_toml_id,
-                } = identity_toml;
                 let conf = PageServerConf {
                     id: {
                         anyhow::ensure!(
-                            identity_toml_id == id,
-                            "id mismatch: identity.toml:id={identity_toml_id} pageserver_(.*) id={id}",
+                            config_toml_id == id,
+                            "id mismatch: config_toml.id={config_toml_id} id={id}",
                         );
                         id
                     },

control_plane/src/pageserver.rs

@@ -1,10 +1,8 @@
 //! Code to manage pageservers
 //!
-//! In the local test environment, the data for each pageserver is stored in
+//! In the local test environment, the pageserver stores its data directly in
 //!
-//! ```text
-//!   .neon/pageserver_<pageserver_id>
-//! ```
+//!   .neon/
 //!
 use std::collections::HashMap;
 
@@ -25,7 +23,6 @@ use pageserver_client::mgmt_api;
 use postgres_backend::AuthType;
 use postgres_connection::{parse_host_port, PgConnectionConfig};
 use utils::auth::{Claims, Scope};
-use utils::id::NodeId;
 use utils::{
     id::{TenantId, TimelineId},
     lsn::Lsn,
@@ -75,10 +72,6 @@ impl PageServerNode {
         }
     }
 
-    fn pageserver_make_identity_toml(&self, node_id: NodeId) -> toml_edit::Document {
-        toml_edit::Document::from_str(&format!("id={node_id}")).unwrap()
-    }
-
     fn pageserver_init_make_toml(
         &self,
         conf: NeonLocalInitPageserverConf,
@@ -127,13 +120,10 @@ impl PageServerNode {
         }
 
         // Apply the user-provided overrides
-        overrides.push({
-            let mut doc =
-                toml_edit::ser::to_document(&conf).expect("we deserialized this from toml earlier");
-            // `id` is written out to `identity.toml` instead of `pageserver.toml`
-            doc.remove("id").expect("it's part of the struct");
-            doc.to_string()
-        });
+        overrides.push(
+            toml_edit::ser::to_string_pretty(&conf)
+                .expect("we deserialized this from toml earlier"),
+        );
 
         // Turn `overrides` into a toml document.
         // TODO: above code is legacy code, it should be refactored to use toml_edit directly.
@@ -194,19 +184,6 @@ impl PageServerNode {
             .write_all(config.to_string().as_bytes())
             .context("write pageserver toml")?;
         drop(config_file);
-
-        let identity_file_path = datadir.join("identity.toml");
-        let mut identity_file = std::fs::OpenOptions::new()
-            .create_new(true)
-            .write(true)
-            .open(identity_file_path)
-            .with_context(|| format!("open identity toml for write: {config_file_path:?}"))?;
-        let identity_toml = self.pageserver_make_identity_toml(node_id);
-        identity_file
-            .write_all(identity_toml.to_string().as_bytes())
-            .context("write identity toml")?;
-        drop(identity_toml);
-
         // TODO: invoke a TBD config-check command to validate that pageserver will start with the written config
 
         // Write metadata file, used by pageserver on startup to register itself with

control_plane/src/storage_controller.rs

@@ -5,9 +5,8 @@ use crate::{
 use camino::{Utf8Path, Utf8PathBuf};
 use pageserver_api::{
     controller_api::{
-        NodeConfigureRequest, NodeDescribeResponse, NodeRegisterRequest, TenantCreateRequest,
-        TenantCreateResponse, TenantLocateResponse, TenantShardMigrateRequest,
-        TenantShardMigrateResponse,
+        NodeConfigureRequest, NodeRegisterRequest, TenantCreateRequest, TenantCreateResponse,
+        TenantLocateResponse, TenantShardMigrateRequest, TenantShardMigrateResponse,
     },
     models::{
         TenantShardSplitRequest, TenantShardSplitResponse, TimelineCreateRequest, TimelineInfo,
@@ -30,6 +29,7 @@ use utils::{
 pub struct StorageController {
     env: LocalEnv,
     listen: String,
+    path: Utf8PathBuf,
     private_key: Option<Vec<u8>>,
     public_key: Option<String>,
     postgres_port: u16,
@@ -41,8 +41,6 @@ const COMMAND: &str = "storage_controller";
 
 const STORAGE_CONTROLLER_POSTGRES_VERSION: u32 = 16;
 
-const DB_NAME: &str = "storage_controller";
-
 #[derive(Serialize, Deserialize)]
 pub struct AttachHookRequest {
     pub tenant_shard_id: TenantShardId,
@@ -67,6 +65,10 @@ pub struct InspectResponse {
 
 impl StorageController {
     pub fn from_env(env: &LocalEnv) -> Self {
+        let path = Utf8PathBuf::from_path_buf(env.base_data_dir.clone())
+            .unwrap()
+            .join("attachments.json");
+
         // Makes no sense to construct this if pageservers aren't going to use it: assume
         // pageservers have control plane API set
         let listen_url = env.control_plane_api.clone().unwrap();
@@ -126,6 +128,7 @@ impl StorageController {
 
         Self {
             env: env.clone(),
+            path,
             listen,
             private_key,
             public_key,
@@ -200,6 +203,7 @@ impl StorageController {
     ///
     /// Returns the database url
     pub async fn setup_database(&self) -> anyhow::Result<String> {
+        const DB_NAME: &str = "storage_controller";
         let database_url = format!("postgresql://localhost:{}/{DB_NAME}", self.postgres_port);
 
         let pg_bin_dir = self.get_pg_bin_dir().await?;
@@ -228,30 +232,6 @@ impl StorageController {
         Ok(database_url)
     }
 
-    pub async fn connect_to_database(
-        &self,
-    ) -> anyhow::Result<(
-        tokio_postgres::Client,
-        tokio_postgres::Connection<tokio_postgres::Socket, tokio_postgres::tls::NoTlsStream>,
-    )> {
-        tokio_postgres::Config::new()
-            .host("localhost")
-            .port(self.postgres_port)
-            // The user is the ambient operating system user name.
-            // That is an impurity which we want to fix in => TODO https://github.com/neondatabase/neon/issues/8400
-            //
-            // Until we get there, use the ambient operating system user name.
-            // Recent tokio-postgres versions default to this if the user isn't specified.
-            // But tokio-postgres fork doesn't have this upstream commit:
-            // https://github.com/sfackler/rust-postgres/commit/cb609be758f3fb5af537f04b584a2ee0cebd5e79
-            // => we should rebase our fork => TODO https://github.com/neondatabase/neon/issues/8399
-            .user(&whoami::username())
-            .dbname(DB_NAME)
-            .connect(tokio_postgres::NoTls)
-            .await
-            .map_err(anyhow::Error::new)
-    }
-
     pub async fn start(&self, retry_timeout: &Duration) -> anyhow::Result<()> {
         // Start a vanilla Postgres process used by the storage controller for persistence.
         let pg_data_path = Utf8PathBuf::from_path_buf(self.env.base_data_dir.clone())
@@ -276,20 +256,17 @@ impl StorageController {
             if !status.success() {
                 anyhow::bail!("initdb failed with status {status}");
             }
-        };
 
-        // Write a minimal config file:
-        // - Specify the port, since this is chosen dynamically
-        // - Switch off fsync, since we're running on lightweight test environments and when e.g. scale testing
-        //   the storage controller we don't want a slow local disk to interfere with that.
-        //
-        // NB: it's important that we rewrite this file on each start command so we propagate changes
-        // from `LocalEnv`'s config file (`.neon/config`).
-        tokio::fs::write(
-            &pg_data_path.join("postgresql.conf"),
-            format!("port = {}\nfsync=off\n", self.postgres_port),
-        )
-        .await?;
+            // Write a minimal config file:
+            // - Specify the port, since this is chosen dynamically
+            // - Switch off fsync, since we're running on lightweight test environments and when e.g. scale testing
+            //   the storage controller we don't want a slow local disk to interfere with that.
+            tokio::fs::write(
+                &pg_data_path.join("postgresql.conf"),
+                format!("port = {}\nfsync=off\n", self.postgres_port),
+            )
+            .await?;
+        };
 
         println!("Starting storage controller database...");
         let db_start_args = [
@@ -319,45 +296,16 @@ impl StorageController {
         // Run migrations on every startup, in case something changed.
         let database_url = self.setup_database().await?;
 
-        // We support running a startup SQL script to fiddle with the database before we launch storcon.
-        // This is used by the test suite.
-        let startup_script_path = self
-            .env
-            .base_data_dir
-            .join("storage_controller_db.startup.sql");
-        let startup_script = match tokio::fs::read_to_string(&startup_script_path).await {
-            Ok(script) => {
-                tokio::fs::remove_file(startup_script_path).await?;
-                script
-            }
-            Err(e) => {
-                if e.kind() == std::io::ErrorKind::NotFound {
-                    // always run some startup script so that this code path doesn't bit rot
-                    "BEGIN; COMMIT;".to_string()
-                } else {
-                    anyhow::bail!("Failed to read startup script: {e}")
-                }
-            }
-        };
-        let (mut client, conn) = self.connect_to_database().await?;
-        let conn = tokio::spawn(conn);
-        let tx = client.build_transaction();
-        let tx = tx.start().await?;
-        tx.batch_execute(&startup_script).await?;
-        tx.commit().await?;
-        drop(client);
-        conn.await??;
-
         let mut args = vec![
             "-l",
             &self.listen,
+            "-p",
+            self.path.as_ref(),
             "--dev",
             "--database-url",
             &database_url,
-            "--max-offline-interval",
-            &humantime::Duration::from(self.config.max_offline).to_string(),
-            "--max-warming-up-interval",
-            &humantime::Duration::from(self.config.max_warming_up).to_string(),
+            "--max-unavailable-interval",
+            &humantime::Duration::from(self.config.max_unavailable).to_string(),
         ]
         .into_iter()
         .map(|s| s.to_string())
@@ -383,10 +331,6 @@ impl StorageController {
             args.push(format!("--split-threshold={split_threshold}"))
         }
 
-        if let Some(lag) = self.config.max_secondary_lag_bytes.as_ref() {
-            args.push(format!("--max-secondary-lag-bytes={lag}"))
-        }
-
         args.push(format!(
             "--neon-local-repo-dir={}",
             self.env.base_data_dir.display()
@@ -632,15 +576,6 @@ impl StorageController {
         .await
     }
 
-    pub async fn node_list(&self) -> anyhow::Result<Vec<NodeDescribeResponse>> {
-        self.dispatch::<(), Vec<NodeDescribeResponse>>(
-            Method::GET,
-            "control/v1/node".to_string(),
-            None,
-        )
-        .await
-    }
-
     #[instrument(skip(self))]
     pub async fn ready(&self) -> anyhow::Result<()> {
         self.dispatch::<(), ()>(Method::GET, "ready".to_string(), None)

control_plane/storcon_cli/Cargo.toml

@@ -17,7 +17,6 @@ pageserver_client.workspace = true
 reqwest.workspace = true
 serde.workspace = true
 serde_json = { workspace = true, features = ["raw_value"] }
-storage_controller_client.workspace = true
 thiserror.workspace = true
 tokio.workspace = true
 tracing.workspace = true

control_plane/storcon_cli/src/main.rs

@@ -14,15 +14,15 @@ use pageserver_api::{
     },
     shard::{ShardStripeSize, TenantShardId},
 };
-use pageserver_client::mgmt_api::{self};
+use pageserver_client::mgmt_api::{self, ResponseErrorMessageExt};
 use reqwest::{Method, StatusCode, Url};
+use serde::{de::DeserializeOwned, Serialize};
 use utils::id::{NodeId, TenantId};
 
 use pageserver_api::controller_api::{
     NodeConfigureRequest, NodeRegisterRequest, NodeSchedulingPolicy, PlacementPolicy,
     TenantShardMigrateRequest, TenantShardMigrateResponse,
 };
-use storage_controller_client::control_api::Client;
 
 #[derive(Subcommand, Debug)]
 enum Command {
@@ -249,6 +249,64 @@ impl FromStr for NodeAvailabilityArg {
     }
 }
 
+struct Client {
+    base_url: Url,
+    jwt_token: Option<String>,
+    client: reqwest::Client,
+}
+
+impl Client {
+    fn new(base_url: Url, jwt_token: Option<String>) -> Self {
+        Self {
+            base_url,
+            jwt_token,
+            client: reqwest::ClientBuilder::new()
+                .build()
+                .expect("Failed to construct http client"),
+        }
+    }
+
+    /// Simple HTTP request wrapper for calling into storage controller
+    async fn dispatch<RQ, RS>(
+        &self,
+        method: Method,
+        path: String,
+        body: Option<RQ>,
+    ) -> mgmt_api::Result<RS>
+    where
+        RQ: Serialize + Sized,
+        RS: DeserializeOwned + Sized,
+    {
+        // The configured URL has the /upcall path prefix for pageservers to use: we will strip that out
+        // for general purpose API access.
+        let url = Url::from_str(&format!(
+            "http://{}:{}/{path}",
+            self.base_url.host_str().unwrap(),
+            self.base_url.port().unwrap()
+        ))
+        .unwrap();
+
+        let mut builder = self.client.request(method, url);
+        if let Some(body) = body {
+            builder = builder.json(&body)
+        }
+        if let Some(jwt_token) = &self.jwt_token {
+            builder = builder.header(
+                reqwest::header::AUTHORIZATION,
+                format!("Bearer {jwt_token}"),
+            );
+        }
+
+        let response = builder.send().await.map_err(mgmt_api::Error::ReceiveBody)?;
+        let response = response.error_from_body().await?;
+
+        response
+            .json()
+            .await
+            .map_err(pageserver_client::mgmt_api::Error::ReceiveBody)
+    }
+}
+
 #[tokio::main]
 async fn main() -> anyhow::Result<()> {
     let cli = Cli::parse();

deny.toml

@@ -4,7 +4,6 @@
 # to your expectations and requirements.
 
 # Root options
-[graph]
 targets = [
     { triple = "x86_64-unknown-linux-gnu" },
     { triple = "aarch64-unknown-linux-gnu" },
@@ -13,7 +12,6 @@ targets = [
 ]
 all-features = false
 no-default-features = false
-[output]
 feature-depth = 1
 
 # This section is considered when running `cargo deny check advisories`
@@ -21,16 +19,17 @@ feature-depth = 1
 # https://embarkstudios.github.io/cargo-deny/checks/advisories/cfg.html
 [advisories]
 db-urls = ["https://github.com/rustsec/advisory-db"]
+vulnerability = "deny"
+unmaintained = "warn"
 yanked = "warn"
-
-[[advisories.ignore]]
-id = "RUSTSEC-2023-0071"
-reason = "the marvin attack only affects private key decryption, not public key signature verification"
+notice = "warn"
+ignore = []
 
 # This section is considered when running `cargo deny check licenses`
 # More documentation for the licenses section can be found here:
 # https://embarkstudios.github.io/cargo-deny/checks/licenses/cfg.html
 [licenses]
+unlicensed = "deny"
 allow = [
     "Apache-2.0",
     "Artistic-2.0",
@@ -43,6 +42,10 @@ allow = [
     "OpenSSL",
     "Unicode-DFS-2016",
 ]
+deny = []
+copyleft = "warn"
+allow-osi-fsf-free = "neither"
+default = "deny"
 confidence-threshold = 0.8
 exceptions = [
     # Zlib license has some restrictions if we decide to change sth

docker-compose/compute_wrapper/shell/compute.sh

@@ -33,7 +33,7 @@ echo $result | jq .
 
 generate_id timeline_id
 PARAMS=(
-     -sbf
+     -sb 
      -X POST
      -H "Content-Type: application/json"
      -d "{\"new_timeline_id\": \"${timeline_id}\", \"pg_version\": ${PG_VERSION}}"

docker-compose/docker-compose.yml

@@ -31,14 +31,25 @@ services:
     restart: always
     image: ${REPOSITORY:-neondatabase}/neon:${TAG:-latest}
     environment:
+      - BROKER_ENDPOINT='http://storage_broker:50051'
       - AWS_ACCESS_KEY_ID=minio
       - AWS_SECRET_ACCESS_KEY=password
       #- RUST_BACKTRACE=1
     ports:
        #- 6400:6400  # pg protocol handler
        - 9898:9898 # http endpoints
-    volumes:
-      - ./pageserver_config:/data/.neon/
+    entrypoint:
+      - "/bin/sh"
+      - "-c"
+    command:
+      - "/usr/local/bin/pageserver -D /data/.neon/
+                                   -c \"broker_endpoint=$$BROKER_ENDPOINT\"
+                                   -c \"listen_pg_addr='0.0.0.0:6400'\"
+                                   -c \"listen_http_addr='0.0.0.0:9898'\"
+                                   -c \"remote_storage={endpoint='http://minio:9000',
+                                                        bucket_name='neon',
+                                                        bucket_region='eu-north-1',
+                                                        prefix_in_bucket='/pageserver/'}\""
     depends_on:
       - storage_broker
       - minio_create_buckets

docker-compose/docker_compose_test.sh

@@ -78,7 +78,7 @@ for pg_version in 14 15 16; do
         docker cp $TMPDIR/data $COMPUTE_CONTAINER_NAME:/ext-src/pg_hint_plan-src/
         rm -rf $TMPDIR
         # 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 \
+        if docker exec -e SKIP=rum-src,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

docker-compose/pageserver_config/identity.toml

@@ -1 +0,0 @@
-id=1234

docker-compose/pageserver_config/pageserver.toml

@@ -1,5 +0,0 @@
-broker_endpoint='http://storage_broker:50051'
-pg_distrib_dir='/usr/local/'
-listen_pg_addr='0.0.0.0:6400'
-listen_http_addr='0.0.0.0:9898'
-remote_storage={ endpoint='http://minio:9000', bucket_name='neon', bucket_region='eu-north-1', prefix_in_bucket='/pageserver' }

docker-compose/run-tests.sh

@@ -1,15 +1,15 @@
 #!/bin/bash
 set -x
 
-cd /ext-src || exit 2
+cd /ext-src
 FAILED=
-LIST=$( (echo "${SKIP//","/"\n"}"; ls -d -- *-src) | sort | uniq -u)
+LIST=$((echo ${SKIP} | sed 's/,/\n/g'; ls -d *-src) | sort | uniq -u)
 for d in ${LIST}
 do
-       [ -d "${d}" ] || continue
+       [ -d ${d} ] || continue
     psql -c "select 1" >/dev/null || break
-       USE_PGXS=1 make -C "${d}" installcheck || FAILED="${d} ${FAILED}"
+       make -C ${d} installcheck || FAILED="${d} ${FAILED}"
 done
 [ -z "${FAILED}" ] && exit 0
-echo "${FAILED}"
+echo ${FAILED}
 exit 1

docs/SUMMARY.md

@@ -1,18 +1,13 @@
 # Summary
 
-# Looking for `neon.tech` docs?
-
-This page linkes to a selection of technical content about the open source code in this repository.
-
-Please visit https://neon.tech/docs for documentation about using the Neon service, which is based on the code
-in this repository.
-
-# Architecture
-
 [Introduction]()
 - [Separation of Compute and Storage](./separation-compute-storage.md)
 
+# Architecture
+
 - [Compute]()
+  - [WAL proposer]()
+  - [WAL Backpressure]()
   - [Postgres changes](./core_changes.md)
 
 - [Pageserver](./pageserver.md)
@@ -21,15 +16,33 @@ in this repository.
     - [WAL Redo](./pageserver-walredo.md)
     - [Page cache](./pageserver-pagecache.md)
     - [Storage](./pageserver-storage.md)
+        - [Datadir mapping]()
+        - [Layer files]()
+        - [Branching]()
+        - [Garbage collection]()
+    - [Cloud Storage]()
     - [Processing a GetPage request](./pageserver-processing-getpage.md)
     - [Processing WAL](./pageserver-processing-wal.md)
+	- [Management API]()
+	- [Tenant Rebalancing]()
 
 - [WAL Service](walservice.md)
   - [Consensus protocol](safekeeper-protocol.md)
+  - [Management API]()
+  - [Rebalancing]()
+
+- [Control Plane]()
+
+- [Proxy]()
 
 - [Source view](./sourcetree.md)
   - [docker.md](./docker.md) — Docker images and building pipeline.
   - [Error handling and logging](./error-handling.md)
+  - [Testing]()
+    - [Unit testing]()
+    - [Integration testing]()
+    - [Benchmarks]()
+
 
 - [Glossary](./glossary.md)
 
@@ -45,6 +58,28 @@ in this repository.
 
 # RFCs
 
-Major changes are documented in RFCS:
-- See [RFCs](./rfcs/README.md) for more information
-- view the RFCs at https://github.com/neondatabase/neon/tree/main/docs/rfcs
+- [RFCs](./rfcs/README.md)
+
+- [002-storage](rfcs/002-storage.md)
+- [003-laptop-cli](rfcs/003-laptop-cli.md)
+- [004-durability](rfcs/004-durability.md)
+- [005-zenith_local](rfcs/005-zenith_local.md)
+- [006-laptop-cli-v2-CLI](rfcs/006-laptop-cli-v2-CLI.md)
+- [006-laptop-cli-v2-repository-structure](rfcs/006-laptop-cli-v2-repository-structure.md)
+- [007-serverless-on-laptop](rfcs/007-serverless-on-laptop.md)
+- [008-push-pull](rfcs/008-push-pull.md)
+- [009-snapshot-first-storage-cli](rfcs/009-snapshot-first-storage-cli.md)
+- [009-snapshot-first-storage](rfcs/009-snapshot-first-storage.md)
+- [009-snapshot-first-storage-pitr](rfcs/009-snapshot-first-storage-pitr.md)
+- [010-storage_details](rfcs/010-storage_details.md)
+- [011-retention-policy](rfcs/011-retention-policy.md)
+- [012-background-tasks](rfcs/012-background-tasks.md)
+- [013-term-history](rfcs/013-term-history.md)
+- [014-safekeepers-gossip](rfcs/014-safekeepers-gossip.md)
+- [014-storage-lsm](rfcs/014-storage-lsm.md)
+- [015-storage-messaging](rfcs/015-storage-messaging.md)
+- [016-connection-routing](rfcs/016-connection-routing.md)
+- [017-timeline-data-management](rfcs/017-timeline-data-management.md)
+- [018-storage-messaging-2](rfcs/018-storage-messaging-2.md)
+- [019-tenant-timeline-lifecycles](rfcs/019-tenant-timeline-lifecycles.md)
+- [cluster-size-limits](rfcs/cluster-size-limits.md)

docs/rfcs/034-ancestor-deletion.md

@@ -1,252 +0,0 @@
-# Ancestor Timeline Deletion
-
-Created on: 2024-02-23
-
-Author: John Spray
-
-# Summary
-
-When a tenant creates a new timeline that they will treat as their 'main' history,
-it is awkward to permanently retain an 'old main' timeline as its ancestor. Currently
-this is necessary because it is forbidden to delete a timeline which has descendents.
-
-A new pageserver API is proposed to 'adopt' data from a parent timeline into
-one of its children, such that the link between ancestor and child can be severed,
-leaving the parent in a state where it may then be deleted.
-
-# Motivation
-
-Retaining parent timelines currently has two costs:
-
-- Cognitive load on users, who have to remember which is the "real" main timeline.
-- Storage capacity cost, as the parent timeline will retain layers up to the
-  child's timeline point, even if the child fully covers its keyspace with image
-  layers and will never actually read from the parent.
-
-# Solution
-
-A new pageserver API `PUT /v1/tenant/:tenant_id/timeline/:timeline_id/detach_ancestor`
-will be added. The `timeline_id` in this URL is that of the _child_ timeline that we
-wish to detach from its parent.
-
-On success, this API will leave the following state:
-
-- The detached child timeline will no longer have an ancestor, and will contain all
-  the data needed to service reads without recursing into an ancestor.
-- Any other children of the parent whose timeline points were at a lower LSN than
-  the detached child timeline will be modified to have the child timeline as their
-  new parent.
-- The parent timeline will still exist, but the child will no longer have it as an
-  ancestor. If this was the last timeline that depended on the parent, then the
-  parent will become deletable.
-
-This API's implementation will consist of a series of retryable steps, such that
-on failures/timeout it can safely be called again to reach the target state.
-
-## Example
-
-### Before
-
-The user has "rolled back" their project to LSN X, resulting in a "new main"
-timeline. The parent "old main" timeline still exists, and they would like
-to clean it up.
-
-They have two other timelines A and B. A is from before the rollback point,
-and B is from after the rollback point.
-
-```
-----"old main" timeline-------X-------------------------------------------->
-                |             |                         |
-                |-> child A   |                         |
-                              |-> "new main" timeline   |
-                                                        -> child B
-
-```
-
-### After calling detach ancestor API
-
-The "new main" timeline is no longer dependent on old main, and neither
-is child A, because it had a branch point before X.
-
-The user may now choose to delete child B and "old main" to get to
-a pristine state. Child B is likely to be unwanted since the user
-chose to roll back to X, and it branches from after X. However, we
-don't assume this in the API; it is up to the user to delete it.
-
-```
-|----"old main" timeline---------------------------------------------------->
-                                                         |
-                                                         |
-                                                         |
-                                                         -> child B
-
-|----"new main" timeline--------->
-                 |
-                 |-> child A
-
-
-```
-
-### After removing timelines
-
-We end up with a totally clean state that leaves no trace that a rollback
-ever happened: there is only one root timeline.
-
-```
-| ----"new main" timeline----------->
-                |
-                |-> child A
-
-
-```
-
-## Caveats
-
-Important things for API users to bear in mind:
-
-- this API does not delete the parent timeline: you must still do that explicitly.
-- if there are other child timelines ahead of the branch point of the detached
-  child, the parent won't be deletable: you must either delete or detach those
-  children.
-- do _not_ simply loop over all children and detach them all: this can have an
-  extremely high storage cost. The detach ancestor API is intended for use on a single
-  timeline to make it the new "main".
-- The detach ancestor API should also not be
-  exposed directly to the user as button/API, because they might decide
-  to click it for all the children and thereby generate many copies of the
-  parent's data -- the detach ancestor API should be used as part
-  of a high level "clean up after rollback" feature.
-
-## `detach_ancestor` API implementation
-
-Terms used in the following sections:
-
-- "the child": the timeline whose ID is specified in the detach ancestor API URL, also
-  called "new main" in the example.
-- "the parent": the parent of "the child". Also called "old main" in the example.
-- "the branch point" the ancestor_lsn of "the child"
-
-### Phase 1: write out adopted layers to S3
-
-The child will "adopt" layers from the parent, such that its end state contains
-all the parent's history as well as its own.
-
-For all layers in the parent's layer map whose high LSN is below the branch
-point, issue S3 CopyObject requests to duplicate them into the child timeline's
-prefix. Do not add them to the child's layer map yet.
-
-For delta layers in the parent's layer map which straddle the branch point, read them
-and write out only content up to the branch point into new layer objects.
-
-This is a long running operation if the parent has many layers: it should be
-implemented in a way that resumes rather than restarting from scratch, if the API
-times out and is called again.
-
-As an optimization, if there are no other timelines that will be adopted into
-the child, _and_ the child's image layers already full cover the branch LSN,
-then we may skip adopting layers.
-
-### Phase 2: update the child's index
-
-Having written out all needed layers in phase 1, atomically link them all
-into the child's IndexPart and upload to S3. This may be done while the
-child Timeline is still running.
-
-### Phase 3: modify timelines ancestry
-
-Modify the child's ancestor to None, and upload its IndexPart to persist the change.
-
-For all timelines which have the same parent as the child, and have a branch
-point lower than our branch point, switch their ancestor_timeline to the child,
-and upload their IndexPart to persist the change.
-
-## Alternatives considered
-
-### Generate full image layer on child, rather than adopting parent deltas
-
-This would work for the case of a single child, but would prevent re-targeting
-other timelines that depended on the parent. If we detached many children this
-way, the storage cost would become prohibitive (consider a 1TB database with
-100 child timelines: it would cost 100TiB if they all generated their own image layers).
-
-### Don't rewrite anything: just fake it in the API
-
-We could add a layer of indirection that let a child "pretend" that it had no
-ancestor, when in reality it still had the parent. The pageserver API could
-accept deletion of ancestor timelines, and just update child metadata to make
-them look like they have no ancestor.
-
-This would not achieve the desired reduction in storage cost, and may well be more
-complex to maintain than simply implementing the API described in this RFC.
-
-### Avoid copying objects: enable child index to use parent layers directly
-
-We could teach IndexPart to store a TimelineId for each layer, such that a child
-timeline could reference a parent's layers directly, rather than copying them
-into the child's prefix.
-
-This would impose a cost for the normal case of indices that only target the
-timeline's own layers, add complexity, and break the useful simplifying
-invariant that timelines "own" their own path. If child timelines were
-referencing layers from the parent, we would have to ensure that the parent
-never runs GC/compaction again, which would make the API less flexible (the
-proposal in this RFC enables deletion of the parent but doesn't require it.)
-
-## Performance
-
-### Adopting layers
-
-- CopyObject is a relatively cheap operation, but we may need to issue tens of thousands
-  of such requests: this can take up to tens of seconds and will compete for RemoteStorage
-  semaphore units with other activity on the pageserver.
-- If we are running on storage backend that doesn't implement CopyObject, then
-  this part will be much more expensive as we would stream all layer content
-  through the pageserver. This is no different to issuing a lot
-  of reads to a timeline that does not have a warm local cache: it will move
-  a lot of gigabytes, but that shouldn't break anything.
-- Generating truncated layers for delta that straddle the branch point will
-  require streaming read/write of all the layers in question.
-
-### Updating timeline ancestry
-
-The simplest way to update timeline ancestry will probably be to stop and start
-all the Timeline objects: this is preferable to the complexity of making their
-ancestry mutable at runtime.
-
-There will be a corresponding "stutter" in the availability of the timelines,
-of the order 10-100ms, which is the time taken to upload their IndexPart, and
-restart the Timeline.
-
-# Interaction with other features
-
-## Concurrent timeline creation
-
-If new historic timelines are created using the parent as an ancestor while the
-detach ancestor API is running, they will not be re-parented to the child. This
-doesn't break anything, but it leaves the parent in a state where it might not
-be possible to delete it.
-
-Since timeline creations are an explicit user action, this is not something we need to
-worry about as the storage layer: a user who wants to delete their parent timeline will not create
-new children, and if they do, they can choose to delete those children to
-enable deleting the parent.
-
-For the least surprise to the user, before starting the detach ancestor branch
-operation, the control plane should wait until all branches are created and not
-allow any branches to be created before the branch point on the ancestor branch
-while the operation is ongoing.
-
-## WAL based disaster recovery
-
-WAL based disaster recovery currently supports only restoring of the main
-branch. Enabling WAL based disaster recovery in the future requires that we
-keep a record which timeline generated the WAL and at which LSN was a parent
-detached. Keep a list of timeline ids and the LSN in which they were detached in
-the `index_part.json`. Limit the size of the list to 100 first entries, after
-which the WAL disaster recovery will not be possible.
-
-## Sharded tenants
-
-For sharded tenants, calls to the detach ancestor API will pass through the storage
-controller, which will handle them the same as timeline creations: invoke first
-on shard zero, and then on all the other shards.

docs/storage_controller.md

@@ -44,7 +44,7 @@ If you need to modify the database schema, here’s how to create a migration:
 - Use `diesel migration generate <name>` to create a new migration
 - Populate the SQL files in the `migrations/` subdirectory
 - Use `DATABASE_URL=... diesel migration run` to apply the migration you just wrote: this will update the `[schema.rs](http://schema.rs)` file automatically.
-  - This requires a running database: the easiest way to do that is to just run `cargo neon init ; cargo neon start`, which will leave a database available at `postgresql://localhost:1235/storage_controller`
+  - This requires a running database: the easiest way to do that is to just run `cargo neon init ; cargo neon start`, which will leave a database available at `postgresql://localhost:1235/attachment_service`
 - Commit the migration files and the changes to schema.rs
 - If you need to iterate, you can rewind migrations with `diesel migration revert -a` and then `diesel migration run` again.
 - The migrations are build into the storage controller binary, and automatically run at startup after it is deployed, so once you’ve committed a migration no further steps are needed.

docs/synthetic-size.md

@@ -21,9 +21,9 @@ implementation where we keep more data than we would need to, do not
 change the synthetic size or incur any costs to the user.
 
 The synthetic size is calculated for the whole project. It is not
-straightforward to attribute size to individual branches. See [What is
-the size of an individual branch?](#what-is-the-size-of-an-individual-branch)
-for a discussion of those difficulties.
+straightforward to attribute size to individual branches. See "What is
+the size of an individual branch?" for discussion on those
+difficulties.
 
 The synthetic size is designed to:
 
@@ -40,9 +40,8 @@ The synthetic size is designed to:
 - logical size is the size of a branch *at a given point in
   time*. It's the total size of all tables in all databases, as you
   see with "\l+" in psql for example, plus the Postgres SLRUs and some
-  small amount of metadata. Note that currently, Neon does not include
-  the SLRUs and metadata in the logical size. Refer to the comment in
-  [`get_current_logical_size_non_incremental()`](/pageserver/src/pgdatadir_mapping.rs#L813-L814).
+  small amount of metadata. NOTE that currently, Neon does not include
+  the SLRUs and metadata in the logical size. See comment to `get_current_logical_size_non_incremental()`.
 
 - a "point in time" is defined as an LSN value. You can convert a
   timestamp to an LSN, but the storage internally works with LSNs.

libs/pageserver_api/src/controller_api.rs

@@ -1,6 +1,4 @@
-use std::collections::HashSet;
 use std::str::FromStr;
-use std::time::{Duration, Instant};
 
 /// Request/response types for the storage controller
 /// API (`/control/v1` prefix).  Implemented by the server
@@ -89,7 +87,7 @@ pub struct TenantLocateResponse {
     pub shard_params: ShardParameters,
 }
 
-#[derive(Serialize, Deserialize, Debug)]
+#[derive(Serialize, Deserialize)]
 pub struct TenantDescribeResponse {
     pub tenant_id: TenantId,
     pub shards: Vec<TenantDescribeResponseShard>,
@@ -112,7 +110,7 @@ pub struct NodeDescribeResponse {
     pub listen_pg_port: u16,
 }
 
-#[derive(Serialize, Deserialize, Debug)]
+#[derive(Serialize, Deserialize)]
 pub struct TenantDescribeResponseShard {
     pub tenant_shard_id: TenantShardId,
 
@@ -152,16 +150,11 @@ impl UtilizationScore {
     }
 }
 
-#[derive(Serialize, Clone, Copy, Debug)]
+#[derive(Serialize, Deserialize, Clone, Copy, Debug)]
 #[serde(into = "NodeAvailabilityWrapper")]
 pub enum NodeAvailability {
     // Normal, happy state
     Active(UtilizationScore),
-    // Node is warming up, but we expect it to become available soon. Covers
-    // the time span between the re-attach response being composed on the storage controller
-    // and the first successful heartbeat after the processing of the re-attach response
-    // finishes on the pageserver.
-    WarmingUp(Instant),
     // Offline: Tenants shouldn't try to attach here, but they may assume that their
     // secondary locations on this node still exist.  Newly added nodes are in this
     // state until we successfully contact them.
@@ -171,10 +164,7 @@ pub enum NodeAvailability {
 impl PartialEq for NodeAvailability {
     fn eq(&self, other: &Self) -> bool {
         use NodeAvailability::*;
-        matches!(
-            (self, other),
-            (Active(_), Active(_)) | (Offline, Offline) | (WarmingUp(_), WarmingUp(_))
-        )
+        matches!((self, other), (Active(_), Active(_)) | (Offline, Offline))
     }
 }
 
@@ -186,7 +176,6 @@ impl Eq for NodeAvailability {}
 #[derive(Serialize, Deserialize, Clone, Copy, Debug)]
 pub enum NodeAvailabilityWrapper {
     Active,
-    WarmingUp,
     Offline,
 }
 
@@ -196,7 +185,6 @@ impl From<NodeAvailabilityWrapper> for NodeAvailability {
             // Assume the worst utilisation score to begin with. It will later be updated by
             // the heartbeats.
             NodeAvailabilityWrapper::Active => NodeAvailability::Active(UtilizationScore::worst()),
-            NodeAvailabilityWrapper::WarmingUp => NodeAvailability::WarmingUp(Instant::now()),
             NodeAvailabilityWrapper::Offline => NodeAvailability::Offline,
         }
     }
@@ -206,7 +194,6 @@ impl From<NodeAvailability> for NodeAvailabilityWrapper {
     fn from(val: NodeAvailability) -> Self {
         match val {
             NodeAvailability::Active(_) => NodeAvailabilityWrapper::Active,
-            NodeAvailability::WarmingUp(_) => NodeAvailabilityWrapper::WarmingUp,
             NodeAvailability::Offline => NodeAvailabilityWrapper::Offline,
         }
     }
@@ -295,42 +282,6 @@ pub enum PlacementPolicy {
 #[derive(Serialize, Deserialize, Debug)]
 pub struct TenantShardMigrateResponse {}
 
-/// Metadata health record posted from scrubber.
-#[derive(Serialize, Deserialize, Debug)]
-pub struct MetadataHealthRecord {
-    pub tenant_shard_id: TenantShardId,
-    pub healthy: bool,
-    pub last_scrubbed_at: chrono::DateTime<chrono::Utc>,
-}
-
-#[derive(Serialize, Deserialize, Debug)]
-pub struct MetadataHealthUpdateRequest {
-    pub healthy_tenant_shards: HashSet<TenantShardId>,
-    pub unhealthy_tenant_shards: HashSet<TenantShardId>,
-}
-
-#[derive(Serialize, Deserialize, Debug)]
-pub struct MetadataHealthUpdateResponse {}
-
-#[derive(Serialize, Deserialize, Debug)]
-
-pub struct MetadataHealthListUnhealthyResponse {
-    pub unhealthy_tenant_shards: Vec<TenantShardId>,
-}
-
-#[derive(Serialize, Deserialize, Debug)]
-
-pub struct MetadataHealthListOutdatedRequest {
-    #[serde(with = "humantime_serde")]
-    pub not_scrubbed_for: Duration,
-}
-
-#[derive(Serialize, Deserialize, Debug)]
-
-pub struct MetadataHealthListOutdatedResponse {
-    pub health_records: Vec<MetadataHealthRecord>,
-}
-
 #[cfg(test)]
 mod test {
     use super::*;

libs/pageserver_api/src/key.rs

@@ -107,10 +107,7 @@ impl Key {
     /// As long as Neon does not support tablespace (because of lack of access to local file system),
     /// we can assume that only some predefined namespace OIDs are used which can fit in u16
     pub fn to_i128(&self) -> i128 {
-        assert!(
-            self.field2 <= 0xFFFF || self.field2 == 0xFFFFFFFF || self.field2 == 0x22222222,
-            "invalid key: {self}",
-        );
+        assert!(self.field2 <= 0xFFFF || self.field2 == 0xFFFFFFFF || self.field2 == 0x22222222);
         (((self.field1 & 0x7F) as i128) << 120)
             | (((self.field2 & 0xFFFF) as i128) << 104)
             | ((self.field3 as i128) << 72)

libs/pageserver_api/src/models.rs

@@ -5,6 +5,7 @@ pub mod utilization;
 pub use utilization::PageserverUtilization;
 
 use std::{
+    borrow::Cow,
     collections::HashMap,
     io::{BufRead, Read},
     num::{NonZeroU64, NonZeroUsize},
@@ -19,6 +20,7 @@ use serde::{Deserialize, Serialize};
 use serde_with::serde_as;
 use utils::{
     completion,
+    history_buffer::HistoryBufferWithDropCounter,
     id::{NodeId, TenantId, TimelineId},
     lsn::Lsn,
     serde_system_time,
@@ -637,13 +639,6 @@ pub struct TenantInfo {
     pub current_physical_size: Option<u64>, // physical size is only included in `tenant_status` endpoint
     pub attachment_status: TenantAttachmentStatus,
     pub generation: u32,
-
-    /// Opaque explanation if gc is being blocked.
-    ///
-    /// Only looked up for the individual tenant detail, not the listing. This is purely for
-    /// debugging, not included in openapi.
-    #[serde(skip_serializing_if = "Option::is_none")]
-    pub gc_blocking: Option<String>,
 }
 
 #[derive(Serialize, Deserialize, Clone)]
@@ -656,17 +651,6 @@ pub struct TenantDetails {
     pub timelines: Vec<TimelineId>,
 }
 
-#[derive(Serialize, Deserialize, PartialEq, Eq, Clone, Copy, Debug)]
-pub enum TimelineArchivalState {
-    Archived,
-    Unarchived,
-}
-
-#[derive(Serialize, Deserialize, PartialEq, Eq, Clone)]
-pub struct TimelineArchivalConfigRequest {
-    pub state: TimelineArchivalState,
-}
-
 /// This represents the output of the "timeline_detail" and "timeline_list" API calls.
 #[derive(Debug, Serialize, Deserialize, Clone)]
 pub struct TimelineInfo {
@@ -731,7 +715,58 @@ pub struct LayerMapInfo {
     pub historic_layers: Vec<HistoricLayerInfo>,
 }
 
-/// The residence status of a layer
+#[derive(Debug, Hash, PartialEq, Eq, Clone, Copy, Serialize, Deserialize, enum_map::Enum)]
+#[repr(usize)]
+pub enum LayerAccessKind {
+    GetValueReconstructData,
+    Iter,
+    KeyIter,
+    Dump,
+}
+
+#[derive(Debug, Clone, Serialize, Deserialize)]
+pub struct LayerAccessStatFullDetails {
+    pub when_millis_since_epoch: u64,
+    pub task_kind: Cow<'static, str>,
+    pub access_kind: LayerAccessKind,
+}
+
+/// An event that impacts the layer's residence status.
+#[serde_as]
+#[derive(Debug, Clone, Serialize, Deserialize)]
+pub struct LayerResidenceEvent {
+    /// The time when the event occurred.
+    /// NB: this timestamp is captured while the residence status changes.
+    /// So, it might be behind/ahead of the actual residence change by a short amount of time.
+    ///
+    #[serde(rename = "timestamp_millis_since_epoch")]
+    #[serde_as(as = "serde_with::TimestampMilliSeconds")]
+    pub timestamp: SystemTime,
+    /// The new residence status of the layer.
+    pub status: LayerResidenceStatus,
+    /// The reason why we had to record this event.
+    pub reason: LayerResidenceEventReason,
+}
+
+/// The reason for recording a given [`LayerResidenceEvent`].
+#[derive(Debug, Clone, Copy, Serialize, Deserialize)]
+pub enum LayerResidenceEventReason {
+    /// The layer map is being populated, e.g. during timeline load or attach.
+    /// This includes [`RemoteLayer`] objects created in [`reconcile_with_remote`].
+    /// We need to record such events because there is no persistent storage for the events.
+    ///
+    // https://github.com/rust-lang/rust/issues/74481
+    /// [`RemoteLayer`]: ../../tenant/storage_layer/struct.RemoteLayer.html
+    /// [`reconcile_with_remote`]: ../../tenant/struct.Timeline.html#method.reconcile_with_remote
+    LayerLoad,
+    /// We just created the layer (e.g., freeze_and_flush or compaction).
+    /// Such layers are always [`LayerResidenceStatus::Resident`].
+    LayerCreate,
+    /// We on-demand downloaded or evicted the given layer.
+    ResidenceChange,
+}
+
+/// The residence status of the layer, after the given [`LayerResidenceEvent`].
 #[derive(Debug, Clone, Copy, Serialize, Deserialize)]
 pub enum LayerResidenceStatus {
     /// Residence status for a layer file that exists locally.
@@ -741,16 +776,23 @@ pub enum LayerResidenceStatus {
     Evicted,
 }
 
-#[serde_as]
+impl LayerResidenceEvent {
+    pub fn new(status: LayerResidenceStatus, reason: LayerResidenceEventReason) -> Self {
+        Self {
+            status,
+            reason,
+            timestamp: SystemTime::now(),
+        }
+    }
+}
+
 #[derive(Debug, Clone, Serialize, Deserialize)]
 pub struct LayerAccessStats {
-    #[serde_as(as = "serde_with::TimestampMilliSeconds")]
-    pub access_time: SystemTime,
-
-    #[serde_as(as = "serde_with::TimestampMilliSeconds")]
-    pub residence_time: SystemTime,
-
-    pub visible: bool,
+    pub access_count_by_access_kind: HashMap<LayerAccessKind, u64>,
+    pub task_kind_access_flag: Vec<Cow<'static, str>>,
+    pub first: Option<LayerAccessStatFullDetails>,
+    pub accesses_history: HistoryBufferWithDropCounter<LayerAccessStatFullDetails, 16>,
+    pub residence_events_history: HistoryBufferWithDropCounter<LayerResidenceEvent, 16>,
 }
 
 #[derive(Debug, Clone, Serialize, Deserialize)]
@@ -947,8 +989,6 @@ pub struct TopTenantShardsResponse {
 }
 
 pub mod virtual_file {
-    use std::path::PathBuf;
-
     #[derive(
         Copy,
         Clone,
@@ -967,53 +1007,6 @@ pub mod virtual_file {
         #[cfg(target_os = "linux")]
         TokioEpollUring,
     }
-
-    /// Direct IO modes for a pageserver.
-    #[derive(Debug, PartialEq, Eq, Clone, serde::Deserialize, serde::Serialize, Default)]
-    #[serde(tag = "mode", rename_all = "kebab-case", deny_unknown_fields)]
-    pub enum DirectIoMode {
-        /// Direct IO disabled (uses usual buffered IO).
-        #[default]
-        Disabled,
-        /// Direct IO disabled (performs checks and perf simulations).
-        Evaluate {
-            /// Alignment check level
-            alignment_check: DirectIoAlignmentCheckLevel,
-            /// Latency padded for performance simulation.
-            latency_padding: DirectIoLatencyPadding,
-        },
-        /// Direct IO enabled.
-        Enabled {
-            /// Actions to perform on alignment error.
-            on_alignment_error: DirectIoOnAlignmentErrorAction,
-        },
-    }
-
-    #[derive(Debug, PartialEq, Eq, Clone, serde::Deserialize, serde::Serialize, Default)]
-    #[serde(rename_all = "kebab-case")]
-    pub enum DirectIoAlignmentCheckLevel {
-        #[default]
-        Error,
-        Log,
-        None,
-    }
-
-    #[derive(Debug, PartialEq, Eq, Clone, serde::Deserialize, serde::Serialize, Default)]
-    #[serde(rename_all = "kebab-case")]
-    pub enum DirectIoOnAlignmentErrorAction {
-        Error,
-        #[default]
-        FallbackToBuffered,
-    }
-
-    #[derive(Debug, PartialEq, Eq, Clone, serde::Deserialize, serde::Serialize, Default)]
-    #[serde(tag = "type", rename_all = "kebab-case")]
-    pub enum DirectIoLatencyPadding {
-        /// Pad virtual file operations with IO to a fake file.
-        FakeFileRW { path: PathBuf },
-        #[default]
-        None,
-    }
 }
 
 // Wrapped in libpq CopyData
@@ -1483,7 +1476,6 @@ mod tests {
             current_physical_size: Some(42),
             attachment_status: TenantAttachmentStatus::Attached,
             generation: 1,
-            gc_blocking: None,
         };
         let expected_active = json!({
             "id": original_active.id.to_string(),
@@ -1506,7 +1498,6 @@ mod tests {
             current_physical_size: Some(42),
             attachment_status: TenantAttachmentStatus::Attached,
             generation: 1,
-            gc_blocking: None,
         };
         let expected_broken = json!({
             "id": original_broken.id.to_string(),

libs/pageserver_api/src/models/detach_ancestor.rs

@@ -1,8 +1,6 @@
-use std::collections::HashSet;
-
 use utils::id::TimelineId;
 
-#[derive(Debug, Default, PartialEq, serde::Serialize, serde::Deserialize)]
+#[derive(Default, serde::Serialize)]
 pub struct AncestorDetached {
-    pub reparented_timelines: HashSet<TimelineId>,
+    pub reparented_timelines: Vec<TimelineId>,
 }

libs/postgres_connection/src/lib.rs

@@ -144,20 +144,7 @@ impl PgConnectionConfig {
             // implement and this function is hardly a bottleneck. The function is only called around
             // establishing a new connection.
             #[allow(unstable_name_collisions)]
-            config.options(
-                &self
-                    .options
-                    .iter()
-                    .map(|s| {
-                        if s.contains(['\\', ' ']) {
-                            Cow::Owned(s.replace('\\', "\\\\").replace(' ', "\\ "))
-                        } else {
-                            Cow::Borrowed(s.as_str())
-                        }
-                    })
-                    .intersperse(Cow::Borrowed(" ")) // TODO: use impl from std once it's stabilized
-                    .collect::<String>(),
-            );
+            config.options(&encode_options(&self.options));
         }
         config
     }
@@ -178,6 +165,21 @@ impl PgConnectionConfig {
     }
 }
 
+#[allow(unstable_name_collisions)]
+fn encode_options(options: &[String]) -> String {
+    options
+        .iter()
+        .map(|s| {
+            if s.contains(['\\', ' ']) {
+                Cow::Owned(s.replace('\\', "\\\\").replace(' ', "\\ "))
+            } else {
+                Cow::Borrowed(s.as_str())
+            }
+        })
+        .intersperse(Cow::Borrowed(" ")) // TODO: use impl from std once it's stabilized
+        .collect::<String>()
+}
+
 impl fmt::Display for PgConnectionConfig {
     fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
         // The password is intentionally hidden and not part of this display string.
@@ -206,7 +208,7 @@ impl fmt::Debug for PgConnectionConfig {
 
 #[cfg(test)]
 mod tests_pg_connection_config {
-    use crate::PgConnectionConfig;
+    use crate::{encode_options, PgConnectionConfig};
     use once_cell::sync::Lazy;
     use url::Host;
 
@@ -255,18 +257,12 @@ mod tests_pg_connection_config {
 
     #[test]
     fn test_with_options() {
-        let cfg = PgConnectionConfig::new_host_port(STUB_HOST.clone(), 123).extend_options([
-            "hello",
-            "world",
-            "with space",
-            "and \\ backslashes",
+        let options = encode_options(&[
+            "hello".to_owned(),
+            "world".to_owned(),
+            "with space".to_owned(),
+            "and \\ backslashes".to_owned(),
         ]);
-        assert_eq!(cfg.host(), &*STUB_HOST);
-        assert_eq!(cfg.port(), 123);
-        assert_eq!(cfg.raw_address(), "stub.host.example:123");
-        assert_eq!(
-            cfg.to_tokio_postgres_config().get_options(),
-            Some("hello world with\\ space and\\ \\\\\\ backslashes")
-        );
+        assert_eq!(options, "hello world with\\ space and\\ \\\\\\ backslashes");
     }
 }

libs/postgres_ffi/src/controlfile_utils.rs

@@ -29,7 +29,7 @@ use anyhow::{bail, Result};
 use bytes::{Bytes, BytesMut};
 
 /// Equivalent to sizeof(ControlFileData) in C
-const SIZEOF_CONTROLDATA: usize = size_of::<ControlFileData>();
+const SIZEOF_CONTROLDATA: usize = std::mem::size_of::<ControlFileData>();
 
 impl ControlFileData {
     /// Compute the offset of the `crc` field within the `ControlFileData` struct.

libs/postgres_ffi/src/pg_constants.rs

@@ -31,7 +31,7 @@ pub const SMGR_TRUNCATE_FSM: u32 = 0x0004;
 //
 
 // Assumes 8 byte alignment
-const SIZEOF_PAGE_HEADER_DATA: usize = size_of::<PageHeaderData>();
+const SIZEOF_PAGE_HEADER_DATA: usize = std::mem::size_of::<PageHeaderData>();
 pub const MAXALIGN_SIZE_OF_PAGE_HEADER_DATA: usize = (SIZEOF_PAGE_HEADER_DATA + 7) & !7;
 
 //
@@ -191,7 +191,7 @@ pub const XLR_RMGR_INFO_MASK: u8 = 0xF0;
 pub const XLOG_TBLSPC_CREATE: u8 = 0x00;
 pub const XLOG_TBLSPC_DROP: u8 = 0x10;
 
-pub const SIZEOF_XLOGRECORD: u32 = size_of::<XLogRecord>() as u32;
+pub const SIZEOF_XLOGRECORD: u32 = std::mem::size_of::<XLogRecord>() as u32;
 
 //
 // from xlogrecord.h

libs/postgres_ffi/src/xlog_utils.rs

@@ -42,9 +42,9 @@ pub const XLP_FIRST_IS_CONTRECORD: u16 = 0x0001;
 pub const XLP_REM_LEN_OFFS: usize = 2 + 2 + 4 + 8;
 pub const XLOG_RECORD_CRC_OFFS: usize = 4 + 4 + 8 + 1 + 1 + 2;
 
-pub const XLOG_SIZE_OF_XLOG_SHORT_PHD: usize = size_of::<XLogPageHeaderData>();
-pub const XLOG_SIZE_OF_XLOG_LONG_PHD: usize = size_of::<XLogLongPageHeaderData>();
-pub const XLOG_SIZE_OF_XLOG_RECORD: usize = size_of::<XLogRecord>();
+pub const XLOG_SIZE_OF_XLOG_SHORT_PHD: usize = std::mem::size_of::<XLogPageHeaderData>();
+pub const XLOG_SIZE_OF_XLOG_LONG_PHD: usize = std::mem::size_of::<XLogLongPageHeaderData>();
+pub const XLOG_SIZE_OF_XLOG_RECORD: usize = std::mem::size_of::<XLogRecord>();
 #[allow(clippy::identity_op)]
 pub const SIZE_OF_XLOG_RECORD_DATA_HEADER_SHORT: usize = 1 * 2;
 
@@ -311,7 +311,7 @@ impl XLogLongPageHeaderData {
     }
 }
 
-pub const SIZEOF_CHECKPOINT: usize = size_of::<CheckPoint>();
+pub const SIZEOF_CHECKPOINT: usize = std::mem::size_of::<CheckPoint>();
 
 impl CheckPoint {
     pub fn encode(&self) -> Result<Bytes, SerializeError> {

libs/postgres_ffi/wal_craft/src/xlog_utils_test.rs

@@ -178,7 +178,7 @@ pub fn test_find_end_of_wal_last_crossing_segment() {
 /// currently 1024.
 #[test]
 pub fn test_update_next_xid() {
-    let checkpoint_buf = [0u8; size_of::<CheckPoint>()];
+    let checkpoint_buf = [0u8; std::mem::size_of::<CheckPoint>()];
     let mut checkpoint = CheckPoint::decode(&checkpoint_buf).unwrap();
 
     checkpoint.nextXid = FullTransactionId { value: 10 };
@@ -204,7 +204,7 @@ pub fn test_update_next_xid() {
 
 #[test]
 pub fn test_update_next_multixid() {
-    let checkpoint_buf = [0u8; size_of::<CheckPoint>()];
+    let checkpoint_buf = [0u8; std::mem::size_of::<CheckPoint>()];
     let mut checkpoint = CheckPoint::decode(&checkpoint_buf).unwrap();
 
     // simple case

libs/remote_storage/Cargo.toml

@@ -7,7 +7,6 @@ license.workspace = true
 [dependencies]
 anyhow.workspace = true
 async-trait.workspace = true
-async-stream.workspace = true
 once_cell.workspace = true
 aws-smithy-async.workspace = true
 aws-smithy-types.workspace = true

libs/remote_storage/src/azure_blob.rs

@@ -15,7 +15,7 @@ use std::time::SystemTime;
 use super::REMOTE_STORAGE_PREFIX_SEPARATOR;
 use anyhow::Result;
 use azure_core::request_options::{MaxResults, Metadata, Range};
-use azure_core::{Continuable, RetryOptions};
+use azure_core::RetryOptions;
 use azure_identity::DefaultAzureCredential;
 use azure_storage::StorageCredentials;
 use azure_storage_blobs::blob::CopyStatus;
@@ -33,7 +33,6 @@ use tracing::debug;
 use utils::backoff;
 
 use crate::metrics::{start_measuring_requests, AttemptOutcome, RequestKind};
-use crate::ListingObject;
 use crate::{
     config::AzureConfig, error::Cancelled, ConcurrencyLimiter, Download, DownloadError, Listing,
     ListingMode, RemotePath, RemoteStorage, StorageMetadata, TimeTravelError, TimeoutOrCancel,
@@ -41,7 +40,6 @@ use crate::{
 
 pub struct AzureBlobStorage {
     client: ContainerClient,
-    container_name: String,
     prefix_in_container: Option<String>,
     max_keys_per_list_response: Option<NonZeroU32>,
     concurrency_limiter: ConcurrencyLimiter,
@@ -87,7 +85,6 @@ impl AzureBlobStorage {
 
         Ok(AzureBlobStorage {
             client,
-            container_name: azure_config.container_name.to_owned(),
             prefix_in_container: azure_config.prefix_in_container.to_owned(),
             max_keys_per_list_response,
             concurrency_limiter: ConcurrencyLimiter::new(azure_config.concurrency_limit.get()),
@@ -241,10 +238,6 @@ impl AzureBlobStorage {
             _ = cancel.cancelled() => Err(Cancelled),
         }
     }
-
-    pub fn container_name(&self) -> &str {
-        &self.container_name
-    }
 }
 
 fn to_azure_metadata(metadata: StorageMetadata) -> Metadata {
@@ -268,30 +261,30 @@ fn to_download_error(error: azure_core::Error) -> DownloadError {
 }
 
 impl RemoteStorage for AzureBlobStorage {
-    fn list_streaming(
+    async fn list(
         &self,
         prefix: Option<&RemotePath>,
         mode: ListingMode,
         max_keys: Option<NonZeroU32>,
         cancel: &CancellationToken,
-    ) -> impl Stream<Item = Result<Listing, DownloadError>> {
-        // get the passed prefix or if it is not set use prefix_in_bucket value
-        let list_prefix = prefix
-            .map(|p| self.relative_path_to_name(p))
-            .or_else(|| self.prefix_in_container.clone())
-            .map(|mut p| {
-                // required to end with a separator
-                // otherwise request will return only the entry of a prefix
-                if matches!(mode, ListingMode::WithDelimiter)
-                    && !p.ends_with(REMOTE_STORAGE_PREFIX_SEPARATOR)
-                {
-                    p.push(REMOTE_STORAGE_PREFIX_SEPARATOR);
-                }
-                p
-            });
+    ) -> anyhow::Result<Listing, DownloadError> {
+        let _permit = self.permit(RequestKind::List, cancel).await?;
 
-        async_stream::stream! {
-            let _permit = self.permit(RequestKind::List, cancel).await?;
+        let op = async {
+            // get the passed prefix or if it is not set use prefix_in_bucket value
+            let list_prefix = prefix
+                .map(|p| self.relative_path_to_name(p))
+                .or_else(|| self.prefix_in_container.clone())
+                .map(|mut p| {
+                    // required to end with a separator
+                    // otherwise request will return only the entry of a prefix
+                    if matches!(mode, ListingMode::WithDelimiter)
+                        && !p.ends_with(REMOTE_STORAGE_PREFIX_SEPARATOR)
+                    {
+                        p.push(REMOTE_STORAGE_PREFIX_SEPARATOR);
+                    }
+                    p
+                });
 
             let mut builder = self.client.list_blobs();
 
@@ -307,43 +300,21 @@ impl RemoteStorage for AzureBlobStorage {
                 builder = builder.max_results(MaxResults::new(limit));
             }
 
-            let mut next_marker = None;
+            let response = builder.into_stream();
+            let response = response.into_stream().map_err(to_download_error);
+            let response = tokio_stream::StreamExt::timeout(response, self.timeout);
+            let response = response.map(|res| match res {
+                Ok(res) => res,
+                Err(_elapsed) => Err(DownloadError::Timeout),
+            });
 
-            'outer: loop {
-                let mut builder = builder.clone();
-                if let Some(marker) = next_marker.clone() {
-                    builder = builder.marker(marker);
-                }
-                let response = builder.into_stream();
-                let response = response.into_stream().map_err(to_download_error);
-                let response = tokio_stream::StreamExt::timeout(response, self.timeout);
-                let response = response.map(|res| match res {
-                    Ok(res) => res,
-                    Err(_elapsed) => Err(DownloadError::Timeout),
-                });
+            let mut response = std::pin::pin!(response);
 
-                let mut response = std::pin::pin!(response);
+            let mut res = Listing::default();
 
-                let mut max_keys = max_keys.map(|mk| mk.get());
-                let next_item = tokio::select! {
-                    op = response.next() => Ok(op),
-                    _ = cancel.cancelled() => Err(DownloadError::Cancelled),
-                }?;
-                let Some(entry) = next_item else {
-                    // The list is complete, so yield it.
-                    break;
-                };
-
-                let mut res = Listing::default();
-                let entry = match entry {
-                    Ok(entry) => entry,
-                    Err(e) => {
-                        // The error is potentially retryable, so we must rewind the loop after yielding.
-                        yield Err(e);
-                        continue;
-                    }
-                };
-                next_marker = entry.continuation();
+            let mut max_keys = max_keys.map(|mk| mk.get());
+            while let Some(entry) = response.next().await {
+                let entry = entry?;
                 let prefix_iter = entry
                     .blobs
                     .prefixes()
@@ -353,12 +324,7 @@ impl RemoteStorage for AzureBlobStorage {
                 let blob_iter = entry
                     .blobs
                     .blobs()
-                    .map(|k| ListingObject{
-                        key: self.name_to_relative_path(&k.name),
-                        last_modified: k.properties.last_modified.into(),
-                        size: k.properties.content_length,
-                    }
-                    );
+                    .map(|k| self.name_to_relative_path(&k.name));
 
                 for key in blob_iter {
                     res.keys.push(key);
@@ -367,19 +333,19 @@ impl RemoteStorage for AzureBlobStorage {
                         assert!(mk > 0);
                         mk -= 1;
                         if mk == 0 {
-                            yield Ok(res); // limit reached
-                            break 'outer;
+                            return Ok(res); // limit reached
                         }
                         max_keys = Some(mk);
                     }
                 }
-                yield Ok(res);
-
-                // We are done here
-                if next_marker.is_none() {
-                    break;
-                }
             }
+
+            Ok(res)
+        };
+
+        tokio::select! {
+            res = op => res,
+            _ = cancel.cancelled() => Err(DownloadError::Cancelled),
         }
     }
 

libs/remote_storage/src/lib.rs

@@ -26,7 +26,7 @@ use anyhow::Context;
 use camino::{Utf8Path, Utf8PathBuf};
 
 use bytes::Bytes;
-use futures::{stream::Stream, StreamExt};
+use futures::stream::Stream;
 use serde::{Deserialize, Serialize};
 use tokio::sync::Semaphore;
 use tokio_util::sync::CancellationToken;
@@ -144,23 +144,15 @@ impl RemotePath {
 ///
 /// The WithDelimiter mode will populate `prefixes` and `keys` in the result.  The
 /// NoDelimiter mode will only populate `keys`.
-#[derive(Copy, Clone)]
 pub enum ListingMode {
     WithDelimiter,
     NoDelimiter,
 }
 
-#[derive(PartialEq, Eq, Debug)]
-pub struct ListingObject {
-    pub key: RemotePath,
-    pub last_modified: SystemTime,
-    pub size: u64,
-}
-
 #[derive(Default)]
 pub struct Listing {
     pub prefixes: Vec<RemotePath>,
-    pub keys: Vec<ListingObject>,
+    pub keys: Vec<RemotePath>,
 }
 
 /// Storage (potentially remote) API to manage its state.
@@ -168,18 +160,13 @@ pub struct Listing {
 /// providing basic CRUD operations for storage files.
 #[allow(async_fn_in_trait)]
 pub trait RemoteStorage: Send + Sync + 'static {
-    /// List objects in remote storage, with semantics matching AWS S3's [`ListObjectsV2`].
-    ///
-    /// The stream is guaranteed to return at least one element, even in the case of errors
-    /// (in that case it's an `Err()`), or an empty `Listing`.
-    ///
-    /// The stream is not ending if it returns an error, as long as [`is_permanent`] returns false on the error.
-    /// The `next` function can be retried, and maybe in a future retry, there will be success.
+    /// List objects in remote storage, with semantics matching AWS S3's ListObjectsV2.
+    /// (see `<https://docs.aws.amazon.com/AmazonS3/latest/API/API_ListObjectsV2.html>`)
     ///
     /// Note that the prefix is relative to any `prefix_in_bucket` configured for the client, not
     /// from the absolute root of the bucket.
     ///
-    /// `mode` configures whether to use a delimiter.  Without a delimiter, all keys
+    /// `mode` configures whether to use a delimiter.  Without a delimiter all keys
     /// within the prefix are listed in the `keys` of the result.  With a delimiter, any "directories" at the top level of
     /// the prefix are returned in the `prefixes` of the result, and keys in the top level of the prefix are
     /// returned in `keys` ().
@@ -188,32 +175,13 @@ pub trait RemoteStorage: Send + Sync + 'static {
     /// will iteratively call listobjects until it runs out of keys.  Note that this is not safe to use on
     /// unlimted size buckets, as the full list of objects is allocated into a monolithic data structure.
     ///
-    /// [`ListObjectsV2`]: <https://docs.aws.amazon.com/AmazonS3/latest/API/API_ListObjectsV2.html>
-    /// [`is_permanent`]: DownloadError::is_permanent
-    fn list_streaming(
-        &self,
-        prefix: Option<&RemotePath>,
-        mode: ListingMode,
-        max_keys: Option<NonZeroU32>,
-        cancel: &CancellationToken,
-    ) -> impl Stream<Item = Result<Listing, DownloadError>> + Send;
-
     async fn list(
         &self,
         prefix: Option<&RemotePath>,
-        mode: ListingMode,
+        _mode: ListingMode,
         max_keys: Option<NonZeroU32>,
         cancel: &CancellationToken,
-    ) -> Result<Listing, DownloadError> {
-        let mut stream = std::pin::pin!(self.list_streaming(prefix, mode, max_keys, cancel));
-        let mut combined = stream.next().await.expect("At least one item required")?;
-        while let Some(list) = stream.next().await {
-            let list = list?;
-            combined.keys.extend(list.keys.into_iter());
-            combined.prefixes.extend_from_slice(&list.prefixes);
-        }
-        Ok(combined)
-    }
+    ) -> Result<Listing, DownloadError>;
 
     /// Streams the local file contents into remote into the remote storage entry.
     ///
@@ -320,8 +288,8 @@ impl Debug for Download {
 
 /// Every storage, currently supported.
 /// Serves as a simple way to pass around the [`RemoteStorage`] without dealing with generics.
-// Require Clone for `Other` due to https://github.com/rust-lang/rust/issues/26925
 #[derive(Clone)]
+// Require Clone for `Other` due to https://github.com/rust-lang/rust/issues/26925
 pub enum GenericRemoteStorage<Other: Clone = Arc<UnreliableWrapper>> {
     LocalFs(LocalFs),
     AwsS3(Arc<S3Bucket>),
@@ -330,14 +298,13 @@ pub enum GenericRemoteStorage<Other: Clone = Arc<UnreliableWrapper>> {
 }
 
 impl<Other: RemoteStorage> GenericRemoteStorage<Arc<Other>> {
-    // See [`RemoteStorage::list`].
     pub async fn list(
         &self,
         prefix: Option<&RemotePath>,
         mode: ListingMode,
         max_keys: Option<NonZeroU32>,
         cancel: &CancellationToken,
-    ) -> Result<Listing, DownloadError> {
+    ) -> anyhow::Result<Listing, DownloadError> {
         match self {
             Self::LocalFs(s) => s.list(prefix, mode, max_keys, cancel).await,
             Self::AwsS3(s) => s.list(prefix, mode, max_keys, cancel).await,
@@ -346,23 +313,6 @@ impl<Other: RemoteStorage> GenericRemoteStorage<Arc<Other>> {
         }
     }
 
-    // See [`RemoteStorage::list_streaming`].
-    pub fn list_streaming<'a>(
-        &'a self,
-        prefix: Option<&'a RemotePath>,
-        mode: ListingMode,
-        max_keys: Option<NonZeroU32>,
-        cancel: &'a CancellationToken,
-    ) -> impl Stream<Item = Result<Listing, DownloadError>> + 'a + Send {
-        match self {
-            Self::LocalFs(s) => Box::pin(s.list_streaming(prefix, mode, max_keys, cancel))
-                as Pin<Box<dyn Stream<Item = Result<Listing, DownloadError>> + Send>>,
-            Self::AwsS3(s) => Box::pin(s.list_streaming(prefix, mode, max_keys, cancel)),
-            Self::AzureBlob(s) => Box::pin(s.list_streaming(prefix, mode, max_keys, cancel)),
-            Self::Unreliable(s) => Box::pin(s.list_streaming(prefix, mode, max_keys, cancel)),
-        }
-    }
-
     /// See [`RemoteStorage::upload`]
     pub async fn upload(
         &self,
@@ -493,7 +443,7 @@ impl<Other: RemoteStorage> GenericRemoteStorage<Arc<Other>> {
 }
 
 impl GenericRemoteStorage {
-    pub async fn from_config(storage_config: &RemoteStorageConfig) -> anyhow::Result<Self> {
+    pub fn from_config(storage_config: &RemoteStorageConfig) -> anyhow::Result<Self> {
         let timeout = storage_config.timeout;
         Ok(match &storage_config.storage {
             RemoteStorageKind::LocalFs { local_path: path } => {
@@ -508,7 +458,7 @@ impl GenericRemoteStorage {
                     std::env::var("AWS_ACCESS_KEY_ID").unwrap_or_else(|_| "<none>".into());
                 info!("Using s3 bucket '{}' in region '{}' as a remote storage, prefix in bucket: '{:?}', bucket endpoint: '{:?}', profile: {profile}, access_key_id: {access_key_id}",
                       s3_config.bucket_name, s3_config.bucket_region, s3_config.prefix_in_bucket, s3_config.endpoint);
-                Self::AwsS3(Arc::new(S3Bucket::new(s3_config, timeout).await?))
+                Self::AwsS3(Arc::new(S3Bucket::new(s3_config, timeout)?))
             }
             RemoteStorageKind::AzureContainer(azure_config) => {
                 let storage_account = azure_config
@@ -554,16 +504,6 @@ impl GenericRemoteStorage {
             None => self.download(from, cancel).await,
         }
     }
-
-    /// The name of the bucket/container/etc.
-    pub fn bucket_name(&self) -> Option<&str> {
-        match self {
-            Self::LocalFs(_s) => None,
-            Self::AwsS3(s) => Some(s.bucket_name()),
-            Self::AzureBlob(s) => Some(s.container_name()),
-            Self::Unreliable(_s) => None,
-        }
-    }
 }
 
 /// Extra set of key-value pairs that contain arbitrary metadata about the storage entry.

libs/remote_storage/src/local_fs.rs

@@ -23,8 +23,8 @@ use tokio_util::{io::ReaderStream, sync::CancellationToken};
 use utils::crashsafe::path_with_suffix_extension;
 
 use crate::{
-    Download, DownloadError, Listing, ListingMode, ListingObject, RemotePath, TimeTravelError,
-    TimeoutOrCancel, REMOTE_STORAGE_PREFIX_SEPARATOR,
+    Download, DownloadError, Listing, ListingMode, RemotePath, TimeTravelError, TimeoutOrCancel,
+    REMOTE_STORAGE_PREFIX_SEPARATOR,
 };
 
 use super::{RemoteStorage, StorageMetadata};
@@ -331,17 +331,6 @@ impl LocalFs {
 }
 
 impl RemoteStorage for LocalFs {
-    fn list_streaming(
-        &self,
-        prefix: Option<&RemotePath>,
-        mode: ListingMode,
-        max_keys: Option<NonZeroU32>,
-        cancel: &CancellationToken,
-    ) -> impl Stream<Item = Result<Listing, DownloadError>> {
-        let listing = self.list(prefix, mode, max_keys, cancel);
-        futures::stream::once(listing)
-    }
-
     async fn list(
         &self,
         prefix: Option<&RemotePath>,
@@ -357,29 +346,19 @@ impl RemoteStorage for LocalFs {
                 .list_recursive(prefix)
                 .await
                 .map_err(DownloadError::Other)?;
-            let objects = keys
+            let keys = keys
                 .into_iter()
-                .filter_map(|k| {
+                .filter(|k| {
                     let path = k.with_base(&self.storage_root);
-                    if path.is_dir() {
-                        None
-                    } else {
-                        Some(ListingObject {
-                            key: k.clone(),
-                            // LocalFs is just for testing, so just specify a dummy time
-                            last_modified: SystemTime::now(),
-                            size: 0,
-                        })
-                    }
+                    !path.is_dir()
                 })
                 .collect();
 
             if let ListingMode::NoDelimiter = mode {
-                result.keys = objects;
+                result.keys = keys;
             } else {
                 let mut prefixes = HashSet::new();
-                for object in objects {
-                    let key = object.key;
+                for key in keys {
                     // If the part after the prefix includes a "/", take only the first part and put it in `prefixes`.
                     let relative_key = if let Some(prefix) = prefix {
                         let mut prefix = prefix.clone();
@@ -408,12 +387,9 @@ impl RemoteStorage for LocalFs {
                             .to_owned();
                         prefixes.insert(first_part);
                     } else {
-                        result.keys.push(ListingObject {
-                            key: RemotePath::from_string(&relative_key).unwrap(),
-                            // LocalFs is just for testing
-                            last_modified: SystemTime::now(),
-                            size: 0,
-                        });
+                        result
+                            .keys
+                            .push(RemotePath::from_string(&relative_key).unwrap());
                     }
                 }
                 result.prefixes = prefixes
@@ -963,11 +939,7 @@ mod fs_tests {
             .await?;
         assert!(listing.prefixes.is_empty());
         assert_eq!(
-            listing
-                .keys
-                .into_iter()
-                .map(|o| o.key)
-                .collect::<HashSet<_>>(),
+            listing.keys.into_iter().collect::<HashSet<_>>(),
             HashSet::from([uncle.clone(), child.clone(), child_sibling.clone()])
         );
 
@@ -992,7 +964,7 @@ mod fs_tests {
             )
             .await?;
         assert_eq!(
-            listing.keys.into_iter().map(|o| o.key).collect::<Vec<_>>(),
+            listing.keys,
             [RemotePath::from_string("uncle").unwrap()].to_vec()
         );
         assert_eq!(
@@ -1009,7 +981,7 @@ mod fs_tests {
                 &cancel,
             )
             .await?;
-        assert_eq!(listing.keys, vec![]);
+        assert_eq!(listing.keys, [].to_vec());
         assert_eq!(
             listing.prefixes,
             [RemotePath::from_string("grandparent").unwrap()].to_vec()
@@ -1024,7 +996,7 @@ mod fs_tests {
                 &cancel,
             )
             .await?;
-        assert_eq!(listing.keys, vec![]);
+        assert_eq!(listing.keys, [].to_vec());
         assert_eq!(
             listing.prefixes,
             [RemotePath::from_string("grandparent").unwrap()].to_vec()
@@ -1057,7 +1029,7 @@ mod fs_tests {
                 &cancel,
             )
             .await?;
-        assert_eq!(listing.keys, vec![]);
+        assert_eq!(listing.keys, [].to_vec());
 
         let mut found_prefixes = listing.prefixes.clone();
         found_prefixes.sort();

libs/remote_storage/src/s3_bucket.rs

@@ -16,10 +16,16 @@ use std::{
 
 use anyhow::{anyhow, Context as _};
 use aws_config::{
-    default_provider::credentials::DefaultCredentialsChain,
+    environment::credentials::EnvironmentVariableCredentialsProvider,
+    imds::credentials::ImdsCredentialsProvider,
+    meta::credentials::CredentialsProviderChain,
+    profile::ProfileFileCredentialsProvider,
+    provider_config::ProviderConfig,
     retry::{RetryConfigBuilder, RetryMode},
+    web_identity_token::WebIdentityTokenCredentialsProvider,
     BehaviorVersion,
 };
+use aws_credential_types::provider::SharedCredentialsProvider;
 use aws_sdk_s3::{
     config::{AsyncSleep, IdentityCache, Region, SharedAsyncSleep},
     error::SdkError,
@@ -44,9 +50,8 @@ use crate::{
     error::Cancelled,
     metrics::{start_counting_cancelled_wait, start_measuring_requests},
     support::PermitCarrying,
-    ConcurrencyLimiter, Download, DownloadError, Listing, ListingMode, ListingObject, RemotePath,
-    RemoteStorage, TimeTravelError, TimeoutOrCancel, MAX_KEYS_PER_DELETE,
-    REMOTE_STORAGE_PREFIX_SEPARATOR,
+    ConcurrencyLimiter, Download, DownloadError, Listing, ListingMode, RemotePath, RemoteStorage,
+    TimeTravelError, TimeoutOrCancel, MAX_KEYS_PER_DELETE, REMOTE_STORAGE_PREFIX_SEPARATOR,
 };
 
 use crate::metrics::AttemptOutcome;
@@ -71,27 +76,40 @@ struct GetObjectRequest {
 }
 impl S3Bucket {
     /// Creates the S3 storage, errors if incorrect AWS S3 configuration provided.
-    pub async fn new(remote_storage_config: &S3Config, timeout: Duration) -> anyhow::Result<Self> {
+    pub fn new(remote_storage_config: &S3Config, timeout: Duration) -> anyhow::Result<Self> {
         tracing::debug!(
             "Creating s3 remote storage for S3 bucket {}",
             remote_storage_config.bucket_name
         );
 
-        let region = Region::new(remote_storage_config.bucket_region.clone());
-        let region_opt = Some(region.clone());
+        let region = Some(Region::new(remote_storage_config.bucket_region.clone()));
 
-        // https://docs.aws.amazon.com/sdkref/latest/guide/standardized-credentials.html
-        // https://docs.rs/aws-config/latest/aws_config/default_provider/credentials/struct.DefaultCredentialsChain.html
-        // Incomplete list of auth methods used by this:
-        // * "AWS_ACCESS_KEY_ID", "AWS_SECRET_ACCESS_KEY"
-        // * "AWS_PROFILE" / `aws sso login --profile <profile>`
-        // * "AWS_WEB_IDENTITY_TOKEN_FILE", "AWS_ROLE_ARN", "AWS_ROLE_SESSION_NAME"
-        // * http (ECS/EKS) container credentials
-        // * imds v2
-        let credentials_provider = DefaultCredentialsChain::builder()
-            .region(region)
-            .build()
-            .await;
+        let provider_conf = ProviderConfig::without_region().with_region(region.clone());
+
+        let credentials_provider = {
+            // uses "AWS_ACCESS_KEY_ID", "AWS_SECRET_ACCESS_KEY"
+            CredentialsProviderChain::first_try(
+                "env",
+                EnvironmentVariableCredentialsProvider::new(),
+            )
+            // uses "AWS_PROFILE" / `aws sso login --profile <profile>`
+            .or_else(
+                "profile-sso",
+                ProfileFileCredentialsProvider::builder()
+                    .configure(&provider_conf)
+                    .build(),
+            )
+            // uses "AWS_WEB_IDENTITY_TOKEN_FILE", "AWS_ROLE_ARN", "AWS_ROLE_SESSION_NAME"
+            // needed to access remote extensions bucket
+            .or_else(
+                "token",
+                WebIdentityTokenCredentialsProvider::builder()
+                    .configure(&provider_conf)
+                    .build(),
+            )
+            // uses imds v2
+            .or_else("imds", ImdsCredentialsProvider::builder().build())
+        };
 
         // AWS SDK requires us to specify how the RetryConfig should sleep when it wants to back off
         let sleep_impl: Arc<dyn AsyncSleep> = Arc::new(TokioSleep::new());
@@ -100,9 +118,9 @@ impl S3Bucket {
             #[allow(deprecated)] /* TODO: https://github.com/neondatabase/neon/issues/7665 */
             BehaviorVersion::v2023_11_09(),
         )
-        .region(region_opt)
+        .region(region)
         .identity_cache(IdentityCache::lazy().build())
-        .credentials_provider(credentials_provider)
+        .credentials_provider(SharedCredentialsProvider::new(credentials_provider))
         .sleep_impl(SharedAsyncSleep::from(sleep_impl));
 
         let sdk_config: aws_config::SdkConfig = std::thread::scope(|s| {
@@ -387,10 +405,6 @@ impl S3Bucket {
         }
         Ok(())
     }
-
-    pub fn bucket_name(&self) -> &str {
-        &self.bucket_name
-    }
 }
 
 pin_project_lite::pin_project! {
@@ -468,16 +482,17 @@ impl<S: Stream<Item = std::io::Result<Bytes>>> Stream for TimedDownload<S> {
 }
 
 impl RemoteStorage for S3Bucket {
-    fn list_streaming(
+    async fn list(
         &self,
         prefix: Option<&RemotePath>,
         mode: ListingMode,
         max_keys: Option<NonZeroU32>,
         cancel: &CancellationToken,
-    ) -> impl Stream<Item = Result<Listing, DownloadError>> {
+    ) -> Result<Listing, DownloadError> {
         let kind = RequestKind::List;
         // s3 sdk wants i32
         let mut max_keys = max_keys.map(|mk| mk.get() as i32);
+        let mut result = Listing::default();
 
         // get the passed prefix or if it is not set use prefix_in_bucket value
         let list_prefix = prefix
@@ -489,119 +504,89 @@ impl RemoteStorage for S3Bucket {
                 })
             });
 
-        async_stream::stream! {
-            let _permit = self.permit(kind, cancel).await?;
+        let _permit = self.permit(kind, cancel).await?;
 
-            let mut continuation_token = None;
-            'outer: loop {
-                let started_at = start_measuring_requests(kind);
+        let mut continuation_token = None;
 
-                // min of two Options, returning Some if one is value and another is
-                // None (None is smaller than anything, so plain min doesn't work).
-                let request_max_keys = self
-                    .max_keys_per_list_response
-                    .into_iter()
-                    .chain(max_keys.into_iter())
-                    .min();
-                let mut request = self
-                    .client
-                    .list_objects_v2()
-                    .bucket(self.bucket_name.clone())
-                    .set_prefix(list_prefix.clone())
-                    .set_continuation_token(continuation_token.clone())
-                    .set_max_keys(request_max_keys);
+        loop {
+            let started_at = start_measuring_requests(kind);
 
-                if let ListingMode::WithDelimiter = mode {
-                    request = request.delimiter(REMOTE_STORAGE_PREFIX_SEPARATOR.to_string());
-                }
+            // min of two Options, returning Some if one is value and another is
+            // None (None is smaller than anything, so plain min doesn't work).
+            let request_max_keys = self
+                .max_keys_per_list_response
+                .into_iter()
+                .chain(max_keys.into_iter())
+                .min();
+            let mut request = self
+                .client
+                .list_objects_v2()
+                .bucket(self.bucket_name.clone())
+                .set_prefix(list_prefix.clone())
+                .set_continuation_token(continuation_token)
+                .set_max_keys(request_max_keys);
 
-                let request = request.send();
-
-                let response = tokio::select! {
-                    res = request => Ok(res),
-                    _ = tokio::time::sleep(self.timeout) => Err(DownloadError::Timeout),
-                    _ = cancel.cancelled() => Err(DownloadError::Cancelled),
-                }?;
-
-                let response = response
-                    .context("Failed to list S3 prefixes")
-                    .map_err(DownloadError::Other);
-
-                let started_at = ScopeGuard::into_inner(started_at);
-
-                crate::metrics::BUCKET_METRICS
-                    .req_seconds
-                    .observe_elapsed(kind, &response, started_at);
-
-                let response = match response {
-                    Ok(response) => response,
-                    Err(e) => {
-                        // The error is potentially retryable, so we must rewind the loop after yielding.
-                        yield Err(e);
-                        continue 'outer;
-                    },
-                };
-
-                let keys = response.contents();
-                let prefixes = response.common_prefixes.as_deref().unwrap_or_default();
-
-                tracing::debug!("list: {} prefixes, {} keys", prefixes.len(), keys.len());
-                let mut result = Listing::default();
-
-                for object in keys {
-                    let key = object.key().expect("response does not contain a key");
-                    let key = self.s3_object_to_relative_path(key);
-
-                    let last_modified = match object.last_modified.map(SystemTime::try_from) {
-                        Some(Ok(t)) => t,
-                        Some(Err(_)) => {
-                            tracing::warn!("Remote storage last_modified {:?} for {} is out of bounds",
-                                object.last_modified, key
-                        );
-                            SystemTime::now()
-                        },
-                        None => {
-                            SystemTime::now()
-                        }
-                    };
-
-                    let size = object.size.unwrap_or(0) as u64;
-
-                    result.keys.push(ListingObject{
-                        key,
-                        last_modified,
-                        size,
-                    });
-                    if let Some(mut mk) = max_keys {
-                        assert!(mk > 0);
-                        mk -= 1;
-                        if mk == 0 {
-                            // limit reached
-                            yield Ok(result);
-                            break 'outer;
-                        }
-                        max_keys = Some(mk);
-                    }
-                }
-
-                // S3 gives us prefixes like "foo/", we return them like "foo"
-                result.prefixes.extend(prefixes.iter().filter_map(|o| {
-                    Some(
-                        self.s3_object_to_relative_path(
-                            o.prefix()?
-                                .trim_end_matches(REMOTE_STORAGE_PREFIX_SEPARATOR),
-                        ),
-                    )
-                }));
-
-                yield Ok(result);
-
-                continuation_token = match response.next_continuation_token {
-                    Some(new_token) => Some(new_token),
-                    None => break,
-                };
+            if let ListingMode::WithDelimiter = mode {
+                request = request.delimiter(REMOTE_STORAGE_PREFIX_SEPARATOR.to_string());
             }
+
+            let request = request.send();
+
+            let response = tokio::select! {
+                res = request => res,
+                _ = tokio::time::sleep(self.timeout) => return Err(DownloadError::Timeout),
+                _ = cancel.cancelled() => return Err(DownloadError::Cancelled),
+            };
+
+            let response = response
+                .context("Failed to list S3 prefixes")
+                .map_err(DownloadError::Other);
+
+            let started_at = ScopeGuard::into_inner(started_at);
+
+            crate::metrics::BUCKET_METRICS
+                .req_seconds
+                .observe_elapsed(kind, &response, started_at);
+
+            let response = response?;
+
+            let keys = response.contents();
+            let empty = Vec::new();
+            let prefixes = response.common_prefixes.as_ref().unwrap_or(&empty);
+
+            tracing::debug!("list: {} prefixes, {} keys", prefixes.len(), keys.len());
+
+            for object in keys {
+                let object_path = object.key().expect("response does not contain a key");
+                let remote_path = self.s3_object_to_relative_path(object_path);
+                result.keys.push(remote_path);
+                if let Some(mut mk) = max_keys {
+                    assert!(mk > 0);
+                    mk -= 1;
+                    if mk == 0 {
+                        return Ok(result); // limit reached
+                    }
+                    max_keys = Some(mk);
+                }
+            }
+
+            // S3 gives us prefixes like "foo/", we return them like "foo"
+            result.prefixes.extend(prefixes.iter().filter_map(|o| {
+                Some(
+                    self.s3_object_to_relative_path(
+                        o.prefix()?
+                            .trim_end_matches(REMOTE_STORAGE_PREFIX_SEPARATOR),
+                    ),
+                )
+            }));
+
+            continuation_token = match response.next_continuation_token {
+                Some(new_token) => Some(new_token),
+                None => break,
+            };
         }
+
+        Ok(result)
     }
 
     async fn upload(
@@ -1056,8 +1041,8 @@ mod tests {
 
     use crate::{RemotePath, S3Bucket, S3Config};
 
-    #[tokio::test]
-    async fn relative_path() {
+    #[test]
+    fn relative_path() {
         let all_paths = ["", "some/path", "some/path/"];
         let all_paths: Vec<RemotePath> = all_paths
             .iter()
@@ -1100,9 +1085,8 @@ mod tests {
                 max_keys_per_list_response: Some(5),
                 upload_storage_class: None,
             };
-            let storage = S3Bucket::new(&config, std::time::Duration::ZERO)
-                .await
-                .expect("remote storage init");
+            let storage =
+                S3Bucket::new(&config, std::time::Duration::ZERO).expect("remote storage init");
             for (test_path_idx, test_path) in all_paths.iter().enumerate() {
                 let result = storage.relative_path_to_s3_object(test_path);
                 let expected = expected_outputs[prefix_idx][test_path_idx];

libs/remote_storage/src/simulate_failures.rs

@@ -3,7 +3,6 @@
 //! testing purposes.
 use bytes::Bytes;
 use futures::stream::Stream;
-use futures::StreamExt;
 use std::collections::HashMap;
 use std::num::NonZeroU32;
 use std::sync::Mutex;
@@ -108,23 +107,6 @@ impl UnreliableWrapper {
 type VoidStorage = crate::LocalFs;
 
 impl RemoteStorage for UnreliableWrapper {
-    fn list_streaming(
-        &self,
-        prefix: Option<&RemotePath>,
-        mode: ListingMode,
-        max_keys: Option<NonZeroU32>,
-        cancel: &CancellationToken,
-    ) -> impl Stream<Item = Result<Listing, DownloadError>> + Send {
-        async_stream::stream! {
-            self.attempt(RemoteOp::ListPrefixes(prefix.cloned()))
-                .map_err(DownloadError::Other)?;
-            let mut stream = self.inner
-                .list_streaming(prefix, mode, max_keys, cancel);
-            while let Some(item) = stream.next().await {
-                yield item;
-            }
-        }
-    }
     async fn list(
         &self,
         prefix: Option<&RemotePath>,

libs/remote_storage/tests/common/mod.rs

@@ -152,7 +152,7 @@ pub(crate) async fn upload_remote_data(
     let mut upload_tasks = JoinSet::new();
     let cancel = CancellationToken::new();
 
-    for i in 1..=upload_tasks_count {
+    for i in 1..upload_tasks_count + 1 {
         let task_client = Arc::clone(client);
         let cancel = cancel.clone();
 

libs/remote_storage/tests/common/tests.rs

@@ -1,6 +1,5 @@
 use anyhow::Context;
 use camino::Utf8Path;
-use futures::StreamExt;
 use remote_storage::ListingMode;
 use remote_storage::RemotePath;
 use std::sync::Arc;
@@ -30,10 +29,10 @@ use super::{
 /// * with no prefix, it lists everything after its `${random_prefix_part}/` — that should be `${base_prefix_str}` value only
 /// * with `${base_prefix_str}/` prefix, it lists every `sub_prefix_${i}`
 ///
-/// In the `MaybeEnabledStorageWithTestBlobs::setup`, we set the `max_keys_in_list_response` param to limit the keys in a single response.
-/// This way, we are able to test the pagination, by ensuring all results are returned from the remote storage and avoid uploading too many blobs to S3,
-/// as the current default AWS S3 pagination limit is 1000.
-/// (see <https://docs.aws.amazon.com/AmazonS3/latest/API/API_ListObjectsV2.html#API_ListObjectsV2_RequestSyntax>).
+/// With the real S3 enabled and `#[cfg(test)]` Rust configuration used, the S3 client test adds a `max-keys` param to limit the response keys.
+/// This way, we are able to test the pagination implicitly, by ensuring all results are returned from the remote storage and avoid uploading too many blobs to S3,
+/// since current default AWS S3 pagination limit is 1000.
+/// (see https://docs.aws.amazon.com/AmazonS3/latest/API/API_ListObjectsV2.html#API_ListObjectsV2_RequestSyntax)
 ///
 /// Lastly, the test attempts to clean up and remove all uploaded S3 files.
 /// If any errors appear during the clean up, they get logged, but the test is not failed or stopped until clean up is finished.
@@ -88,41 +87,6 @@ async fn pagination_should_work(ctx: &mut MaybeEnabledStorageWithTestBlobs) -> a
         "remote storage nested prefixes list mismatches with the uploads. Remote only prefixes: {remote_only_prefixes:?}, missing uploaded prefixes: {missing_uploaded_prefixes:?}",
     );
 
-    // list_streaming
-
-    let prefix_with_slash = base_prefix.add_trailing_slash();
-    let mut nested_remote_prefixes_st = test_client.list_streaming(
-        Some(&prefix_with_slash),
-        ListingMode::WithDelimiter,
-        None,
-        &cancel,
-    );
-    let mut nested_remote_prefixes_combined = HashSet::new();
-    let mut segments = 0;
-    let mut segment_max_size = 0;
-    while let Some(st) = nested_remote_prefixes_st.next().await {
-        let st = st?;
-        segment_max_size = segment_max_size.max(st.prefixes.len());
-        nested_remote_prefixes_combined.extend(st.prefixes.into_iter());
-        segments += 1;
-    }
-    assert!(segments > 1, "less than 2 segments: {segments}");
-    assert!(
-        segment_max_size * 2 <= nested_remote_prefixes_combined.len(),
-        "double of segment_max_size={segment_max_size} larger number of remote prefixes of {}",
-        nested_remote_prefixes_combined.len()
-    );
-    let remote_only_prefixes = nested_remote_prefixes_combined
-        .difference(&expected_remote_prefixes)
-        .collect::<HashSet<_>>();
-    let missing_uploaded_prefixes = expected_remote_prefixes
-        .difference(&nested_remote_prefixes_combined)
-        .collect::<HashSet<_>>();
-    assert_eq!(
-        remote_only_prefixes.len() + missing_uploaded_prefixes.len(), 0,
-        "remote storage nested prefixes list mismatches with the uploads. Remote only prefixes: {remote_only_prefixes:?}, missing uploaded prefixes: {missing_uploaded_prefixes:?}",
-    );
-
     Ok(())
 }
 
@@ -156,7 +120,6 @@ async fn list_no_delimiter_works(
         .context("client list root files failure")?
         .keys
         .into_iter()
-        .map(|o| o.key)
         .collect::<HashSet<_>>();
     assert_eq!(
         root_files,
@@ -183,7 +146,6 @@ async fn list_no_delimiter_works(
         .context("client list nested files failure")?
         .keys
         .into_iter()
-        .map(|o| o.key)
         .collect::<HashSet<_>>();
     let trim_remote_blobs: HashSet<_> = ctx
         .remote_blobs

libs/remote_storage/tests/test_real_azure.rs

@@ -31,7 +31,6 @@ struct EnabledAzure {
 impl EnabledAzure {
     async fn setup(max_keys_in_list_response: Option<i32>) -> Self {
         let client = create_azure_client(max_keys_in_list_response)
-            .await
             .context("Azure client creation")
             .expect("Azure client creation failed");
 
@@ -188,7 +187,7 @@ impl AsyncTestContext for MaybeEnabledStorageWithSimpleTestBlobs {
     }
 }
 
-async fn create_azure_client(
+fn create_azure_client(
     max_keys_per_list_response: Option<i32>,
 ) -> anyhow::Result<Arc<GenericRemoteStorage>> {
     use rand::Rng;
@@ -222,8 +221,6 @@ async fn create_azure_client(
         timeout: Duration::from_secs(120),
     };
     Ok(Arc::new(
-        GenericRemoteStorage::from_config(&remote_storage_config)
-            .await
-            .context("remote storage init")?,
+        GenericRemoteStorage::from_config(&remote_storage_config).context("remote storage init")?,
     ))
 }

libs/remote_storage/tests/test_real_s3.rs

@@ -81,7 +81,6 @@ async fn s3_time_travel_recovery_works(ctx: &mut MaybeEnabledStorage) -> anyhow:
                 .context("list root files failure")?
                 .keys
                 .into_iter()
-                .map(|o| o.key)
                 .collect::<HashSet<_>>(),
         )
     }
@@ -198,7 +197,6 @@ struct EnabledS3 {
 impl EnabledS3 {
     async fn setup(max_keys_in_list_response: Option<i32>) -> Self {
         let client = create_s3_client(max_keys_in_list_response)
-            .await
             .context("S3 client creation")
             .expect("S3 client creation failed");
 
@@ -354,7 +352,7 @@ impl AsyncTestContext for MaybeEnabledStorageWithSimpleTestBlobs {
     }
 }
 
-async fn create_s3_client(
+fn create_s3_client(
     max_keys_per_list_response: Option<i32>,
 ) -> anyhow::Result<Arc<GenericRemoteStorage>> {
     use rand::Rng;
@@ -387,9 +385,7 @@ async fn create_s3_client(
         timeout: RemoteStorageConfig::DEFAULT_TIMEOUT,
     };
     Ok(Arc::new(
-        GenericRemoteStorage::from_config(&remote_storage_config)
-            .await
-            .context("remote storage init")?,
+        GenericRemoteStorage::from_config(&remote_storage_config).context("remote storage init")?,
     ))
 }
 

libs/utils/Cargo.toml

@@ -20,6 +20,7 @@ bincode.workspace = true
 bytes.workspace = true
 camino.workspace = true
 chrono.workspace = true
+heapless.workspace = true
 hex = { workspace = true, features = ["serde"] }
 humantime.workspace = true
 hyper = { workspace = true, features = ["full"] }

libs/utils/src/auth.rs

@@ -18,25 +18,21 @@ const STORAGE_TOKEN_ALGORITHM: Algorithm = Algorithm::EdDSA;
 #[derive(Debug, Serialize, Deserialize, Clone, Copy, PartialEq)]
 #[serde(rename_all = "lowercase")]
 pub enum Scope {
-    /// Provides access to all data for a specific tenant (specified in `struct Claims` below)
+    // Provides access to all data for a specific tenant (specified in `struct Claims` below)
     // TODO: join these two?
     Tenant,
-    /// Provides blanket access to all tenants on the pageserver plus pageserver-wide APIs.
-    /// Should only be used e.g. for status check/tenant creation/list.
+    // Provides blanket access to all tenants on the pageserver plus pageserver-wide APIs.
+    // Should only be used e.g. for status check/tenant creation/list.
     PageServerApi,
-    /// Provides blanket access to all data on the safekeeper plus safekeeper-wide APIs.
-    /// Should only be used e.g. for status check.
-    /// Currently also used for connection from any pageserver to any safekeeper.
+    // Provides blanket access to all data on the safekeeper plus safekeeper-wide APIs.
+    // Should only be used e.g. for status check.
+    // Currently also used for connection from any pageserver to any safekeeper.
     SafekeeperData,
-    /// The scope used by pageservers in upcalls to storage controller and cloud control plane
+    // The scope used by pageservers in upcalls to storage controller and cloud control plane
     #[serde(rename = "generations_api")]
     GenerationsApi,
-    /// Allows access to control plane managment API and some storage controller endpoints.
+    // Allows access to control plane managment API and some storage controller endpoints.
     Admin,
-
-    /// Allows access to storage controller APIs used by the scrubber, to interrogate the state
-    /// of a tenant & post scrub results.
-    Scrubber,
 }
 
 /// JWT payload. See docs/authentication.md for the format

libs/utils/src/history_buffer.rs

@@ -0,0 +1,196 @@
+//! A heapless buffer for events of sorts.
+
+use std::ops;
+
+use heapless::HistoryBuffer;
+
+#[derive(Debug, Clone)]
+pub struct HistoryBufferWithDropCounter<T, const L: usize> {
+    buffer: HistoryBuffer<T, L>,
+    drop_count: u64,
+}
+
+impl<T, const L: usize> HistoryBufferWithDropCounter<T, L> {
+    pub fn write(&mut self, data: T) {
+        let len_before = self.buffer.len();
+        self.buffer.write(data);
+        let len_after = self.buffer.len();
+        self.drop_count += u64::from(len_before == len_after);
+    }
+    pub fn drop_count(&self) -> u64 {
+        self.drop_count
+    }
+    pub fn map<U, F: Fn(&T) -> U>(&self, f: F) -> HistoryBufferWithDropCounter<U, L> {
+        let mut buffer = HistoryBuffer::new();
+        buffer.extend(self.buffer.oldest_ordered().map(f));
+        HistoryBufferWithDropCounter::<U, L> {
+            buffer,
+            drop_count: self.drop_count,
+        }
+    }
+}
+
+impl<T, const L: usize> Default for HistoryBufferWithDropCounter<T, L> {
+    fn default() -> Self {
+        Self {
+            buffer: HistoryBuffer::default(),
+            drop_count: 0,
+        }
+    }
+}
+
+impl<T, const L: usize> ops::Deref for HistoryBufferWithDropCounter<T, L> {
+    type Target = HistoryBuffer<T, L>;
+
+    fn deref(&self) -> &Self::Target {
+        &self.buffer
+    }
+}
+
+#[derive(serde::Serialize, serde::Deserialize)]
+struct SerdeRepr<T> {
+    buffer: Vec<T>,
+    buffer_size: usize,
+    drop_count: u64,
+}
+
+impl<'a, T, const L: usize> From<&'a HistoryBufferWithDropCounter<T, L>> for SerdeRepr<T>
+where
+    T: Clone + serde::Serialize,
+{
+    fn from(value: &'a HistoryBufferWithDropCounter<T, L>) -> Self {
+        let HistoryBufferWithDropCounter { buffer, drop_count } = value;
+        SerdeRepr {
+            buffer: buffer.iter().cloned().collect(),
+            buffer_size: L,
+            drop_count: *drop_count,
+        }
+    }
+}
+
+impl<T, const L: usize> serde::Serialize for HistoryBufferWithDropCounter<T, L>
+where
+    T: Clone + serde::Serialize,
+{
+    fn serialize<S>(&self, serializer: S) -> Result<S::Ok, S::Error>
+    where
+        S: serde::Serializer,
+    {
+        SerdeRepr::from(self).serialize(serializer)
+    }
+}
+
+impl<'de, T, const L: usize> serde::de::Deserialize<'de> for HistoryBufferWithDropCounter<T, L>
+where
+    T: Clone + serde::Deserialize<'de>,
+{
+    fn deserialize<D>(deserializer: D) -> Result<Self, D::Error>
+    where
+        D: serde::Deserializer<'de>,
+    {
+        let SerdeRepr {
+            buffer: des_buffer,
+            drop_count,
+            buffer_size,
+        } = SerdeRepr::<T>::deserialize(deserializer)?;
+        if buffer_size != L {
+            use serde::de::Error;
+            return Err(D::Error::custom(format!(
+                "invalid buffer_size, expecting {L} got {buffer_size}"
+            )));
+        }
+        let mut buffer = HistoryBuffer::new();
+        buffer.extend(des_buffer);
+        Ok(HistoryBufferWithDropCounter { buffer, drop_count })
+    }
+}
+
+#[cfg(test)]
+mod test {
+    use super::HistoryBufferWithDropCounter;
+
+    #[test]
+    fn test_basics() {
+        let mut b = HistoryBufferWithDropCounter::<usize, 2>::default();
+        b.write(1);
+        b.write(2);
+        b.write(3);
+        assert!(b.iter().any(|e| *e == 2));
+        assert!(b.iter().any(|e| *e == 3));
+        assert!(!b.iter().any(|e| *e == 1));
+
+        // round-trip serde
+        let round_tripped: HistoryBufferWithDropCounter<usize, 2> =
+            serde_json::from_str(&serde_json::to_string(&b).unwrap()).unwrap();
+        assert_eq!(
+            round_tripped.iter().cloned().collect::<Vec<_>>(),
+            b.iter().cloned().collect::<Vec<_>>()
+        );
+    }
+
+    #[test]
+    fn test_drop_count_works() {
+        let mut b = HistoryBufferWithDropCounter::<_, 2>::default();
+        b.write(1);
+        assert_eq!(b.drop_count(), 0);
+        b.write(2);
+        assert_eq!(b.drop_count(), 0);
+        b.write(3);
+        assert_eq!(b.drop_count(), 1);
+        b.write(4);
+        assert_eq!(b.drop_count(), 2);
+    }
+
+    #[test]
+    fn test_clone_works() {
+        let mut b = HistoryBufferWithDropCounter::<_, 2>::default();
+        b.write(1);
+        b.write(2);
+        b.write(3);
+        assert_eq!(b.drop_count(), 1);
+        let mut c = b.clone();
+        assert_eq!(c.drop_count(), 1);
+        assert!(c.iter().any(|e| *e == 2));
+        assert!(c.iter().any(|e| *e == 3));
+        assert!(!c.iter().any(|e| *e == 1));
+
+        c.write(4);
+        assert!(c.iter().any(|e| *e == 4));
+        assert!(!b.iter().any(|e| *e == 4));
+    }
+
+    #[test]
+    fn test_map() {
+        let mut b = HistoryBufferWithDropCounter::<_, 2>::default();
+
+        b.write(1);
+        assert_eq!(b.drop_count(), 0);
+        {
+            let c = b.map(|i| i + 10);
+            assert_eq!(c.oldest_ordered().cloned().collect::<Vec<_>>(), vec![11]);
+            assert_eq!(c.drop_count(), 0);
+        }
+
+        b.write(2);
+        assert_eq!(b.drop_count(), 0);
+        {
+            let c = b.map(|i| i + 10);
+            assert_eq!(
+                c.oldest_ordered().cloned().collect::<Vec<_>>(),
+                vec![11, 12]
+            );
+            assert_eq!(c.drop_count(), 0);
+        }
+
+        b.write(3);
+        assert_eq!(b.drop_count(), 1);
+        {
+            let c = b.map(|i| i + 10);
+            assert_eq!(
+                c.oldest_ordered().cloned().collect::<Vec<_>>(),
+                vec![12, 13]
+            );
+            assert_eq!(c.drop_count(), 1);
+        }
+    }
+}

libs/utils/src/lib.rs

@@ -59,6 +59,8 @@ pub mod signals;
 
 pub mod fs_ext;
 
+pub mod history_buffer;
+
 pub mod measured_stream;
 
 pub mod serde_percent;
@@ -128,7 +130,7 @@ pub mod circuit_breaker;
 ///
 /// #############################################################################################
 /// TODO this macro is not the way the library is intended to be used, see <https://github.com/neondatabase/neon/issues/1565> for details.
-/// We used `cachepot` to reduce our current CI build times: <https://github.com/neondatabase/cloud/pull/1033#issuecomment-1100935036>
+/// We use `cachepot` to reduce our current CI build times: <https://github.com/neondatabase/cloud/pull/1033#issuecomment-1100935036>
 /// Yet, it seems to ignore the GIT_VERSION env variable, passed to Docker build, even with build.rs that contains
 /// `println!("cargo:rerun-if-env-changed=GIT_VERSION");` code for cachepot cache invalidation.
 /// The problem needs further investigation and regular `const` declaration instead of a macro.

libs/utils/src/shard.rs

@@ -49,7 +49,6 @@ pub struct TenantShardId {
 
 impl ShardCount {
     pub const MAX: Self = Self(u8::MAX);
-    pub const MIN: Self = Self(0);
 
     /// The internal value of a ShardCount may be zero, which means "1 shard, but use
     /// legacy format for TenantShardId that excludes the shard suffix", also known

libs/utils/src/sync/gate.rs

@@ -78,9 +78,8 @@ impl Drop for GateGuard {
     }
 }
 
-#[derive(Debug, thiserror::Error)]
+#[derive(Debug)]
 pub enum GateError {
-    #[error("gate is closed")]
     GateClosed,
 }
 

pageserver/Cargo.toml

@@ -49,7 +49,6 @@ postgres_backend.workspace = true
 postgres-protocol.workspace = true
 postgres-types.workspace = true
 rand.workspace = true
-range-set-blaze = { version = "0.1.16", features = ["alloc"] }
 regex.workspace = true
 scopeguard.workspace = true
 serde.workspace = true
@@ -108,7 +107,3 @@ harness = false
 [[bench]]
 name = "bench_walredo"
 harness = false
-
-[[bench]]
-name = "bench_ingest"
-harness = false

pageserver/benches/bench_ingest.rs

@@ -1,239 +0,0 @@
-use std::{env, num::NonZeroUsize};
-
-use bytes::Bytes;
-use camino::Utf8PathBuf;
-use criterion::{criterion_group, criterion_main, Criterion};
-use pageserver::{
-    config::PageServerConf,
-    context::{DownloadBehavior, RequestContext},
-    l0_flush::{L0FlushConfig, L0FlushGlobalState},
-    page_cache,
-    repository::Value,
-    task_mgr::TaskKind,
-    tenant::storage_layer::InMemoryLayer,
-    virtual_file,
-};
-use pageserver_api::{key::Key, shard::TenantShardId};
-use utils::{
-    bin_ser::BeSer,
-    id::{TenantId, TimelineId},
-};
-
-// A very cheap hash for generating non-sequential keys.
-fn murmurhash32(mut h: u32) -> u32 {
-    h ^= h >> 16;
-    h = h.wrapping_mul(0x85ebca6b);
-    h ^= h >> 13;
-    h = h.wrapping_mul(0xc2b2ae35);
-    h ^= h >> 16;
-    h
-}
-
-enum KeyLayout {
-    /// Sequential unique keys
-    Sequential,
-    /// Random unique keys
-    Random,
-    /// Random keys, but only use the bits from the mask of them
-    RandomReuse(u32),
-}
-
-enum WriteDelta {
-    Yes,
-    No,
-}
-
-async fn ingest(
-    conf: &'static PageServerConf,
-    put_size: usize,
-    put_count: usize,
-    key_layout: KeyLayout,
-    write_delta: WriteDelta,
-) -> anyhow::Result<()> {
-    let mut lsn = utils::lsn::Lsn(1000);
-    let mut key = Key::from_i128(0x0);
-
-    let timeline_id = TimelineId::generate();
-    let tenant_id = TenantId::generate();
-    let tenant_shard_id = TenantShardId::unsharded(tenant_id);
-
-    tokio::fs::create_dir_all(conf.timeline_path(&tenant_shard_id, &timeline_id)).await?;
-
-    let ctx = RequestContext::new(TaskKind::DebugTool, DownloadBehavior::Error);
-
-    let gate = utils::sync::gate::Gate::default();
-    let entered = gate.enter().unwrap();
-
-    let layer =
-        InMemoryLayer::create(conf, timeline_id, tenant_shard_id, lsn, entered, &ctx).await?;
-
-    let data = Value::Image(Bytes::from(vec![0u8; put_size])).ser()?;
-    let ctx = RequestContext::new(
-        pageserver::task_mgr::TaskKind::WalReceiverConnectionHandler,
-        pageserver::context::DownloadBehavior::Download,
-    );
-
-    for i in 0..put_count {
-        lsn += put_size as u64;
-
-        // Generate lots of keys within a single relation, which simulates the typical bulk ingest case: people
-        // usually care the most about write performance when they're blasting a huge batch of data into a huge table.
-        match key_layout {
-            KeyLayout::Sequential => {
-                // Use sequential order to illustrate the experience a user is likely to have
-                // when ingesting bulk data.
-                key.field6 = i as u32;
-            }
-            KeyLayout::Random => {
-                // Use random-order keys to avoid giving a false advantage to data structures that are
-                // faster when inserting on the end.
-                key.field6 = murmurhash32(i as u32);
-            }
-            KeyLayout::RandomReuse(mask) => {
-                // Use low bits only, to limit cardinality
-                key.field6 = murmurhash32(i as u32) & mask;
-            }
-        }
-
-        layer.put_value(key, lsn, &data, &ctx).await?;
-    }
-    layer.freeze(lsn + 1).await;
-
-    if matches!(write_delta, WriteDelta::Yes) {
-        let l0_flush_state = L0FlushGlobalState::new(L0FlushConfig::Direct {
-            max_concurrency: NonZeroUsize::new(1).unwrap(),
-        });
-        let (_desc, path) = layer
-            .write_to_disk(&ctx, None, l0_flush_state.inner())
-            .await?
-            .unwrap();
-        tokio::fs::remove_file(path).await?;
-    }
-
-    Ok(())
-}
-
-/// Wrapper to instantiate a tokio runtime
-fn ingest_main(
-    conf: &'static PageServerConf,
-    put_size: usize,
-    put_count: usize,
-    key_layout: KeyLayout,
-    write_delta: WriteDelta,
-) {
-    let runtime = tokio::runtime::Builder::new_current_thread()
-        .enable_all()
-        .build()
-        .unwrap();
-
-    runtime.block_on(async move {
-        let r = ingest(conf, put_size, put_count, key_layout, write_delta).await;
-        if let Err(e) = r {
-            panic!("{e:?}");
-        }
-    });
-}
-
-/// Declare a series of benchmarks for the Pageserver's ingest write path.
-///
-/// This benchmark does not include WAL decode: it starts at InMemoryLayer::put_value, and ends either
-/// at freezing the ephemeral layer, or writing the ephemeral layer out to an L0 (depending on whether WriteDelta is set).
-///
-/// Genuine disk I/O is used, so expect results to differ depending on storage.  However, when running on
-/// a fast disk, CPU is the bottleneck at time of writing.
-fn criterion_benchmark(c: &mut Criterion) {
-    let temp_dir_parent: Utf8PathBuf = env::current_dir().unwrap().try_into().unwrap();
-    let temp_dir = camino_tempfile::tempdir_in(temp_dir_parent).unwrap();
-    eprintln!("Data directory: {}", temp_dir.path());
-
-    let conf: &'static PageServerConf = Box::leak(Box::new(
-        pageserver::config::PageServerConf::dummy_conf(temp_dir.path().to_path_buf()),
-    ));
-    virtual_file::init(16384, virtual_file::io_engine_for_bench());
-    page_cache::init(conf.page_cache_size);
-
-    {
-        let mut group = c.benchmark_group("ingest-small-values");
-        let put_size = 100usize;
-        let put_count = 128 * 1024 * 1024 / put_size;
-        group.throughput(criterion::Throughput::Bytes((put_size * put_count) as u64));
-        group.sample_size(10);
-        group.bench_function("ingest 128MB/100b seq", |b| {
-            b.iter(|| {
-                ingest_main(
-                    conf,
-                    put_size,
-                    put_count,
-                    KeyLayout::Sequential,
-                    WriteDelta::Yes,
-                )
-            })
-        });
-        group.bench_function("ingest 128MB/100b rand", |b| {
-            b.iter(|| {
-                ingest_main(
-                    conf,
-                    put_size,
-                    put_count,
-                    KeyLayout::Random,
-                    WriteDelta::Yes,
-                )
-            })
-        });
-        group.bench_function("ingest 128MB/100b rand-1024keys", |b| {
-            b.iter(|| {
-                ingest_main(
-                    conf,
-                    put_size,
-                    put_count,
-                    KeyLayout::RandomReuse(0x3ff),
-                    WriteDelta::Yes,
-                )
-            })
-        });
-        group.bench_function("ingest 128MB/100b seq, no delta", |b| {
-            b.iter(|| {
-                ingest_main(
-                    conf,
-                    put_size,
-                    put_count,
-                    KeyLayout::Sequential,
-                    WriteDelta::No,
-                )
-            })
-        });
-    }
-
-    {
-        let mut group = c.benchmark_group("ingest-big-values");
-        let put_size = 8192usize;
-        let put_count = 128 * 1024 * 1024 / put_size;
-        group.throughput(criterion::Throughput::Bytes((put_size * put_count) as u64));
-        group.sample_size(10);
-        group.bench_function("ingest 128MB/8k seq", |b| {
-            b.iter(|| {
-                ingest_main(
-                    conf,
-                    put_size,
-                    put_count,
-                    KeyLayout::Sequential,
-                    WriteDelta::Yes,
-                )
-            })
-        });
-        group.bench_function("ingest 128MB/8k seq, no delta", |b| {
-            b.iter(|| {
-                ingest_main(
-                    conf,
-                    put_size,
-                    put_count,
-                    KeyLayout::Sequential,
-                    WriteDelta::No,
-                )
-            })
-        });
-    }
-}
-
-criterion_group!(benches, criterion_benchmark);
-criterion_main!(benches);

pageserver/benches/bench_layer_map.rs

@@ -1,4 +1,3 @@
-use criterion::measurement::WallTime;
 use pageserver::keyspace::{KeyPartitioning, KeySpace};
 use pageserver::repository::Key;
 use pageserver::tenant::layer_map::LayerMap;
@@ -16,11 +15,7 @@ use utils::id::{TenantId, TimelineId};
 
 use utils::lsn::Lsn;
 
-use criterion::{black_box, criterion_group, criterion_main, BenchmarkGroup, Criterion};
-
-fn fixture_path(relative: &str) -> PathBuf {
-    PathBuf::from(env!("CARGO_MANIFEST_DIR")).join(relative)
-}
+use criterion::{black_box, criterion_group, criterion_main, Criterion};
 
 fn build_layer_map(filename_dump: PathBuf) -> LayerMap {
     let mut layer_map = LayerMap::default();
@@ -114,7 +109,7 @@ fn uniform_key_partitioning(layer_map: &LayerMap, _lsn: Lsn) -> KeyPartitioning
 // between each test run.
 fn bench_from_captest_env(c: &mut Criterion) {
     // TODO consider compressing this file
-    let layer_map = build_layer_map(fixture_path("benches/odd-brook-layernames.txt"));
+    let layer_map = build_layer_map(PathBuf::from("benches/odd-brook-layernames.txt"));
     let queries: Vec<(Key, Lsn)> = uniform_query_pattern(&layer_map);
 
     // Test with uniform query pattern
@@ -144,7 +139,7 @@ fn bench_from_captest_env(c: &mut Criterion) {
 fn bench_from_real_project(c: &mut Criterion) {
     // Init layer map
     let now = Instant::now();
-    let layer_map = build_layer_map(fixture_path("benches/odd-brook-layernames.txt"));
+    let layer_map = build_layer_map(PathBuf::from("benches/odd-brook-layernames.txt"));
     println!("Finished layer map init in {:?}", now.elapsed());
 
     // Choose uniformly distributed queries
@@ -247,72 +242,7 @@ fn bench_sequential(c: &mut Criterion) {
     group.finish();
 }
 
-fn bench_visibility_with_map(
-    group: &mut BenchmarkGroup<WallTime>,
-    layer_map: LayerMap,
-    read_points: Vec<Lsn>,
-    bench_name: &str,
-) {
-    group.bench_function(bench_name, |b| {
-        b.iter(|| black_box(layer_map.get_visibility(read_points.clone())));
-    });
-}
-
-// Benchmark using synthetic data. Arrange image layers on stacked diagonal lines.
-fn bench_visibility(c: &mut Criterion) {
-    let mut group = c.benchmark_group("visibility");
-    {
-        // Init layer map. Create 100_000 layers arranged in 1000 diagonal lines.
-        let now = Instant::now();
-        let mut layer_map = LayerMap::default();
-        let mut updates = layer_map.batch_update();
-        for i in 0..100_000 {
-            let i32 = (i as u32) % 100;
-            let zero = Key::from_hex("000000000000000000000000000000000000").unwrap();
-            let layer = PersistentLayerDesc::new_img(
-                TenantShardId::unsharded(TenantId::generate()),
-                TimelineId::generate(),
-                zero.add(10 * i32)..zero.add(10 * i32 + 1),
-                Lsn(i),
-                0,
-            );
-            updates.insert_historic(layer);
-        }
-        updates.flush();
-        println!("Finished layer map init in {:?}", now.elapsed());
-
-        let mut read_points = Vec::new();
-        for i in (0..100_000).step_by(1000) {
-            read_points.push(Lsn(i));
-        }
-
-        bench_visibility_with_map(&mut group, layer_map, read_points, "sequential");
-    }
-
-    {
-        let layer_map = build_layer_map(fixture_path("benches/odd-brook-layernames.txt"));
-        let read_points = vec![Lsn(0x1C760FA190)];
-        bench_visibility_with_map(&mut group, layer_map, read_points, "real_map");
-
-        let layer_map = build_layer_map(fixture_path("benches/odd-brook-layernames.txt"));
-        let read_points = vec![
-            Lsn(0x1C760FA190),
-            Lsn(0x000000931BEAD539),
-            Lsn(0x000000931BF63011),
-            Lsn(0x000000931B33AE68),
-            Lsn(0x00000038E67ABFA0),
-            Lsn(0x000000931B33AE68),
-            Lsn(0x000000914E3F38F0),
-            Lsn(0x000000931B33AE68),
-        ];
-        bench_visibility_with_map(&mut group, layer_map, read_points, "real_map_many_branches");
-    }
-
-    group.finish();
-}
-
 criterion_group!(group_1, bench_from_captest_env);
 criterion_group!(group_2, bench_from_real_project);
 criterion_group!(group_3, bench_sequential);
-criterion_group!(group_4, bench_visibility);
-criterion_main!(group_1, group_2, group_3, group_4);
+criterion_main!(group_1, group_2, group_3);

pageserver/client/src/mgmt_api.rs

@@ -1,7 +1,6 @@
 use std::collections::HashMap;
 
 use bytes::Bytes;
-use detach_ancestor::AncestorDetached;
 use pageserver_api::{models::*, shard::TenantShardId};
 use reqwest::{IntoUrl, Method, StatusCode};
 use utils::{
@@ -419,23 +418,6 @@ impl Client {
         }
     }
 
-    pub async fn timeline_detach_ancestor(
-        &self,
-        tenant_shard_id: TenantShardId,
-        timeline_id: TimelineId,
-    ) -> Result<AncestorDetached> {
-        let uri = format!(
-            "{}/v1/tenant/{tenant_shard_id}/timeline/{timeline_id}/detach_ancestor",
-            self.mgmt_api_endpoint
-        );
-
-        self.request(Method::PUT, &uri, ())
-            .await?
-            .json()
-            .await
-            .map_err(Error::ReceiveBody)
-    }
-
     pub async fn tenant_reset(&self, tenant_shard_id: TenantShardId) -> Result<()> {
         let uri = format!(
             "{}/v1/tenant/{}/reset",

pageserver/ctl/src/main.rs

@@ -179,7 +179,7 @@ async fn main() -> anyhow::Result<()> {
                 .get("remote_storage")
                 .expect("need remote_storage");
             let config = RemoteStorageConfig::from_toml(toml_item)?;
-            let storage = remote_storage::GenericRemoteStorage::from_config(&config).await;
+            let storage = remote_storage::GenericRemoteStorage::from_config(&config);
             let cancel = CancellationToken::new();
             storage
                 .unwrap()

pageserver/src/auth.rs

@@ -14,14 +14,12 @@ pub fn check_permission(claims: &Claims, tenant_id: Option<TenantId>) -> Result<
         }
         (Scope::PageServerApi, None) => Ok(()), // access to management api for PageServerApi scope
         (Scope::PageServerApi, Some(_)) => Ok(()), // access to tenant api using PageServerApi scope
-        (Scope::Admin | Scope::SafekeeperData | Scope::GenerationsApi | Scope::Scrubber, _) => {
-            Err(AuthError(
-                format!(
-                    "JWT scope '{:?}' is ineligible for Pageserver auth",
-                    claims.scope
-                )
-                .into(),
-            ))
-        }
+        (Scope::Admin | Scope::SafekeeperData | Scope::GenerationsApi, _) => Err(AuthError(
+            format!(
+                "JWT scope '{:?}' is ineligible for Pageserver auth",
+                claims.scope
+            )
+            .into(),
+        )),
     }
 }

pageserver/src/bin/pageserver.rs

@@ -2,36 +2,35 @@
 
 //! Main entry point for the Page Server executable.
 
-use std::env;
 use std::env::{var, VarError};
 use std::io::Read;
 use std::sync::Arc;
 use std::time::Duration;
+use std::{env, ops::ControlFlow, str::FromStr};
 
 use anyhow::{anyhow, Context};
 use camino::Utf8Path;
 use clap::{Arg, ArgAction, Command};
 
 use metrics::launch_timestamp::{set_launch_timestamp_metric, LaunchTimestamp};
-use pageserver::config::PageserverIdentity;
 use pageserver::control_plane_client::ControlPlaneClient;
 use pageserver::disk_usage_eviction_task::{self, launch_disk_usage_global_eviction_task};
 use pageserver::metrics::{STARTUP_DURATION, STARTUP_IS_LOADING};
-use pageserver::task_mgr::{COMPUTE_REQUEST_RUNTIME, WALRECEIVER_RUNTIME};
+use pageserver::task_mgr::WALRECEIVER_RUNTIME;
 use pageserver::tenant::{secondary, TenantSharedResources};
-use pageserver::{CancellableTask, ConsumptionMetricsTasks, HttpEndpointListener};
 use remote_storage::GenericRemoteStorage;
 use tokio::signal::unix::SignalKind;
 use tokio::time::Instant;
-use tokio_util::sync::CancellationToken;
 use tracing::*;
 
 use metrics::set_build_info_metric;
 use pageserver::{
-    config::PageServerConf,
+    config::{defaults::*, PageServerConf},
+    context::{DownloadBehavior, RequestContext},
     deletion_queue::DeletionQueue,
     http, page_cache, page_service, task_mgr,
-    task_mgr::{BACKGROUND_RUNTIME, MGMT_REQUEST_RUNTIME},
+    task_mgr::TaskKind,
+    task_mgr::{BACKGROUND_RUNTIME, COMPUTE_REQUEST_RUNTIME, MGMT_REQUEST_RUNTIME},
     tenant::mgr,
     virtual_file,
 };
@@ -85,13 +84,18 @@ fn main() -> anyhow::Result<()> {
         .with_context(|| format!("Error opening workdir '{workdir}'"))?;
 
     let cfg_file_path = workdir.join("pageserver.toml");
-    let identity_file_path = workdir.join("identity.toml");
 
     // Set CWD to workdir for non-daemon modes
     env::set_current_dir(&workdir)
         .with_context(|| format!("Failed to set application's current dir to '{workdir}'"))?;
 
-    let conf = initialize_config(&identity_file_path, &cfg_file_path, &workdir)?;
+    let conf = match initialize_config(&cfg_file_path, arg_matches, &workdir)? {
+        ControlFlow::Continue(conf) => conf,
+        ControlFlow::Break(()) => {
+            info!("Pageserver config init successful");
+            return Ok(());
+        }
+    };
 
     // Initialize logging.
     //
@@ -123,10 +127,8 @@ fn main() -> anyhow::Result<()> {
 
     // after setting up logging, log the effective IO engine choice and read path implementations
     info!(?conf.virtual_file_io_engine, "starting with virtual_file IO engine");
-    info!(?conf.virtual_file_direct_io, "starting with virtual_file Direct IO settings");
     info!(?conf.get_impl, "starting with get page implementation");
     info!(?conf.get_vectored_impl, "starting with vectored get page implementation");
-    info!(?conf.compact_level0_phase1_value_access, "starting with setting for compact_level0_phase1_value_access");
 
     let tenants_path = conf.tenants_path();
     if !tenants_path.exists() {
@@ -148,55 +150,70 @@ fn main() -> anyhow::Result<()> {
 }
 
 fn initialize_config(
-    identity_file_path: &Utf8Path,
     cfg_file_path: &Utf8Path,
+    arg_matches: clap::ArgMatches,
     workdir: &Utf8Path,
-) -> anyhow::Result<&'static PageServerConf> {
-    // The deployment orchestrator writes out an indentity file containing the node id
-    // for all pageservers. This file is the source of truth for the node id. In order
-    // to allow for rolling back pageserver releases, the node id is also included in
-    // the pageserver config that the deployment orchestrator writes to disk for the pageserver.
-    // A rolled back version of the pageserver will get the node id from the pageserver.toml
-    // config file.
-    let identity = match std::fs::File::open(identity_file_path) {
+) -> anyhow::Result<ControlFlow<(), &'static PageServerConf>> {
+    let init = arg_matches.get_flag("init");
+
+    let file_contents: Option<toml_edit::Document> = match std::fs::File::open(cfg_file_path) {
         Ok(mut f) => {
-            let md = f.metadata().context("stat config file")?;
-            if !md.is_file() {
-                anyhow::bail!("Pageserver found identity file but it is a dir entry: {identity_file_path}. Aborting start up ...");
+            if init {
+                anyhow::bail!("config file already exists: {cfg_file_path}");
             }
-
-            let mut s = String::new();
-            f.read_to_string(&mut s).context("read identity file")?;
-            toml_edit::de::from_str::<PageserverIdentity>(&s)?
-        }
-        Err(e) => {
-            anyhow::bail!("Pageserver could not read identity file: {identity_file_path}: {e}. Aborting start up ...");
-        }
-    };
-
-    let config: toml_edit::Document = match std::fs::File::open(cfg_file_path) {
-        Ok(mut f) => {
             let md = f.metadata().context("stat config file")?;
             if md.is_file() {
                 let mut s = String::new();
                 f.read_to_string(&mut s).context("read config file")?;
-                s.parse().context("parse config file toml")?
+                Some(s.parse().context("parse config file toml")?)
             } else {
                 anyhow::bail!("directory entry exists but is not a file: {cfg_file_path}");
             }
         }
+        Err(e) if e.kind() == std::io::ErrorKind::NotFound => None,
         Err(e) => {
             anyhow::bail!("open pageserver config: {e}: {cfg_file_path}");
         }
     };
 
-    debug!("Using pageserver toml: {config}");
+    let mut effective_config = file_contents.unwrap_or_else(|| {
+        DEFAULT_CONFIG_FILE
+            .parse()
+            .expect("unit tests ensure this works")
+    });
+
+    // Patch with overrides from the command line
+    if let Some(values) = arg_matches.get_many::<String>("config-override") {
+        for option_line in values {
+            let doc = toml_edit::Document::from_str(option_line).with_context(|| {
+                format!("Option '{option_line}' could not be parsed as a toml document")
+            })?;
+
+            for (key, item) in doc.iter() {
+                effective_config.insert(key, item.clone());
+            }
+        }
+    }
+
+    debug!("Resulting toml: {effective_config}");
 
     // Construct the runtime representation
-    let conf = PageServerConf::parse_and_validate(identity.id, &config, workdir)
+    let conf = PageServerConf::parse_and_validate(&effective_config, workdir)
         .context("Failed to parse pageserver configuration")?;
 
-    Ok(Box::leak(Box::new(conf)))
+    if init {
+        info!("Writing pageserver config to '{cfg_file_path}'");
+
+        std::fs::write(cfg_file_path, effective_config.to_string())
+            .with_context(|| format!("Failed to write pageserver config to '{cfg_file_path}'"))?;
+        info!("Config successfully written to '{cfg_file_path}'")
+    }
+
+    Ok(if init {
+        ControlFlow::Break(())
+    } else {
+        ControlFlow::Continue(Box::leak(Box::new(conf)))
+    })
 }
 
 struct WaitForPhaseResult<F: std::future::Future + Unpin> {
@@ -288,7 +305,6 @@ fn start_pageserver(
     // Create and lock PID file. This ensures that there cannot be more than one
     // pageserver process running at the same time.
     let lock_file_path = conf.workdir.join(PID_FILE_NAME);
-    info!("Claiming pid file at {lock_file_path:?}...");
     let lock_file =
         utils::pid_file::claim_for_current_process(&lock_file_path).context("claim pid file")?;
     info!("Claimed pid file at {lock_file_path:?}");
@@ -369,7 +385,7 @@ fn start_pageserver(
     let shutdown_pageserver = tokio_util::sync::CancellationToken::new();
 
     // Set up remote storage client
-    let remote_storage = BACKGROUND_RUNTIME.block_on(create_remote_storage_client(conf))?;
+    let remote_storage = create_remote_storage_client(conf)?;
 
     // Set up deletion queue
     let (deletion_queue, deletion_workers) = DeletionQueue::new(
@@ -414,10 +430,8 @@ fn start_pageserver(
 
     // Scan the local 'tenants/' directory and start loading the tenants
     let deletion_queue_client = deletion_queue.new_client();
-    let background_purges = mgr::BackgroundPurges::default();
     let tenant_manager = BACKGROUND_RUNTIME.block_on(mgr::init_tenant_mgr(
         conf,
-        background_purges.clone(),
         TenantSharedResources {
             broker_client: broker_client.clone(),
             remote_storage: remote_storage.clone(),
@@ -509,7 +523,7 @@ fn start_pageserver(
         }
     });
 
-    let (secondary_controller, secondary_controller_tasks) = secondary::spawn_tasks(
+    let secondary_controller = secondary::spawn_tasks(
         tenant_manager.clone(),
         remote_storage.clone(),
         background_jobs_barrier.clone(),
@@ -522,19 +536,18 @@ fn start_pageserver(
     // been configured.
     let disk_usage_eviction_state: Arc<disk_usage_eviction_task::State> = Arc::default();
 
-    let disk_usage_eviction_task = launch_disk_usage_global_eviction_task(
+    launch_disk_usage_global_eviction_task(
         conf,
         remote_storage.clone(),
         disk_usage_eviction_state.clone(),
         tenant_manager.clone(),
         background_jobs_barrier.clone(),
-    );
+    )?;
 
     // Start up the service to handle HTTP mgmt API request. We created the
     // listener earlier already.
-    let http_endpoint_listener = {
-        let _rt_guard = MGMT_REQUEST_RUNTIME.enter(); // for hyper
-        let cancel = CancellationToken::new();
+    {
+        let _rt_guard = MGMT_REQUEST_RUNTIME.enter();
 
         let router_state = Arc::new(
             http::routes::State::new(
@@ -555,49 +568,109 @@ fn start_pageserver(
         let service = utils::http::RouterService::new(router).unwrap();
         let server = hyper::Server::from_tcp(http_listener)?
             .serve(service)
-            .with_graceful_shutdown({
-                let cancel = cancel.clone();
-                async move { cancel.clone().cancelled().await }
-            });
+            .with_graceful_shutdown(task_mgr::shutdown_watcher());
 
-        let task = MGMT_REQUEST_RUNTIME.spawn(task_mgr::exit_on_panic_or_error(
+        task_mgr::spawn(
+            MGMT_REQUEST_RUNTIME.handle(),
+            TaskKind::HttpEndpointListener,
+            None,
+            None,
             "http endpoint listener",
-            server,
-        ));
-        HttpEndpointListener(CancellableTask { task, cancel })
-    };
+            true,
+            async {
+                server.await?;
+                Ok(())
+            },
+        );
+    }
 
-    let consumption_metrics_tasks = {
-        let cancel = shutdown_pageserver.child_token();
-        let task = crate::BACKGROUND_RUNTIME.spawn({
-            let tenant_manager = tenant_manager.clone();
-            let cancel = cancel.clone();
-            async move {
-                // first wait until background jobs are cleared to launch.
-                //
-                // this is because we only process active tenants and timelines, and the
-                // Timeline::get_current_logical_size will spawn the logical size calculation,
-                // which will not be rate-limited.
-                tokio::select! {
-                    _ = cancel.cancelled() => { return; },
-                    _ = background_jobs_barrier.wait() => {}
-                };
+    if let Some(metric_collection_endpoint) = &conf.metric_collection_endpoint {
+        let metrics_ctx = RequestContext::todo_child(
+            TaskKind::MetricsCollection,
+            // This task itself shouldn't download anything.
+            // The actual size calculation does need downloads, and
+            // creates a child context with the right DownloadBehavior.
+            DownloadBehavior::Error,
+        );
 
-                pageserver::consumption_metrics::run(conf, tenant_manager, cancel).await;
-            }
-        });
-        ConsumptionMetricsTasks(CancellableTask { task, cancel })
-    };
+        let local_disk_storage = conf.workdir.join("last_consumption_metrics.json");
+
+        task_mgr::spawn(
+            crate::BACKGROUND_RUNTIME.handle(),
+            TaskKind::MetricsCollection,
+            None,
+            None,
+            "consumption metrics collection",
+            true,
+            {
+                let tenant_manager = tenant_manager.clone();
+                async move {
+                    // first wait until background jobs are cleared to launch.
+                    //
+                    // this is because we only process active tenants and timelines, and the
+                    // Timeline::get_current_logical_size will spawn the logical size calculation,
+                    // which will not be rate-limited.
+                    let cancel = task_mgr::shutdown_token();
+
+                    tokio::select! {
+                        _ = cancel.cancelled() => { return Ok(()); },
+                        _ = background_jobs_barrier.wait() => {}
+                    };
+
+                    pageserver::consumption_metrics::collect_metrics(
+                        tenant_manager,
+                        metric_collection_endpoint,
+                        &conf.metric_collection_bucket,
+                        conf.metric_collection_interval,
+                        conf.cached_metric_collection_interval,
+                        conf.synthetic_size_calculation_interval,
+                        conf.id,
+                        local_disk_storage,
+                        cancel,
+                        metrics_ctx,
+                    )
+                    .instrument(info_span!("metrics_collection"))
+                    .await?;
+                    Ok(())
+                }
+            },
+        );
+    }
 
     // Spawn a task to listen for libpq connections. It will spawn further tasks
     // for each connection. We created the listener earlier already.
-    let page_service = page_service::spawn(conf, tenant_manager.clone(), pg_auth, {
-        let _entered = COMPUTE_REQUEST_RUNTIME.enter(); // TcpListener::from_std requires it
-        pageserver_listener
-            .set_nonblocking(true)
-            .context("set listener to nonblocking")?;
-        tokio::net::TcpListener::from_std(pageserver_listener).context("create tokio listener")?
-    });
+    {
+        let libpq_ctx = RequestContext::todo_child(
+            TaskKind::LibpqEndpointListener,
+            // listener task shouldn't need to download anything. (We will
+            // create a separate sub-contexts for each connection, with their
+            // own download behavior. This context is used only to listen and
+            // accept connections.)
+            DownloadBehavior::Error,
+        );
+        task_mgr::spawn(
+            COMPUTE_REQUEST_RUNTIME.handle(),
+            TaskKind::LibpqEndpointListener,
+            None,
+            None,
+            "libpq endpoint listener",
+            true,
+            {
+                let tenant_manager = tenant_manager.clone();
+                async move {
+                    page_service::libpq_listener_main(
+                        tenant_manager,
+                        pg_auth,
+                        pageserver_listener,
+                        conf.pg_auth_type,
+                        libpq_ctx,
+                        task_mgr::shutdown_token(),
+                    )
+                    .await
+                }
+            },
+        );
+    }
 
     let mut shutdown_pageserver = Some(shutdown_pageserver.drop_guard());
 
@@ -623,24 +696,13 @@ fn start_pageserver(
             // Right now that tree doesn't reach very far, and `task_mgr` is used instead.
             // The plan is to change that over time.
             shutdown_pageserver.take();
-            pageserver::shutdown_pageserver(
-                http_endpoint_listener,
-                page_service,
-                consumption_metrics_tasks,
-                disk_usage_eviction_task,
-                &tenant_manager,
-                background_purges,
-                deletion_queue.clone(),
-                secondary_controller_tasks,
-                0,
-            )
-            .await;
+            pageserver::shutdown_pageserver(&tenant_manager, deletion_queue.clone(), 0).await;
             unreachable!()
         })
     }
 }
 
-async fn create_remote_storage_client(
+fn create_remote_storage_client(
     conf: &'static PageServerConf,
 ) -> anyhow::Result<GenericRemoteStorage> {
     let config = if let Some(config) = &conf.remote_storage_config {
@@ -650,7 +712,7 @@ async fn create_remote_storage_client(
     };
 
     // Create the client
-    let mut remote_storage = GenericRemoteStorage::from_config(config).await?;
+    let mut remote_storage = GenericRemoteStorage::from_config(config)?;
 
     // If `test_remote_failures` is non-zero, wrap the client with a
     // wrapper that simulates failures.
@@ -673,12 +735,28 @@ fn cli() -> Command {
     Command::new("Neon page server")
         .about("Materializes WAL stream to pages and serves them to the postgres")
         .version(version())
+        .arg(
+            Arg::new("init")
+                .long("init")
+                .action(ArgAction::SetTrue)
+                .help("Initialize pageserver with all given config overrides"),
+        )
         .arg(
             Arg::new("workdir")
                 .short('D')
                 .long("workdir")
                 .help("Working directory for the pageserver"),
         )
+        // See `settings.md` for more details on the extra configuration patameters pageserver can process
+        .arg(
+            Arg::new("config-override")
+                .long("config-override")
+                .short('c')
+                .num_args(1)
+                .action(ArgAction::Append)
+                .help("Additional configuration overrides of the ones from the toml config file (or new ones to add there). \
+                Any option has to be a valid toml document, example: `-c=\"foo='hey'\"` `-c=\"foo={value=1}\"`"),
+        )
         .arg(
             Arg::new("enabled-features")
                 .long("enabled-features")

pageserver/src/config.rs

@@ -7,8 +7,8 @@
 use anyhow::{anyhow, bail, ensure, Context, Result};
 use pageserver_api::{models::ImageCompressionAlgorithm, shard::TenantShardId};
 use remote_storage::{RemotePath, RemoteStorageConfig};
+use serde;
 use serde::de::IntoDeserializer;
-use serde::{self, Deserialize};
 use std::env;
 use storage_broker::Uri;
 use utils::crashsafe::path_with_suffix_extension;
@@ -29,7 +29,6 @@ use utils::{
     logging::LogFormat,
 };
 
-use crate::tenant::timeline::compaction::CompactL0Phase1ValueAccess;
 use crate::tenant::vectored_blob_io::MaxVectoredReadBytes;
 use crate::tenant::{config::TenantConfOpt, timeline::GetImpl};
 use crate::tenant::{TENANTS_SEGMENT_NAME, TIMELINES_SEGMENT_NAME};
@@ -53,7 +52,7 @@ pub mod defaults {
     use pageserver_api::models::ImageCompressionAlgorithm;
     pub use storage_broker::DEFAULT_ENDPOINT as BROKER_DEFAULT_ENDPOINT;
 
-    pub const DEFAULT_WAIT_LSN_TIMEOUT: &str = "300 s";
+    pub const DEFAULT_WAIT_LSN_TIMEOUT: &str = "60 s";
     pub const DEFAULT_WAL_REDO_TIMEOUT: &str = "60 s";
 
     pub const DEFAULT_SUPERUSER: &str = "cloud_admin";
@@ -69,6 +68,7 @@ pub mod defaults {
         super::ConfigurableSemaphore::DEFAULT_INITIAL.get();
 
     pub const DEFAULT_METRIC_COLLECTION_INTERVAL: &str = "10 min";
+    pub const DEFAULT_CACHED_METRIC_COLLECTION_INTERVAL: &str = "0s";
     pub const DEFAULT_METRIC_COLLECTION_ENDPOINT: Option<reqwest::Url> = None;
     pub const DEFAULT_SYNTHETIC_SIZE_CALCULATION_INTERVAL: &str = "10 min";
     pub const DEFAULT_BACKGROUND_TASK_MAXIMUM_DELAY: &str = "10s";
@@ -84,16 +84,16 @@ pub mod defaults {
     #[cfg(not(target_os = "linux"))]
     pub const DEFAULT_VIRTUAL_FILE_IO_ENGINE: &str = "std-fs";
 
-    pub const DEFAULT_GET_VECTORED_IMPL: &str = "vectored";
+    pub const DEFAULT_GET_VECTORED_IMPL: &str = "sequential";
 
-    pub const DEFAULT_GET_IMPL: &str = "vectored";
+    pub const DEFAULT_GET_IMPL: &str = "legacy";
 
     pub const DEFAULT_MAX_VECTORED_READ_BYTES: usize = 128 * 1024; // 128 KiB
 
     pub const DEFAULT_IMAGE_COMPRESSION: ImageCompressionAlgorithm =
         ImageCompressionAlgorithm::Disabled;
 
-    pub const DEFAULT_VALIDATE_VECTORED_GET: bool = false;
+    pub const DEFAULT_VALIDATE_VECTORED_GET: bool = true;
 
     pub const DEFAULT_EPHEMERAL_BYTES_PER_MEMORY_KB: usize = 0;
 
@@ -123,6 +123,7 @@ pub mod defaults {
 #concurrent_tenant_warmup = '{DEFAULT_CONCURRENT_TENANT_WARMUP}'
 
 #metric_collection_interval = '{DEFAULT_METRIC_COLLECTION_INTERVAL}'
+#cached_metric_collection_interval = '{DEFAULT_CACHED_METRIC_COLLECTION_INTERVAL}'
 #synthetic_size_calculation_interval = '{DEFAULT_SYNTHETIC_SIZE_CALCULATION_INTERVAL}'
 
 #disk_usage_based_eviction = {{ max_usage_pct = .., min_avail_bytes = .., period = "10s"}}
@@ -237,6 +238,7 @@ pub struct PageServerConf {
     // How often to collect metrics and send them to the metrics endpoint.
     pub metric_collection_interval: Duration,
     // How often to send unchanged cached metrics to the metrics endpoint.
+    pub cached_metric_collection_interval: Duration,
     pub metric_collection_endpoint: Option<Url>,
     pub metric_collection_bucket: Option<RemoteStorageConfig>,
     pub synthetic_size_calculation_interval: Duration,
@@ -296,13 +298,6 @@ pub struct PageServerConf {
     pub ephemeral_bytes_per_memory_kb: usize,
 
     pub l0_flush: L0FlushConfig,
-
-    /// This flag is temporary and will be removed after gradual rollout.
-    /// See <https://github.com/neondatabase/neon/issues/8184>.
-    pub compact_level0_phase1_value_access: CompactL0Phase1ValueAccess,
-
-    /// Direct IO settings
-    pub virtual_file_direct_io: virtual_file::DirectIoMode,
 }
 
 /// We do not want to store this in a PageServerConf because the latter may be logged
@@ -364,6 +359,8 @@ struct PageServerConfigBuilder {
     auth_validation_public_key_path: BuilderValue<Option<Utf8PathBuf>>,
     remote_storage_config: BuilderValue<Option<RemoteStorageConfig>>,
 
+    id: BuilderValue<NodeId>,
+
     broker_endpoint: BuilderValue<Uri>,
     broker_keepalive_interval: BuilderValue<Duration>,
 
@@ -373,6 +370,7 @@ struct PageServerConfigBuilder {
     concurrent_tenant_size_logical_size_queries: BuilderValue<NonZeroUsize>,
 
     metric_collection_interval: BuilderValue<Duration>,
+    cached_metric_collection_interval: BuilderValue<Duration>,
     metric_collection_endpoint: BuilderValue<Option<Url>>,
     synthetic_size_calculation_interval: BuilderValue<Duration>,
     metric_collection_bucket: BuilderValue<Option<RemoteStorageConfig>>,
@@ -409,17 +407,9 @@ struct PageServerConfigBuilder {
     ephemeral_bytes_per_memory_kb: BuilderValue<usize>,
 
     l0_flush: BuilderValue<L0FlushConfig>,
-
-    compact_level0_phase1_value_access: BuilderValue<CompactL0Phase1ValueAccess>,
-
-    virtual_file_direct_io: BuilderValue<virtual_file::DirectIoMode>,
 }
 
 impl PageServerConfigBuilder {
-    fn new() -> Self {
-        Self::default()
-    }
-
     #[inline(always)]
     fn default_values() -> Self {
         use self::BuilderValue::*;
@@ -445,6 +435,7 @@ impl PageServerConfigBuilder {
             pg_auth_type: Set(AuthType::Trust),
             auth_validation_public_key_path: Set(None),
             remote_storage_config: Set(None),
+            id: NotSet,
             broker_endpoint: Set(storage_broker::DEFAULT_ENDPOINT
                 .parse()
                 .expect("failed to parse default broker endpoint")),
@@ -463,6 +454,10 @@ impl PageServerConfigBuilder {
                 DEFAULT_METRIC_COLLECTION_INTERVAL,
             )
             .expect("cannot parse default metric collection interval")),
+            cached_metric_collection_interval: Set(humantime::parse_duration(
+                DEFAULT_CACHED_METRIC_COLLECTION_INTERVAL,
+            )
+            .expect("cannot parse default cached_metric_collection_interval")),
             synthetic_size_calculation_interval: Set(humantime::parse_duration(
                 DEFAULT_SYNTHETIC_SIZE_CALCULATION_INTERVAL,
             )
@@ -502,8 +497,6 @@ impl PageServerConfigBuilder {
             validate_vectored_get: Set(DEFAULT_VALIDATE_VECTORED_GET),
             ephemeral_bytes_per_memory_kb: Set(DEFAULT_EPHEMERAL_BYTES_PER_MEMORY_KB),
             l0_flush: Set(L0FlushConfig::default()),
-            compact_level0_phase1_value_access: Set(CompactL0Phase1ValueAccess::default()),
-            virtual_file_direct_io: Set(virtual_file::DirectIoMode::default()),
         }
     }
 }
@@ -576,6 +569,10 @@ impl PageServerConfigBuilder {
         self.broker_keepalive_interval = BuilderValue::Set(broker_keepalive_interval)
     }
 
+    pub fn id(&mut self, node_id: NodeId) {
+        self.id = BuilderValue::Set(node_id)
+    }
+
     pub fn log_format(&mut self, log_format: LogFormat) {
         self.log_format = BuilderValue::Set(log_format)
     }
@@ -592,6 +589,14 @@ impl PageServerConfigBuilder {
         self.metric_collection_interval = BuilderValue::Set(metric_collection_interval)
     }
 
+    pub fn cached_metric_collection_interval(
+        &mut self,
+        cached_metric_collection_interval: Duration,
+    ) {
+        self.cached_metric_collection_interval =
+            BuilderValue::Set(cached_metric_collection_interval)
+    }
+
     pub fn metric_collection_endpoint(&mut self, metric_collection_endpoint: Option<Url>) {
         self.metric_collection_endpoint = BuilderValue::Set(metric_collection_endpoint)
     }
@@ -687,15 +692,7 @@ impl PageServerConfigBuilder {
         self.l0_flush = BuilderValue::Set(value);
     }
 
-    pub fn compact_level0_phase1_value_access(&mut self, value: CompactL0Phase1ValueAccess) {
-        self.compact_level0_phase1_value_access = BuilderValue::Set(value);
-    }
-
-    pub fn virtual_file_direct_io(&mut self, value: virtual_file::DirectIoMode) {
-        self.virtual_file_direct_io = BuilderValue::Set(value);
-    }
-
-    pub fn build(self, id: NodeId) -> anyhow::Result<PageServerConf> {
+    pub fn build(self) -> anyhow::Result<PageServerConf> {
         let default = Self::default_values();
 
         macro_rules! conf {
@@ -728,10 +725,12 @@ impl PageServerConfigBuilder {
                 pg_auth_type,
                 auth_validation_public_key_path,
                 remote_storage_config,
+                id,
                 broker_endpoint,
                 broker_keepalive_interval,
                 log_format,
                 metric_collection_interval,
+                cached_metric_collection_interval,
                 metric_collection_endpoint,
                 metric_collection_bucket,
                 synthetic_size_calculation_interval,
@@ -752,12 +751,9 @@ impl PageServerConfigBuilder {
                 image_compression,
                 ephemeral_bytes_per_memory_kb,
                 l0_flush,
-                compact_level0_phase1_value_access,
-                virtual_file_direct_io,
             }
             CUSTOM LOGIC
             {
-                id: id,
                 // TenantConf is handled separately
                 default_tenant_conf: TenantConf::default(),
                 concurrent_tenant_warmup: ConfigurableSemaphore::new({
@@ -902,12 +898,8 @@ impl PageServerConf {
     /// validating the input and failing on errors.
     ///
     /// This leaves any options not present in the file in the built-in defaults.
-    pub fn parse_and_validate(
-        node_id: NodeId,
-        toml: &Document,
-        workdir: &Utf8Path,
-    ) -> anyhow::Result<Self> {
-        let mut builder = PageServerConfigBuilder::new();
+    pub fn parse_and_validate(toml: &Document, workdir: &Utf8Path) -> anyhow::Result<Self> {
+        let mut builder = PageServerConfigBuilder::default();
         builder.workdir(workdir.to_owned());
 
         let mut t_conf = TenantConfOpt::default();
@@ -938,6 +930,7 @@ impl PageServerConf {
                 "tenant_config" => {
                     t_conf = TenantConfOpt::try_from(item.to_owned()).context(format!("failed to parse: '{key}'"))?;
                 }
+                "id" => builder.id(NodeId(parse_toml_u64(key, item)?)),
                 "broker_endpoint" => builder.broker_endpoint(parse_toml_string(key, item)?.parse().context("failed to parse broker endpoint")?),
                 "broker_keepalive_interval" => builder.broker_keepalive_interval(parse_toml_duration(key, item)?),
                 "log_format" => builder.log_format(
@@ -954,6 +947,7 @@ impl PageServerConf {
                     NonZeroUsize::new(permits).context("initial semaphore permits out of range: 0, use other configuration to disable a feature")?
                 }),
                 "metric_collection_interval" => builder.metric_collection_interval(parse_toml_duration(key, item)?),
+                "cached_metric_collection_interval" => builder.cached_metric_collection_interval(parse_toml_duration(key, item)?),
                 "metric_collection_endpoint" => {
                     let endpoint = parse_toml_string(key, item)?.parse().context("failed to parse metric_collection_endpoint")?;
                     builder.metric_collection_endpoint(Some(endpoint));
@@ -1026,17 +1020,11 @@ impl PageServerConf {
                 "l0_flush" => {
                     builder.l0_flush(utils::toml_edit_ext::deserialize_item(item).context("l0_flush")?)
                 }
-                "compact_level0_phase1_value_access" => {
-                    builder.compact_level0_phase1_value_access(utils::toml_edit_ext::deserialize_item(item).context("compact_level0_phase1_value_access")?)
-                }
-                "virtual_file_direct_io" => {
-                    builder.virtual_file_direct_io(utils::toml_edit_ext::deserialize_item(item).context("virtual_file_direct_io")?)
-                }
                 _ => bail!("unrecognized pageserver option '{key}'"),
             }
         }
 
-        let mut conf = builder.build(node_id).context("invalid config")?;
+        let mut conf = builder.build().context("invalid config")?;
 
         if conf.http_auth_type == AuthType::NeonJWT || conf.pg_auth_type == AuthType::NeonJWT {
             let auth_validation_public_key_path = conf
@@ -1092,6 +1080,7 @@ impl PageServerConf {
             eviction_task_immitated_concurrent_logical_size_queries: ConfigurableSemaphore::default(
             ),
             metric_collection_interval: Duration::from_secs(60),
+            cached_metric_collection_interval: Duration::from_secs(60 * 60),
             metric_collection_endpoint: defaults::DEFAULT_METRIC_COLLECTION_ENDPOINT,
             metric_collection_bucket: None,
             synthetic_size_calculation_interval: Duration::from_secs(60),
@@ -1116,18 +1105,10 @@ impl PageServerConf {
             validate_vectored_get: defaults::DEFAULT_VALIDATE_VECTORED_GET,
             ephemeral_bytes_per_memory_kb: defaults::DEFAULT_EPHEMERAL_BYTES_PER_MEMORY_KB,
             l0_flush: L0FlushConfig::default(),
-            compact_level0_phase1_value_access: CompactL0Phase1ValueAccess::default(),
-            virtual_file_direct_io: virtual_file::DirectIoMode::default(),
         }
     }
 }
 
-#[derive(Deserialize)]
-#[serde(deny_unknown_fields)]
-pub struct PageserverIdentity {
-    pub id: NodeId,
-}
-
 // Helper functions to parse a toml Item
 
 fn parse_toml_string(name: &str, item: &Item) -> Result<String> {
@@ -1275,8 +1256,10 @@ max_file_descriptors = 333
 
 # initial superuser role name to use when creating a new tenant
 initial_superuser_name = 'zzzz'
+id = 10
 
 metric_collection_interval = '222 s'
+cached_metric_collection_interval = '22200 s'
 metric_collection_endpoint = 'http://localhost:80/metrics'
 synthetic_size_calculation_interval = '333 s'
 
@@ -1291,11 +1274,12 @@ background_task_maximum_delay = '334 s'
         let (workdir, pg_distrib_dir) = prepare_fs(&tempdir)?;
         let broker_endpoint = storage_broker::DEFAULT_ENDPOINT;
         // we have to create dummy values to overcome the validation errors
-        let config_string =
-            format!("pg_distrib_dir='{pg_distrib_dir}'\nbroker_endpoint = '{broker_endpoint}'",);
+        let config_string = format!(
+            "pg_distrib_dir='{pg_distrib_dir}'\nid=10\nbroker_endpoint = '{broker_endpoint}'",
+        );
         let toml = config_string.parse()?;
 
-        let parsed_config = PageServerConf::parse_and_validate(NodeId(10), &toml, &workdir)
+        let parsed_config = PageServerConf::parse_and_validate(&toml, &workdir)
             .unwrap_or_else(|e| panic!("Failed to parse config '{config_string}', reason: {e:?}"));
 
         assert_eq!(
@@ -1331,6 +1315,9 @@ background_task_maximum_delay = '334 s'
                 metric_collection_interval: humantime::parse_duration(
                     defaults::DEFAULT_METRIC_COLLECTION_INTERVAL
                 )?,
+                cached_metric_collection_interval: humantime::parse_duration(
+                    defaults::DEFAULT_CACHED_METRIC_COLLECTION_INTERVAL
+                )?,
                 metric_collection_endpoint: defaults::DEFAULT_METRIC_COLLECTION_ENDPOINT,
                 metric_collection_bucket: None,
                 synthetic_size_calculation_interval: humantime::parse_duration(
@@ -1359,8 +1346,6 @@ background_task_maximum_delay = '334 s'
                 image_compression: defaults::DEFAULT_IMAGE_COMPRESSION,
                 ephemeral_bytes_per_memory_kb: defaults::DEFAULT_EPHEMERAL_BYTES_PER_MEMORY_KB,
                 l0_flush: L0FlushConfig::default(),
-                compact_level0_phase1_value_access: CompactL0Phase1ValueAccess::default(),
-                virtual_file_direct_io: virtual_file::DirectIoMode::default(),
             },
             "Correct defaults should be used when no config values are provided"
         );
@@ -1379,7 +1364,7 @@ background_task_maximum_delay = '334 s'
         );
         let toml = config_string.parse()?;
 
-        let parsed_config = PageServerConf::parse_and_validate(NodeId(10), &toml, &workdir)
+        let parsed_config = PageServerConf::parse_and_validate(&toml, &workdir)
             .unwrap_or_else(|e| panic!("Failed to parse config '{config_string}', reason: {e:?}"));
 
         assert_eq!(
@@ -1411,6 +1396,7 @@ background_task_maximum_delay = '334 s'
                 eviction_task_immitated_concurrent_logical_size_queries:
                     ConfigurableSemaphore::default(),
                 metric_collection_interval: Duration::from_secs(222),
+                cached_metric_collection_interval: Duration::from_secs(22200),
                 metric_collection_endpoint: Some(Url::parse("http://localhost:80/metrics")?),
                 metric_collection_bucket: None,
                 synthetic_size_calculation_interval: Duration::from_secs(333),
@@ -1435,8 +1421,6 @@ background_task_maximum_delay = '334 s'
                 image_compression: defaults::DEFAULT_IMAGE_COMPRESSION,
                 ephemeral_bytes_per_memory_kb: defaults::DEFAULT_EPHEMERAL_BYTES_PER_MEMORY_KB,
                 l0_flush: L0FlushConfig::default(),
-                compact_level0_phase1_value_access: CompactL0Phase1ValueAccess::default(),
-                virtual_file_direct_io: virtual_file::DirectIoMode::default(),
             },
             "Should be able to parse all basic config values correctly"
         );
@@ -1471,13 +1455,12 @@ broker_endpoint = '{broker_endpoint}'
 
             let toml = config_string.parse()?;
 
-            let parsed_remote_storage_config =
-                PageServerConf::parse_and_validate(NodeId(10), &toml, &workdir)
-                    .unwrap_or_else(|e| {
-                        panic!("Failed to parse config '{config_string}', reason: {e:?}")
-                    })
-                    .remote_storage_config
-                    .expect("Should have remote storage config for the local FS");
+            let parsed_remote_storage_config = PageServerConf::parse_and_validate(&toml, &workdir)
+                .unwrap_or_else(|e| {
+                    panic!("Failed to parse config '{config_string}', reason: {e:?}")
+                })
+                .remote_storage_config
+                .expect("Should have remote storage config for the local FS");
 
             assert_eq!(
                 parsed_remote_storage_config,
@@ -1533,13 +1516,12 @@ broker_endpoint = '{broker_endpoint}'
 
             let toml = config_string.parse()?;
 
-            let parsed_remote_storage_config =
-                PageServerConf::parse_and_validate(NodeId(10), &toml, &workdir)
-                    .unwrap_or_else(|e| {
-                        panic!("Failed to parse config '{config_string}', reason: {e:?}")
-                    })
-                    .remote_storage_config
-                    .expect("Should have remote storage config for S3");
+            let parsed_remote_storage_config = PageServerConf::parse_and_validate(&toml, &workdir)
+                .unwrap_or_else(|e| {
+                    panic!("Failed to parse config '{config_string}', reason: {e:?}")
+                })
+                .remote_storage_config
+                .expect("Should have remote storage config for S3");
 
             assert_eq!(
                 parsed_remote_storage_config,
@@ -1601,6 +1583,7 @@ broker_endpoint = '{broker_endpoint}'
             r#"pg_distrib_dir = "{pg_distrib_dir}"
 metric_collection_endpoint = "http://sample.url"
 metric_collection_interval = "10min"
+id = 222
 
 [disk_usage_based_eviction]
 max_usage_pct = 80
@@ -1617,7 +1600,7 @@ threshold = "20m"
 "#,
         );
         let toml: Document = pageserver_conf_toml.parse()?;
-        let conf = PageServerConf::parse_and_validate(NodeId(333), &toml, &workdir)?;
+        let conf = PageServerConf::parse_and_validate(&toml, &workdir)?;
 
         assert_eq!(conf.pg_distrib_dir, pg_distrib_dir);
         assert_eq!(
@@ -1633,11 +1616,7 @@ threshold = "20m"
                 .evictions_low_residence_duration_metric_threshold,
             Duration::from_secs(20 * 60)
         );
-
-        // Assert that the node id provided by the indentity file (threaded
-        // through the call to [`PageServerConf::parse_and_validate`] is
-        // used.
-        assert_eq!(conf.id, NodeId(333));
+        assert_eq!(conf.id, NodeId(222));
         assert_eq!(
             conf.disk_usage_based_eviction,
             Some(DiskUsageEvictionTaskConfig {
@@ -1646,7 +1625,7 @@ threshold = "20m"
                 period: Duration::from_secs(10),
                 #[cfg(feature = "testing")]
                 mock_statvfs: None,
-                eviction_order: Default::default(),
+                eviction_order: crate::disk_usage_eviction_task::EvictionOrder::AbsoluteAccessed,
             })
         );
 
@@ -1670,6 +1649,7 @@ threshold = "20m"
             r#"pg_distrib_dir = "{pg_distrib_dir}"
 metric_collection_endpoint = "http://sample.url"
 metric_collection_interval = "10min"
+id = 222
 
 [tenant_config]
 evictions_low_residence_duration_metric_threshold = "20m"
@@ -1681,7 +1661,7 @@ threshold = "20m"
 "#,
         );
         let toml: Document = pageserver_conf_toml.parse().unwrap();
-        let conf = PageServerConf::parse_and_validate(NodeId(222), &toml, &workdir).unwrap();
+        let conf = PageServerConf::parse_and_validate(&toml, &workdir).unwrap();
 
         match &conf.default_tenant_conf.eviction_policy {
             EvictionPolicy::OnlyImitiate(t) => {
@@ -1700,7 +1680,7 @@ threshold = "20m"
 remote_storage = {}
         "#;
         let doc = toml_edit::Document::from_str(input).unwrap();
-        let err = PageServerConf::parse_and_validate(NodeId(222), &doc, &workdir)
+        let err = PageServerConf::parse_and_validate(&doc, &workdir)
             .expect_err("empty remote_storage field should fail, don't specify it if you want no remote_storage");
         assert!(format!("{err}").contains("remote_storage"), "{err}");
     }

pageserver/src/consumption_metrics.rs

@@ -1,6 +1,5 @@
 //! Periodically collect consumption metrics for all active tenants
 //! and push them to a HTTP endpoint.
-use crate::config::PageServerConf;
 use crate::context::{DownloadBehavior, RequestContext};
 use crate::task_mgr::{self, TaskKind, BACKGROUND_RUNTIME};
 use crate::tenant::size::CalculateSyntheticSizeError;
@@ -40,74 +39,56 @@ type RawMetric = (MetricsKey, (EventType, u64));
 /// for deduplication, but that is no longer needed.
 type Cache = HashMap<MetricsKey, (EventType, u64)>;
 
-pub async fn run(
-    conf: &'static PageServerConf,
-    tenant_manager: Arc<TenantManager>,
-    cancel: CancellationToken,
-) {
-    let Some(metric_collection_endpoint) = conf.metric_collection_endpoint.as_ref() else {
-        return;
-    };
-
-    let local_disk_storage = conf.workdir.join("last_consumption_metrics.json");
-
-    let metrics_ctx = RequestContext::todo_child(
-        TaskKind::MetricsCollection,
-        // This task itself shouldn't download anything.
-        // The actual size calculation does need downloads, and
-        // creates a child context with the right DownloadBehavior.
-        DownloadBehavior::Error,
-    );
-    let collect_metrics = BACKGROUND_RUNTIME.spawn(task_mgr::exit_on_panic_or_error(
-        "consumption metrics collection",
-        collect_metrics(
-            tenant_manager.clone(),
-            metric_collection_endpoint,
-            &conf.metric_collection_bucket,
-            conf.metric_collection_interval,
-            conf.id,
-            local_disk_storage,
-            cancel.clone(),
-            metrics_ctx,
-        )
-        .instrument(info_span!("metrics_collection")),
-    ));
-
-    let worker_ctx =
-        RequestContext::todo_child(TaskKind::CalculateSyntheticSize, DownloadBehavior::Download);
-    let synthetic_size_worker = BACKGROUND_RUNTIME.spawn(task_mgr::exit_on_panic_or_error(
-        "synthetic size calculation",
-        calculate_synthetic_size_worker(
-            tenant_manager.clone(),
-            conf.synthetic_size_calculation_interval,
-            cancel.clone(),
-            worker_ctx,
-        )
-        .instrument(info_span!("synthetic_size_worker")),
-    ));
-
-    let (collect_metrics, synthetic_size_worker) =
-        futures::future::join(collect_metrics, synthetic_size_worker).await;
-    collect_metrics
-        .expect("unreachable: exit_on_panic_or_error would catch the panic and exit the process");
-    synthetic_size_worker
-        .expect("unreachable: exit_on_panic_or_error would catch the panic and exit the process");
-}
-
 /// Main thread that serves metrics collection
 #[allow(clippy::too_many_arguments)]
-async fn collect_metrics(
+pub async fn collect_metrics(
     tenant_manager: Arc<TenantManager>,
     metric_collection_endpoint: &Url,
     metric_collection_bucket: &Option<RemoteStorageConfig>,
     metric_collection_interval: Duration,
+    _cached_metric_collection_interval: Duration,
+    synthetic_size_calculation_interval: Duration,
     node_id: NodeId,
     local_disk_storage: Utf8PathBuf,
     cancel: CancellationToken,
     ctx: RequestContext,
 ) -> anyhow::Result<()> {
+    if _cached_metric_collection_interval != Duration::ZERO {
+        tracing::warn!(
+            "cached_metric_collection_interval is no longer used, please set it to zero."
+        )
+    }
+
+    // spin up background worker that caclulates tenant sizes
+    let worker_ctx =
+        ctx.detached_child(TaskKind::CalculateSyntheticSize, DownloadBehavior::Download);
+    task_mgr::spawn(
+        BACKGROUND_RUNTIME.handle(),
+        TaskKind::CalculateSyntheticSize,
+        None,
+        None,
+        "synthetic size calculation",
+        false,
+        {
+            let tenant_manager = tenant_manager.clone();
+            async move {
+                calculate_synthetic_size_worker(
+                    tenant_manager,
+                    synthetic_size_calculation_interval,
+                    &cancel,
+                    &worker_ctx,
+                )
+                .instrument(info_span!("synthetic_size_worker"))
+                .await?;
+                Ok(())
+            }
+        },
+    );
+
     let path: Arc<Utf8PathBuf> = Arc::new(local_disk_storage);
 
+    let cancel = task_mgr::shutdown_token();
+
     let restore_and_reschedule = restore_and_reschedule(&path, metric_collection_interval);
 
     let mut cached_metrics = tokio::select! {
@@ -122,7 +103,7 @@ async fn collect_metrics(
         .expect("Failed to create http client with timeout");
 
     let bucket_client = if let Some(bucket_config) = metric_collection_bucket {
-        match GenericRemoteStorage::from_config(bucket_config).await {
+        match GenericRemoteStorage::from_config(bucket_config) {
             Ok(client) => Some(client),
             Err(e) => {
                 // Non-fatal error: if we were given an invalid config, we will proceed
@@ -194,9 +175,11 @@ async fn collect_metrics(
             BackgroundLoopKind::ConsumptionMetricsCollectMetrics,
         );
 
-        let res =
-            tokio::time::timeout_at(started_at + metric_collection_interval, cancel.cancelled())
-                .await;
+        let res = tokio::time::timeout_at(
+            started_at + metric_collection_interval,
+            task_mgr::shutdown_token().cancelled(),
+        )
+        .await;
         if res.is_ok() {
             return Ok(());
         }
@@ -296,8 +279,8 @@ async fn reschedule(
 async fn calculate_synthetic_size_worker(
     tenant_manager: Arc<TenantManager>,
     synthetic_size_calculation_interval: Duration,
-    cancel: CancellationToken,
-    ctx: RequestContext,
+    cancel: &CancellationToken,
+    ctx: &RequestContext,
 ) -> anyhow::Result<()> {
     info!("starting calculate_synthetic_size_worker");
     scopeguard::defer! {
@@ -337,7 +320,7 @@ async fn calculate_synthetic_size_worker(
             // there is never any reason to exit calculate_synthetic_size_worker following any
             // return value -- we don't need to care about shutdown because no tenant is found when
             // pageserver is shut down.
-            calculate_and_log(&tenant, &cancel, &ctx).await;
+            calculate_and_log(&tenant, cancel, ctx).await;
         }
 
         crate::tenant::tasks::warn_when_period_overrun(

pageserver/src/control_plane_client.rs

@@ -171,14 +171,14 @@ impl ControlPlaneGenerationsApi for ControlPlaneClient {
             register,
         };
 
+        fail::fail_point!("control-plane-client-re-attach");
+
         let response: ReAttachResponse = self.retry_http_forever(&re_attach_path, request).await?;
         tracing::info!(
             "Received re-attach response with {} tenants",
             response.tenants.len()
         );
 
-        failpoint_support::sleep_millis_async!("control-plane-client-re-attach");
-
         Ok(response
             .tenants
             .into_iter()

pageserver/src/deletion_queue.rs

@@ -828,9 +828,9 @@ mod test {
         }
     }
 
-    async fn setup(test_name: &str) -> anyhow::Result<TestSetup> {
+    fn setup(test_name: &str) -> anyhow::Result<TestSetup> {
         let test_name = Box::leak(Box::new(format!("deletion_queue__{test_name}")));
-        let harness = TenantHarness::create(test_name).await?;
+        let harness = TenantHarness::create(test_name)?;
 
         // We do not load() the harness: we only need its config and remote_storage
 
@@ -844,9 +844,7 @@ mod test {
             },
             timeout: RemoteStorageConfig::DEFAULT_TIMEOUT,
         };
-        let storage = GenericRemoteStorage::from_config(&storage_config)
-            .await
-            .unwrap();
+        let storage = GenericRemoteStorage::from_config(&storage_config).unwrap();
 
         let mock_control_plane = MockControlPlane::new();
 
@@ -924,9 +922,7 @@ mod test {
     #[tokio::test]
     async fn deletion_queue_smoke() -> anyhow::Result<()> {
         // Basic test that the deletion queue processes the deletions we pass into it
-        let ctx = setup("deletion_queue_smoke")
-            .await
-            .expect("Failed test setup");
+        let ctx = setup("deletion_queue_smoke").expect("Failed test setup");
         let client = ctx.deletion_queue.new_client();
         client.recover(HashMap::new())?;
 
@@ -996,9 +992,7 @@ mod test {
 
     #[tokio::test]
     async fn deletion_queue_validation() -> anyhow::Result<()> {
-        let ctx = setup("deletion_queue_validation")
-            .await
-            .expect("Failed test setup");
+        let ctx = setup("deletion_queue_validation").expect("Failed test setup");
         let client = ctx.deletion_queue.new_client();
         client.recover(HashMap::new())?;
 
@@ -1057,9 +1051,7 @@ mod test {
     #[tokio::test]
     async fn deletion_queue_recovery() -> anyhow::Result<()> {
         // Basic test that the deletion queue processes the deletions we pass into it
-        let mut ctx = setup("deletion_queue_recovery")
-            .await
-            .expect("Failed test setup");
+        let mut ctx = setup("deletion_queue_recovery").expect("Failed test setup");
         let client = ctx.deletion_queue.new_client();
         client.recover(HashMap::new())?;
 

pageserver/src/disk_usage_eviction_task.rs

@@ -59,14 +59,13 @@ use utils::{completion, id::TimelineId};
 use crate::{
     config::PageServerConf,
     metrics::disk_usage_based_eviction::METRICS,
-    task_mgr::{self, BACKGROUND_RUNTIME},
+    task_mgr::{self, TaskKind, BACKGROUND_RUNTIME},
     tenant::{
         mgr::TenantManager,
         remote_timeline_client::LayerFileMetadata,
         secondary::SecondaryTenant,
         storage_layer::{AsLayerDesc, EvictionError, Layer, LayerName},
     },
-    CancellableTask, DiskUsageEvictionTask,
 };
 
 #[derive(Debug, Clone, PartialEq, Eq, Serialize, Deserialize)]
@@ -84,9 +83,17 @@ pub struct DiskUsageEvictionTaskConfig {
 
 /// Selects the sort order for eviction candidates *after* per tenant `min_resident_size`
 /// partitioning.
-#[derive(Debug, Clone, Copy, PartialEq, Eq, Serialize, Deserialize)]
+#[derive(Default, Debug, Clone, Copy, PartialEq, Eq, Serialize, Deserialize)]
 #[serde(tag = "type", content = "args")]
 pub enum EvictionOrder {
+    /// Order the layers to be evicted by how recently they have been accessed in absolute
+    /// time.
+    ///
+    /// This strategy is unfair when some tenants grow faster than others towards the slower
+    /// growing.
+    #[default]
+    AbsoluteAccessed,
+
     /// Order the layers to be evicted by how recently they have been accessed relatively within
     /// the set of resident layers of a tenant.
     RelativeAccessed {
@@ -101,14 +108,6 @@ pub enum EvictionOrder {
     },
 }
 
-impl Default for EvictionOrder {
-    fn default() -> Self {
-        Self::RelativeAccessed {
-            highest_layer_count_loses_first: true,
-        }
-    }
-}
-
 fn default_highest_layer_count_loses_first() -> bool {
     true
 }
@@ -118,6 +117,11 @@ impl EvictionOrder {
         use EvictionOrder::*;
 
         match self {
+            AbsoluteAccessed => {
+                candidates.sort_unstable_by_key(|(partition, candidate)| {
+                    (*partition, candidate.last_activity_ts)
+                });
+            }
             RelativeAccessed { .. } => candidates.sort_unstable_by_key(|(partition, candidate)| {
                 (*partition, candidate.relative_last_activity)
             }),
@@ -130,6 +134,7 @@ impl EvictionOrder {
         use EvictionOrder::*;
 
         match self {
+            AbsoluteAccessed => finite_f32::FiniteF32::ZERO,
             RelativeAccessed {
                 highest_layer_count_loses_first,
             } => {
@@ -187,34 +192,36 @@ pub fn launch_disk_usage_global_eviction_task(
     state: Arc<State>,
     tenant_manager: Arc<TenantManager>,
     background_jobs_barrier: completion::Barrier,
-) -> Option<DiskUsageEvictionTask> {
+) -> anyhow::Result<()> {
     let Some(task_config) = &conf.disk_usage_based_eviction else {
         info!("disk usage based eviction task not configured");
-        return None;
+        return Ok(());
     };
 
     info!("launching disk usage based eviction task");
 
-    let cancel = CancellationToken::new();
-    let task = BACKGROUND_RUNTIME.spawn(task_mgr::exit_on_panic_or_error(
+    task_mgr::spawn(
+        BACKGROUND_RUNTIME.handle(),
+        TaskKind::DiskUsageEviction,
+        None,
+        None,
         "disk usage based eviction",
-        {
-            let cancel = cancel.clone();
-            async move {
-                // wait until initial load is complete, because we cannot evict from loading tenants.
-                tokio::select! {
-                    _ = cancel.cancelled() => { return anyhow::Ok(()); },
-                    _ = background_jobs_barrier.wait() => { }
-                };
+        false,
+        async move {
+            let cancel = task_mgr::shutdown_token();
 
-                disk_usage_eviction_task(&state, task_config, &storage, tenant_manager, cancel)
-                    .await;
-                anyhow::Ok(())
-            }
+            // wait until initial load is complete, because we cannot evict from loading tenants.
+            tokio::select! {
+                _ = cancel.cancelled() => { return Ok(()); },
+                _ = background_jobs_barrier.wait() => { }
+            };
+
+            disk_usage_eviction_task(&state, task_config, &storage, tenant_manager, cancel).await;
+            Ok(())
         },
-    ));
+    );
 
-    Some(DiskUsageEvictionTask(CancellableTask { cancel, task }))
+    Ok(())
 }
 
 #[instrument(skip_all)]

pageserver/src/http/openapi_spec.yml

@@ -308,45 +308,6 @@ paths:
             application/json:
               schema:
                 type: string
-
-  /v1/tenant/{tenant_shard_id}/timeline/{timeline_id}/block_gc:
-    parameters:
-      - name: tenant_shard_id
-        in: path
-        required: true
-        schema:
-          type: string
-      - name: timeline_id
-        in: path
-        required: true
-        schema:
-          type: string
-          format: hex
-    post:
-      description: Persistently add a gc blocking at the tenant level because of this timeline
-      responses:
-        "200":
-          description: OK
-
-  /v1/tenant/{tenant_shard_id}/timeline/{timeline_id}/unblock_gc:
-    parameters:
-      - name: tenant_shard_id
-        in: path
-        required: true
-        schema:
-          type: string
-      - name: timeline_id
-        in: path
-        required: true
-        schema:
-          type: string
-          format: hex
-    post:
-      description: Persistently remove a tenant level gc blocking for this timeline
-      responses:
-        "200":
-          description: OK
-
   /v1/tenant/{tenant_shard_id}/location_config:
     parameters:
       - name: tenant_shard_id
@@ -416,7 +377,7 @@ paths:
               schema:
                 $ref: "#/components/schemas/ConflictError"
 
-  /v1/tenant/{tenant_id}/timeline/{timeline_id}/preserve_initdb_archive:
+  /v1/tenant/{tenant_id}/{timeline_id}/preserve_initdb_archive:
     parameters:
       - name: tenant_id
         in: path
@@ -436,51 +397,6 @@ paths:
         "202":
           description: Tenant scheduled to load successfully
 
-  /v1/tenant/{tenant_shard_id}/timeline/{timeline_id}/archival_config:
-    parameters:
-      - name: tenant_shard_id
-        in: path
-        required: true
-        schema:
-          type: string
-      - name: timeline_id
-        in: path
-        required: true
-        schema:
-          type: string
-    put:
-      description: |
-        Either archives or unarchives the given timeline.
-        An archived timeline may not have any non-archived children.
-      requestBody:
-        required: true
-        content:
-          application/json:
-            schema:
-              $ref: "#/components/schemas/ArchivalConfigRequest"
-      responses:
-        "200":
-          description: Timeline (un)archived successfully
-        "409":
-          description: |
-            The tenant/timeline is already being modified, perhaps by a concurrent call to this API
-          content:
-            application/json:
-              schema:
-                $ref: "#/components/schemas/ConflictError"
-        "500":
-          description: Generic operation error
-          content:
-            application/json:
-              schema:
-                $ref: "#/components/schemas/Error"
-        "503":
-          description: Temporarily unavailable, please retry.
-          content:
-            application/json:
-              schema:
-                $ref: "#/components/schemas/ServiceUnavailableError"
-
   /v1/tenant/{tenant_id}/synthetic_size:
     parameters:
       - name: tenant_id
@@ -513,9 +429,7 @@ paths:
               schema:
                 $ref: "#/components/schemas/SyntheticSizeResponse"
             text/html:
-              schema:
-                type: string
-                description: SVG representation of the tenant and its timelines.
+              description: SVG representation of the tenant and it's timelines.
         "401":
           description: Unauthorized Error
           content:
@@ -654,7 +568,7 @@ paths:
           type: string
       - name: timeline_id
         in: path
-        required: true
+        ŕequired: true
         schema:
           type: string
 
@@ -860,13 +774,15 @@ components:
     TenantCreateRequest:
       allOf:
         - $ref: '#/components/schemas/TenantConfig'
-        - $ref: '#/components/schemas/TenantLoadRequest'
         - type: object
           required:
             - new_tenant_id
           properties:
             new_tenant_id:
               type: string
+            generation:
+              type: integer
+              description: Attachment generation number.
     TenantLoadRequest:
       type: object
       properties:
@@ -930,15 +846,6 @@ components:
         warm:
           type: boolean
           description: Whether to poll remote storage for layers to download.  If false, secondary locations don't download anything.
-    ArchivalConfigRequest:
-      type: object
-      required:
-        - state
-      properties:
-        state:
-          description: The archival state of a timeline
-          type: string
-          enum: ["Archived", "Unarchived"]
     TenantConfig:
       type: object
       properties:
@@ -1199,7 +1106,7 @@ components:
         reparented_timelines:
           type: array
           description: Set of reparented timeline ids
-          items:
+          properties:
             type: string
             format: hex
             description: TimelineId

pageserver/src/http/routes.rs

@@ -18,17 +18,14 @@ use hyper::StatusCode;
 use hyper::{Body, Request, Response, Uri};
 use metrics::launch_timestamp::LaunchTimestamp;
 use pageserver_api::models::AuxFilePolicy;
-use pageserver_api::models::DownloadRemoteLayersTaskSpawnRequest;
 use pageserver_api::models::IngestAuxFilesRequest;
 use pageserver_api::models::ListAuxFilesRequest;
 use pageserver_api::models::LocationConfig;
 use pageserver_api::models::LocationConfigListResponse;
-use pageserver_api::models::LocationConfigMode;
 use pageserver_api::models::LsnLease;
 use pageserver_api::models::LsnLeaseRequest;
 use pageserver_api::models::ShardParameters;
 use pageserver_api::models::TenantDetails;
-use pageserver_api::models::TenantLocationConfigRequest;
 use pageserver_api::models::TenantLocationConfigResponse;
 use pageserver_api::models::TenantScanRemoteStorageResponse;
 use pageserver_api::models::TenantScanRemoteStorageShard;
@@ -36,10 +33,12 @@ use pageserver_api::models::TenantShardLocation;
 use pageserver_api::models::TenantShardSplitRequest;
 use pageserver_api::models::TenantShardSplitResponse;
 use pageserver_api::models::TenantSorting;
-use pageserver_api::models::TimelineArchivalConfigRequest;
 use pageserver_api::models::TopTenantShardItem;
 use pageserver_api::models::TopTenantShardsRequest;
 use pageserver_api::models::TopTenantShardsResponse;
+use pageserver_api::models::{
+    DownloadRemoteLayersTaskSpawnRequest, LocationConfigMode, TenantLocationConfigRequest,
+};
 use pageserver_api::shard::ShardCount;
 use pageserver_api::shard::TenantShardId;
 use remote_storage::DownloadError;
@@ -296,11 +295,6 @@ impl From<GetActiveTenantError> for ApiError {
             GetActiveTenantError::WaitForActiveTimeout { .. } => {
                 ApiError::ResourceUnavailable(format!("{}", e).into())
             }
-            GetActiveTenantError::SwitchedTenant => {
-                // in our HTTP handlers, this error doesn't happen
-                // TODO: separate error types
-                ApiError::ResourceUnavailable("switched tenant".into())
-            }
         }
     }
 }
@@ -670,39 +664,6 @@ async fn timeline_preserve_initdb_handler(
     json_response(StatusCode::OK, ())
 }
 
-async fn timeline_archival_config_handler(
-    mut request: Request<Body>,
-    _cancel: CancellationToken,
-) -> Result<Response<Body>, ApiError> {
-    let tenant_shard_id: TenantShardId = parse_request_param(&request, "tenant_shard_id")?;
-    let timeline_id: TimelineId = parse_request_param(&request, "timeline_id")?;
-
-    let request_data: TimelineArchivalConfigRequest = json_request(&mut request).await?;
-    check_permission(&request, Some(tenant_shard_id.tenant_id))?;
-    let state = get_state(&request);
-
-    async {
-        let tenant = state
-            .tenant_manager
-            .get_attached_tenant_shard(tenant_shard_id)?;
-
-        tenant
-            .apply_timeline_archival_config(timeline_id, request_data.state)
-            .await
-            .context("applying archival config")
-            .map_err(ApiError::InternalServerError)?;
-        Ok::<_, ApiError>(())
-    }
-    .instrument(info_span!("timeline_archival_config",
-                tenant_id = %tenant_shard_id.tenant_id,
-                shard_id = %tenant_shard_id.shard_slug(),
-                state = ?request_data.state,
-                %timeline_id))
-    .await?;
-
-    json_response(StatusCode::OK, ())
-}
-
 async fn timeline_detail_handler(
     request: Request<Body>,
     _cancel: CancellationToken,
@@ -935,7 +896,6 @@ async fn tenant_list_handler(
             generation: (*gen)
                 .into()
                 .expect("Tenants are always attached with a generation"),
-            gc_blocking: None,
         })
         .collect::<Vec<TenantInfo>>();
 
@@ -987,7 +947,6 @@ async fn tenant_status(
                     .generation()
                     .into()
                     .expect("Tenants are always attached with a generation"),
-                gc_blocking: tenant.gc_block.summary().map(|x| format!("{x:?}")),
             },
             walredo: tenant.wal_redo_manager_status(),
             timelines: tenant.list_timeline_ids(),
@@ -1162,10 +1121,7 @@ async fn layer_map_info_handler(
     let timeline =
         active_timeline_of_active_tenant(&state.tenant_manager, tenant_shard_id, timeline_id)
             .await?;
-    let layer_map_info = timeline
-        .layer_map_info(reset)
-        .await
-        .map_err(|_shutdown| ApiError::ShuttingDown)?;
+    let layer_map_info = timeline.layer_map_info(reset).await;
 
     json_response(StatusCode::OK, layer_map_info)
 }
@@ -1231,72 +1187,6 @@ async fn evict_timeline_layer_handler(
     }
 }
 
-async fn timeline_gc_blocking_handler(
-    request: Request<Body>,
-    _cancel: CancellationToken,
-) -> Result<Response<Body>, ApiError> {
-    block_or_unblock_gc(request, true).await
-}
-
-async fn timeline_gc_unblocking_handler(
-    request: Request<Body>,
-    _cancel: CancellationToken,
-) -> Result<Response<Body>, ApiError> {
-    block_or_unblock_gc(request, false).await
-}
-
-/// Adding a block is `POST ../block_gc`, removing a block is `POST ../unblock_gc`.
-///
-/// Both are technically unsafe because they might fire off index uploads, thus they are POST.
-async fn block_or_unblock_gc(
-    request: Request<Body>,
-    block: bool,
-) -> Result<Response<Body>, ApiError> {
-    use crate::tenant::{
-        remote_timeline_client::WaitCompletionError, upload_queue::NotInitialized,
-    };
-    let tenant_shard_id: TenantShardId = parse_request_param(&request, "tenant_shard_id")?;
-    check_permission(&request, Some(tenant_shard_id.tenant_id))?;
-    let timeline_id: TimelineId = parse_request_param(&request, "timeline_id")?;
-    let state = get_state(&request);
-
-    let tenant = state
-        .tenant_manager
-        .get_attached_tenant_shard(tenant_shard_id)?;
-
-    tenant.wait_to_become_active(ACTIVE_TENANT_TIMEOUT).await?;
-
-    let timeline = tenant.get_timeline(timeline_id, true)?;
-
-    let fut = async {
-        if block {
-            timeline.block_gc(&tenant).await.map(|_| ())
-        } else {
-            timeline.unblock_gc(&tenant).await
-        }
-    };
-
-    let span = tracing::info_span!(
-        "block_or_unblock_gc",
-        tenant_id = %tenant_shard_id.tenant_id,
-        shard_id = %tenant_shard_id.shard_slug(),
-        timeline_id = %timeline_id,
-        block = block,
-    );
-
-    let res = fut.instrument(span).await;
-
-    res.map_err(|e| {
-        if e.is::<NotInitialized>() || e.is::<WaitCompletionError>() {
-            ApiError::ShuttingDown
-        } else {
-            ApiError::InternalServerError(e)
-        }
-    })?;
-
-    json_response(StatusCode::OK, ())
-}
-
 /// Get tenant_size SVG graph along with the JSON data.
 fn synthetic_size_html_response(
     inputs: ModelInputs,
@@ -1726,9 +1616,7 @@ async fn timeline_compact_handler(
             .await
             .map_err(|e| ApiError::InternalServerError(e.into()))?;
         if wait_until_uploaded {
-            timeline.remote_client.wait_completion().await
-            // XXX map to correct ApiError for the cases where it's due to shutdown
-            .context("wait completion").map_err(ApiError::InternalServerError)?;
+            timeline.remote_client.wait_completion().await.map_err(ApiError::InternalServerError)?;
         }
         json_response(StatusCode::OK, ())
     }
@@ -1754,10 +1642,6 @@ async fn timeline_checkpoint_handler(
     if Some(true) == parse_query_param::<_, bool>(&request, "force_image_layer_creation")? {
         flags |= CompactFlags::ForceImageLayerCreation;
     }
-
-    // By default, checkpoints come with a compaction, but this may be optionally disabled by tests that just want to flush + upload.
-    let compact = parse_query_param::<_, bool>(&request, "compact")?.unwrap_or(true);
-
     let wait_until_uploaded =
         parse_query_param::<_, bool>(&request, "wait_until_uploaded")?.unwrap_or(false);
 
@@ -1774,22 +1658,18 @@ async fn timeline_checkpoint_handler(
 
                 }
             })?;
-        if compact {
-            timeline
-                .compact(&cancel, flags, &ctx)
-                .await
-                .map_err(|e|
-                    match e {
-                        CompactionError::ShuttingDown => ApiError::ShuttingDown,
-                        CompactionError::Other(e) => ApiError::InternalServerError(e)
-                    }
-                )?;
-        }
+        timeline
+            .compact(&cancel, flags, &ctx)
+            .await
+            .map_err(|e|
+                match e {
+                    CompactionError::ShuttingDown => ApiError::ShuttingDown,
+                    CompactionError::Other(e) => ApiError::InternalServerError(e)
+                }
+            )?;
 
         if wait_until_uploaded {
-            timeline.remote_client.wait_completion().await
-            // XXX map to correct ApiError for the cases where it's due to shutdown
-            .context("wait completion").map_err(ApiError::InternalServerError)?;
+            timeline.remote_client.wait_completion().await.map_err(ApiError::InternalServerError)?;
         }
 
         json_response(StatusCode::OK, ())
@@ -1841,9 +1721,7 @@ async fn timeline_detach_ancestor_handler(
     request: Request<Body>,
     _cancel: CancellationToken,
 ) -> Result<Response<Body>, ApiError> {
-    use crate::tenant::timeline::detach_ancestor;
-    use pageserver_api::models::detach_ancestor::AncestorDetached;
-
+    use crate::tenant::timeline::detach_ancestor::Options;
     let tenant_shard_id: TenantShardId = parse_request_param(&request, "tenant_shard_id")?;
     check_permission(&request, Some(tenant_shard_id.tenant_id))?;
     let timeline_id: TimelineId = parse_request_param(&request, "timeline_id")?;
@@ -1851,7 +1729,7 @@ async fn timeline_detach_ancestor_handler(
     let span = tracing::info_span!("detach_ancestor", tenant_id=%tenant_shard_id.tenant_id, shard_id=%tenant_shard_id.shard_slug(), %timeline_id);
 
     async move {
-        let mut options = detach_ancestor::Options::default();
+        let mut options = Options::default();
 
         let rewrite_concurrency =
             parse_query_param::<_, std::num::NonZeroUsize>(&request, "rewrite_concurrency")?;
@@ -1879,36 +1757,27 @@ async fn timeline_detach_ancestor_handler(
 
         let timeline = tenant.get_timeline(timeline_id, true)?;
 
-        let progress = timeline
+        let (_guard, prepared) = timeline
             .prepare_to_detach_from_ancestor(&tenant, options, ctx)
             .await?;
 
-        // uncomment to allow early as possible Tenant::drop
-        // drop(tenant);
+        let res = state
+            .tenant_manager
+            .complete_detaching_timeline_ancestor(tenant_shard_id, timeline_id, prepared, ctx)
+            .await;
 
-        let resp = match progress {
-            detach_ancestor::Progress::Prepared(_guard, prepared) => {
-                // it would be great to tag the guard on to the tenant activation future
-                let reparented_timelines = state
-                    .tenant_manager
-                    .complete_detaching_timeline_ancestor(
-                        tenant_shard_id,
-                        timeline_id,
-                        prepared,
-                        ctx,
-                    )
-                    .await
-                    .context("timeline detach ancestor completion")
-                    .map_err(ApiError::InternalServerError)?;
-
-                AncestorDetached {
+        match res {
+            Ok(reparented_timelines) => {
+                let resp = pageserver_api::models::detach_ancestor::AncestorDetached {
                     reparented_timelines,
-                }
-            }
-            detach_ancestor::Progress::Done(resp) => resp,
-        };
+                };
 
-        json_response(StatusCode::OK, resp)
+                json_response(StatusCode::OK, resp)
+            }
+            Err(e) => Err(ApiError::InternalServerError(
+                e.context("timeline detach completion"),
+            )),
+        }
     }
     .instrument(span)
     .await
@@ -2205,24 +2074,14 @@ async fn secondary_download_handler(
 
     let timeout = wait.unwrap_or(Duration::MAX);
 
-    let result = tokio::time::timeout(
+    let status = match tokio::time::timeout(
         timeout,
         state.secondary_controller.download_tenant(tenant_shard_id),
     )
-    .await;
-
-    let progress = secondary_tenant.progress.lock().unwrap().clone();
-
-    let status = match result {
-        Ok(Ok(())) => {
-            if progress.layers_downloaded >= progress.layers_total {
-                // Download job ran to completion
-                StatusCode::OK
-            } else {
-                // Download dropped out without errors because it ran out of time budget
-                StatusCode::ACCEPTED
-            }
-        }
+    .await
+    {
+        // Download job ran to completion.
+        Ok(Ok(())) => StatusCode::OK,
         // Edge case: downloads aren't usually fallible: things like a missing heatmap are considered
         // okay.  We could get an error here in the unlikely edge case that the tenant
         // was detached between our check above and executing the download job.
@@ -2232,6 +2091,8 @@ async fn secondary_download_handler(
         Err(_) => StatusCode::ACCEPTED,
     };
 
+    let progress = secondary_tenant.progress.lock().unwrap().clone();
+
     json_response(status, progress)
 }
 
@@ -2917,10 +2778,6 @@ pub fn make_router(
             "/v1/tenant/:tenant_shard_id/timeline/:timeline_id/preserve_initdb_archive",
             |r| api_handler(r, timeline_preserve_initdb_handler),
         )
-        .post(
-            "/v1/tenant/:tenant_shard_id/timeline/:timeline_id/archival_config",
-            |r| api_handler(r, timeline_archival_config_handler),
-        )
         .get("/v1/tenant/:tenant_shard_id/timeline/:timeline_id", |r| {
             api_handler(r, timeline_detail_handler)
         })
@@ -2975,14 +2832,6 @@ pub fn make_router(
             "/v1/tenant/:tenant_shard_id/timeline/:timeline_id/layer/:layer_file_name",
             |r| api_handler(r, evict_timeline_layer_handler),
         )
-        .post(
-            "/v1/tenant/:tenant_shard_id/timeline/:timeline_id/block_gc",
-            |r| api_handler(r, timeline_gc_blocking_handler),
-        )
-        .post(
-            "/v1/tenant/:tenant_shard_id/timeline/:timeline_id/unblock_gc",
-            |r| api_handler(r, timeline_gc_unblocking_handler),
-        )
         .post("/v1/tenant/:tenant_shard_id/heatmap_upload", |r| {
             api_handler(r, secondary_upload_handler)
         })

pageserver/src/l0_flush.rs

@@ -2,29 +2,19 @@ use std::{num::NonZeroUsize, sync::Arc};
 
 use crate::tenant::ephemeral_file;
 
-#[derive(Debug, PartialEq, Eq, Clone, serde::Deserialize)]
+#[derive(Default, Debug, PartialEq, Eq, Clone, serde::Deserialize)]
 #[serde(tag = "mode", rename_all = "kebab-case", deny_unknown_fields)]
 pub enum L0FlushConfig {
+    #[default]
     PageCached,
     #[serde(rename_all = "snake_case")]
-    Direct {
-        max_concurrency: NonZeroUsize,
-    },
-}
-
-impl Default for L0FlushConfig {
-    fn default() -> Self {
-        Self::Direct {
-            // TODO: using num_cpus results in different peak memory usage on different instance types.
-            max_concurrency: NonZeroUsize::new(usize::max(1, num_cpus::get())).unwrap(),
-        }
-    }
+    Direct { max_concurrency: NonZeroUsize },
 }
 
 #[derive(Clone)]
 pub struct L0FlushGlobalState(Arc<Inner>);
 
-pub enum Inner {
+pub(crate) enum Inner {
     PageCached,
     Direct { semaphore: tokio::sync::Semaphore },
 }
@@ -40,7 +30,7 @@ impl L0FlushGlobalState {
         }
     }
 
-    pub fn inner(&self) -> &Arc<Inner> {
+    pub(crate) fn inner(&self) -> &Arc<Inner> {
         &self.0
     }
 }

pageserver/src/lib.rs

@@ -12,10 +12,7 @@ pub mod disk_usage_eviction_task;
 pub mod http;
 pub mod import_datadir;
 pub mod l0_flush;
-
-use futures::{stream::FuturesUnordered, StreamExt};
 pub use pageserver_api::keyspace;
-use tokio_util::sync::CancellationToken;
 pub mod aux_file;
 pub mod metrics;
 pub mod page_cache;
@@ -32,13 +29,11 @@ pub mod walingest;
 pub mod walrecord;
 pub mod walredo;
 
+use crate::task_mgr::TaskKind;
 use camino::Utf8Path;
 use deletion_queue::DeletionQueue;
-use tenant::{
-    mgr::{BackgroundPurges, TenantManager},
-    secondary,
-};
-use tracing::{info, info_span};
+use tenant::mgr::TenantManager;
+use tracing::info;
 
 /// Current storage format version
 ///
@@ -59,111 +54,17 @@ static ZERO_PAGE: bytes::Bytes = bytes::Bytes::from_static(&[0u8; 8192]);
 
 pub use crate::metrics::preinitialize_metrics;
 
-pub struct CancellableTask {
-    pub task: tokio::task::JoinHandle<()>,
-    pub cancel: CancellationToken,
-}
-pub struct HttpEndpointListener(pub CancellableTask);
-pub struct ConsumptionMetricsTasks(pub CancellableTask);
-pub struct DiskUsageEvictionTask(pub CancellableTask);
-impl CancellableTask {
-    pub async fn shutdown(self) {
-        self.cancel.cancel();
-        self.task.await.unwrap();
-    }
-}
-
 #[tracing::instrument(skip_all, fields(%exit_code))]
-#[allow(clippy::too_many_arguments)]
 pub async fn shutdown_pageserver(
-    http_listener: HttpEndpointListener,
-    page_service: page_service::Listener,
-    consumption_metrics_worker: ConsumptionMetricsTasks,
-    disk_usage_eviction_task: Option<DiskUsageEvictionTask>,
     tenant_manager: &TenantManager,
-    background_purges: BackgroundPurges,
     mut deletion_queue: DeletionQueue,
-    secondary_controller_tasks: secondary::GlobalTasks,
     exit_code: i32,
 ) {
     use std::time::Duration;
-
-    // If the orderly shutdown below takes too long, we still want to make
-    // sure that all walredo processes are killed and wait()ed on by us, not systemd.
-    //
-    // (Leftover walredo processes are the hypothesized trigger for the systemd freezes
-    //  that we keep seeing in prod => https://github.com/neondatabase/cloud/issues/11387.
-    //
-    // We use a thread instead of a tokio task because the background runtime is likely busy
-    // with the final flushing / uploads. This activity here has priority, and due to lack
-    // of scheduling priority feature sin the tokio scheduler, using a separate thread is
-    // an effective priority booster.
-    let walredo_extraordinary_shutdown_thread_span = {
-        let span = info_span!(parent: None, "walredo_extraordinary_shutdown_thread");
-        span.follows_from(tracing::Span::current());
-        span
-    };
-    let walredo_extraordinary_shutdown_thread_cancel = CancellationToken::new();
-    let walredo_extraordinary_shutdown_thread = std::thread::spawn({
-        let walredo_extraordinary_shutdown_thread_cancel =
-            walredo_extraordinary_shutdown_thread_cancel.clone();
-        move || {
-            let rt = tokio::runtime::Builder::new_current_thread()
-                .enable_all()
-                .build()
-                .unwrap();
-            let _entered = rt.enter();
-            let _entered = walredo_extraordinary_shutdown_thread_span.enter();
-            if let Ok(()) = rt.block_on(tokio::time::timeout(
-                Duration::from_secs(8),
-                walredo_extraordinary_shutdown_thread_cancel.cancelled(),
-            )) {
-                info!("cancellation requested");
-                return;
-            }
-            let managers = tenant::WALREDO_MANAGERS
-                .lock()
-                .unwrap()
-                // prevents new walredo managers from being inserted
-                .take()
-                .expect("only we take()");
-            // Use FuturesUnordered to get in queue early for each manager's
-            // heavier_once_cell semaphore wait list.
-            // Also, for idle tenants that for some reason haven't
-            // shut down yet, it's quite likely that we're not going
-            // to get Poll::Pending once.
-            let mut futs: FuturesUnordered<_> = managers
-                .into_iter()
-                .filter_map(|(_, mgr)| mgr.upgrade())
-                .map(|mgr| async move { tokio::task::unconstrained(mgr.shutdown()).await })
-                .collect();
-            info!(count=%futs.len(), "built FuturesUnordered");
-            let mut last_log_at = std::time::Instant::now();
-            #[derive(Debug, Default)]
-            struct Results {
-                initiated: u64,
-                already: u64,
-            }
-            let mut results = Results::default();
-            while let Some(we_initiated) = rt.block_on(futs.next()) {
-                if we_initiated {
-                    results.initiated += 1;
-                } else {
-                    results.already += 1;
-                }
-                if last_log_at.elapsed() > Duration::from_millis(100) {
-                    info!(remaining=%futs.len(), ?results, "progress");
-                    last_log_at = std::time::Instant::now();
-                }
-            }
-            info!(?results, "done");
-        }
-    });
-
     // Shut down the libpq endpoint task. This prevents new connections from
     // being accepted.
-    let remaining_connections = timed(
-        page_service.stop_accepting(),
+    timed(
+        task_mgr::shutdown_tasks(Some(TaskKind::LibpqEndpointListener), None, None),
         "shutdown LibpqEndpointListener",
         Duration::from_secs(1),
     )
@@ -181,7 +82,7 @@ pub async fn shutdown_pageserver(
     // Shut down any page service tasks: any in-progress work for particular timelines or tenants
     // should already have been canclled via mgr::shutdown_all_tenants
     timed(
-        remaining_connections.shutdown(),
+        task_mgr::shutdown_tasks(Some(TaskKind::PageRequestHandler), None, None),
         "shutdown PageRequestHandlers",
         Duration::from_secs(1),
     )
@@ -190,44 +91,16 @@ pub async fn shutdown_pageserver(
     // Best effort to persist any outstanding deletions, to avoid leaking objects
     deletion_queue.shutdown(Duration::from_secs(5)).await;
 
-    timed(
-        consumption_metrics_worker.0.shutdown(),
-        "shutdown consumption metrics",
-        Duration::from_secs(1),
-    )
-    .await;
-
-    timed(
-        futures::future::OptionFuture::from(disk_usage_eviction_task.map(|t| t.0.shutdown())),
-        "shutdown disk usage eviction",
-        Duration::from_secs(1),
-    )
-    .await;
-
-    timed(
-        background_purges.shutdown(),
-        "shutdown background purges",
-        Duration::from_secs(1),
-    )
-    .await;
-
     // Shut down the HTTP endpoint last, so that you can still check the server's
     // status while it's shutting down.
     // FIXME: We should probably stop accepting commands like attach/detach earlier.
     timed(
-        http_listener.0.shutdown(),
+        task_mgr::shutdown_tasks(Some(TaskKind::HttpEndpointListener), None, None),
         "shutdown http",
         Duration::from_secs(1),
     )
     .await;
 
-    timed(
-        secondary_controller_tasks.wait(), // cancellation happened in caller
-        "secondary controller wait",
-        Duration::from_secs(1),
-    )
-    .await;
-
     // There should be nothing left, but let's be sure
     timed(
         task_mgr::shutdown_tasks(None, None, None),
@@ -235,12 +108,6 @@ pub async fn shutdown_pageserver(
         Duration::from_secs(1),
     )
     .await;
-
-    info!("cancel & join walredo_extraordinary_shutdown_thread");
-    walredo_extraordinary_shutdown_thread_cancel.cancel();
-    walredo_extraordinary_shutdown_thread.join().unwrap();
-    info!("walredo_extraordinary_shutdown_thread done");
-
     info!("Shut down successfully completed");
     std::process::exit(exit_code);
 }

pageserver/src/metrics.rs

@@ -473,31 +473,6 @@ static PITR_HISTORY_SIZE: Lazy<UIntGaugeVec> = Lazy::new(|| {
     .expect("failed to define a metric")
 });
 
-#[derive(strum_macros::EnumString, strum_macros::Display, strum_macros::IntoStaticStr)]
-#[strum(serialize_all = "kebab_case")]
-pub(crate) enum MetricLayerKind {
-    Delta,
-    Image,
-}
-
-static TIMELINE_LAYER_SIZE: Lazy<UIntGaugeVec> = Lazy::new(|| {
-    register_uint_gauge_vec!(
-        "pageserver_layer_bytes",
-        "Sum of layer physical sizes in bytes",
-        &["tenant_id", "shard_id", "timeline_id", "kind"]
-    )
-    .expect("failed to define a metric")
-});
-
-static TIMELINE_LAYER_COUNT: Lazy<UIntGaugeVec> = Lazy::new(|| {
-    register_uint_gauge_vec!(
-        "pageserver_layer_count",
-        "Number of layers that exist",
-        &["tenant_id", "shard_id", "timeline_id", "kind"]
-    )
-    .expect("failed to define a metric")
-});
-
 static TIMELINE_ARCHIVE_SIZE: Lazy<UIntGaugeVec> = Lazy::new(|| {
     register_uint_gauge_vec!(
         "pageserver_archive_size",
@@ -525,15 +500,6 @@ static RESIDENT_PHYSICAL_SIZE: Lazy<UIntGaugeVec> = Lazy::new(|| {
     .expect("failed to define a metric")
 });
 
-static VISIBLE_PHYSICAL_SIZE: Lazy<UIntGaugeVec> = Lazy::new(|| {
-    register_uint_gauge_vec!(
-        "pageserver_visible_physical_size",
-        "The size of the layer files present in the pageserver's filesystem.",
-        &["tenant_id", "shard_id", "timeline_id"]
-    )
-    .expect("failed to define a metric")
-});
-
 pub(crate) static RESIDENT_PHYSICAL_SIZE_GLOBAL: Lazy<UIntGauge> = Lazy::new(|| {
     register_uint_gauge!(
         "pageserver_resident_physical_size_global",
@@ -619,38 +585,6 @@ pub(crate) static CIRCUIT_BREAKERS_UNBROKEN: Lazy<IntCounter> = Lazy::new(|| {
     .expect("failed to define a metric")
 });
 
-pub(crate) static COMPRESSION_IMAGE_INPUT_BYTES: Lazy<IntCounter> = Lazy::new(|| {
-    register_int_counter!(
-        "pageserver_compression_image_in_bytes_total",
-        "Size of data written into image layers before compression"
-    )
-    .expect("failed to define a metric")
-});
-
-pub(crate) static COMPRESSION_IMAGE_INPUT_BYTES_CONSIDERED: Lazy<IntCounter> = Lazy::new(|| {
-    register_int_counter!(
-        "pageserver_compression_image_in_bytes_considered",
-        "Size of potentially compressible data written into image layers before compression"
-    )
-    .expect("failed to define a metric")
-});
-
-pub(crate) static COMPRESSION_IMAGE_INPUT_BYTES_CHOSEN: Lazy<IntCounter> = Lazy::new(|| {
-    register_int_counter!(
-        "pageserver_compression_image_in_bytes_chosen",
-        "Size of data whose compressed form was written into image layers"
-    )
-    .expect("failed to define a metric")
-});
-
-pub(crate) static COMPRESSION_IMAGE_OUTPUT_BYTES: Lazy<IntCounter> = Lazy::new(|| {
-    register_int_counter!(
-        "pageserver_compression_image_out_bytes_total",
-        "Size of compressed image layer written"
-    )
-    .expect("failed to define a metric")
-});
-
 pub(crate) mod initial_logical_size {
     use metrics::{register_int_counter, register_int_counter_vec, IntCounter, IntCounterVec};
     use once_cell::sync::Lazy;
@@ -1556,6 +1490,7 @@ pub(crate) enum ComputeCommandKind {
     Basebackup,
     Fullbackup,
     LeaseLsn,
+    Show,
 }
 
 pub(crate) struct ComputeCommandCounters {
@@ -2207,13 +2142,8 @@ pub(crate) struct TimelineMetrics {
     pub last_record_gauge: IntGauge,
     pub pitr_history_size: UIntGauge,
     pub archival_size: UIntGauge,
-    pub(crate) layer_size_image: UIntGauge,
-    pub(crate) layer_count_image: UIntGauge,
-    pub(crate) layer_size_delta: UIntGauge,
-    pub(crate) layer_count_delta: UIntGauge,
     pub standby_horizon_gauge: IntGauge,
     pub resident_physical_size_gauge: UIntGauge,
-    pub visible_physical_size_gauge: UIntGauge,
     /// copy of LayeredTimeline.current_logical_size
     pub current_logical_size_gauge: UIntGauge,
     pub aux_file_size_gauge: IntGauge,
@@ -2294,51 +2224,12 @@ impl TimelineMetrics {
             .get_metric_with_label_values(&[&tenant_id, &shard_id, &timeline_id])
             .unwrap();
 
-        let layer_size_image = TIMELINE_LAYER_SIZE
-            .get_metric_with_label_values(&[
-                &tenant_id,
-                &shard_id,
-                &timeline_id,
-                MetricLayerKind::Image.into(),
-            ])
-            .unwrap();
-
-        let layer_count_image = TIMELINE_LAYER_COUNT
-            .get_metric_with_label_values(&[
-                &tenant_id,
-                &shard_id,
-                &timeline_id,
-                MetricLayerKind::Image.into(),
-            ])
-            .unwrap();
-
-        let layer_size_delta = TIMELINE_LAYER_SIZE
-            .get_metric_with_label_values(&[
-                &tenant_id,
-                &shard_id,
-                &timeline_id,
-                MetricLayerKind::Delta.into(),
-            ])
-            .unwrap();
-
-        let layer_count_delta = TIMELINE_LAYER_COUNT
-            .get_metric_with_label_values(&[
-                &tenant_id,
-                &shard_id,
-                &timeline_id,
-                MetricLayerKind::Delta.into(),
-            ])
-            .unwrap();
-
         let standby_horizon_gauge = STANDBY_HORIZON
             .get_metric_with_label_values(&[&tenant_id, &shard_id, &timeline_id])
             .unwrap();
         let resident_physical_size_gauge = RESIDENT_PHYSICAL_SIZE
             .get_metric_with_label_values(&[&tenant_id, &shard_id, &timeline_id])
             .unwrap();
-        let visible_physical_size_gauge = VISIBLE_PHYSICAL_SIZE
-            .get_metric_with_label_values(&[&tenant_id, &shard_id, &timeline_id])
-            .unwrap();
         // TODO: we shouldn't expose this metric
         let current_logical_size_gauge = CURRENT_LOGICAL_SIZE
             .get_metric_with_label_values(&[&tenant_id, &shard_id, &timeline_id])
@@ -2387,13 +2278,8 @@ impl TimelineMetrics {
             last_record_gauge,
             pitr_history_size,
             archival_size,
-            layer_size_image,
-            layer_count_image,
-            layer_size_delta,
-            layer_count_delta,
             standby_horizon_gauge,
             resident_physical_size_gauge,
-            visible_physical_size_gauge,
             current_logical_size_gauge,
             aux_file_size_gauge,
             directory_entries_count_gauge,
@@ -2445,7 +2331,6 @@ impl TimelineMetrics {
             RESIDENT_PHYSICAL_SIZE_GLOBAL.sub(self.resident_physical_size_get());
             let _ = RESIDENT_PHYSICAL_SIZE.remove_label_values(&[tenant_id, shard_id, timeline_id]);
         }
-        let _ = VISIBLE_PHYSICAL_SIZE.remove_label_values(&[tenant_id, shard_id, timeline_id]);
         let _ = CURRENT_LOGICAL_SIZE.remove_label_values(&[tenant_id, shard_id, timeline_id]);
         if let Some(metric) = Lazy::get(&DIRECTORY_ENTRIES_COUNT) {
             let _ = metric.remove_label_values(&[tenant_id, shard_id, timeline_id]);
@@ -2454,31 +2339,6 @@ impl TimelineMetrics {
         let _ = TIMELINE_ARCHIVE_SIZE.remove_label_values(&[tenant_id, shard_id, timeline_id]);
         let _ = PITR_HISTORY_SIZE.remove_label_values(&[tenant_id, shard_id, timeline_id]);
 
-        let _ = TIMELINE_LAYER_SIZE.remove_label_values(&[
-            tenant_id,
-            shard_id,
-            timeline_id,
-            MetricLayerKind::Image.into(),
-        ]);
-        let _ = TIMELINE_LAYER_COUNT.remove_label_values(&[
-            tenant_id,
-            shard_id,
-            timeline_id,
-            MetricLayerKind::Image.into(),
-        ]);
-        let _ = TIMELINE_LAYER_SIZE.remove_label_values(&[
-            tenant_id,
-            shard_id,
-            timeline_id,
-            MetricLayerKind::Delta.into(),
-        ]);
-        let _ = TIMELINE_LAYER_COUNT.remove_label_values(&[
-            tenant_id,
-            shard_id,
-            timeline_id,
-            MetricLayerKind::Delta.into(),
-        ]);
-
         let _ = EVICTIONS.remove_label_values(&[tenant_id, shard_id, timeline_id]);
         let _ = AUX_FILE_SIZE.remove_label_values(&[tenant_id, shard_id, timeline_id]);
         let _ = VALID_LSN_LEASE_COUNT.remove_label_values(&[tenant_id, shard_id, timeline_id]);
@@ -3135,8 +2995,6 @@ pub fn preinitialize_metrics() {
         &tokio_epoll_uring::THREAD_LOCAL_LAUNCH_SUCCESSES,
         &REMOTE_ONDEMAND_DOWNLOADED_LAYERS,
         &REMOTE_ONDEMAND_DOWNLOADED_BYTES,
-        &CIRCUIT_BREAKERS_BROKEN,
-        &CIRCUIT_BREAKERS_UNBROKEN,
     ]
     .into_iter()
     .for_each(|c| {

pageserver/src/pgdatadir_mapping.rs

@@ -284,16 +284,6 @@ impl Timeline {
         if let Some(_nblocks) = self.get_cached_rel_size(&tag, version.get_lsn()) {
             return Ok(true);
         }
-        // then check if the database was already initialized.
-        // get_rel_exists can be called before dbdir is created.
-        let buf = version.get(self, DBDIR_KEY, ctx).await?;
-        let dbdirs = match DbDirectory::des(&buf).context("deserialization failure") {
-            Ok(dir) => Ok(dir.dbdirs),
-            Err(e) => Err(PageReconstructError::from(e)),
-        }?;
-        if !dbdirs.contains_key(&(tag.spcnode, tag.dbnode)) {
-            return Ok(false);
-        }
         // fetch directory listing
         let key = rel_dir_to_key(tag.spcnode, tag.dbnode);
         let buf = version.get(self, key, ctx).await?;
@@ -2041,7 +2031,7 @@ mod tests {
     #[tokio::test]
     async fn aux_files_round_trip() -> anyhow::Result<()> {
         let name = "aux_files_round_trip";
-        let harness = TenantHarness::create(name).await?;
+        let harness = TenantHarness::create(name)?;
 
         pub const TIMELINE_ID: TimelineId =
             TimelineId::from_array(hex!("11223344556677881122334455667788"));

pageserver/src/repository.rs

@@ -8,7 +8,8 @@ use std::time::Duration;
 pub use pageserver_api::key::{Key, KEY_SIZE};
 
 /// A 'value' stored for a one Key.
-#[derive(Debug, Clone, Serialize, Deserialize, PartialEq, Eq)]
+#[derive(Debug, Clone, Serialize, Deserialize)]
+#[cfg_attr(test, derive(PartialEq))]
 pub enum Value {
     /// An Image value contains a full copy of the value
     Image(Bytes),

pageserver/src/statvfs.rs

@@ -56,6 +56,7 @@ impl Statvfs {
 }
 
 pub mod mock {
+    use anyhow::Context;
     use camino::Utf8Path;
     use regex::Regex;
     use tracing::log::info;
@@ -134,30 +135,14 @@ pub mod mock {
             {
                 continue;
             }
-            let m = match entry.metadata() {
-                Ok(m) => m,
-                Err(e) if is_not_found(&e) => {
-                    // some temp file which got removed right as we are walking
-                    continue;
-                }
-                Err(e) => {
-                    return Err(anyhow::Error::new(e)
-                        .context(format!("get metadata of {:?}", entry.path())))
-                }
-            };
-            total += m.len();
+            total += entry
+                .metadata()
+                .with_context(|| format!("get metadata of {:?}", entry.path()))?
+                .len();
         }
         Ok(total)
     }
 
-    fn is_not_found(e: &walkdir::Error) -> bool {
-        let Some(io_error) = e.io_error() else {
-            return false;
-        };
-        let kind = io_error.kind();
-        matches!(kind, std::io::ErrorKind::NotFound)
-    }
-
     pub struct Statvfs {
         pub blocks: u64,
         pub blocks_available: u64,