Add GUCs to control libpq SSL parameters in walproposer and pagestore

In preparation for TLS enablement of compute to {pageserver,safekeeper}
connections, add GUCs to control the various SSL connection parameters.
Start with the Postgres defaults, and change them when we have all the
files in place for enabling TLS encryption on the connections.

Note that the Neon DBaaS won't make use of all these parameters, but
open source users may want to use them.

Part-of: https://github.com/neondatabase/cloud/issues/25823
Signed-off-by: Tristan Partin <tristan@neon.tech>

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

@@ -49,10 +49,6 @@ inputs:
     description: 'A JSON object with project settings'
     required: false
     default: '{}'
-  default_endpoint_settings:
-    description: 'A JSON object with the default endpoint settings'
-    required: false
-    default: '{}'
 
 outputs:
   dsn:
@@ -70,9 +66,9 @@ runs:
       # A shell without `set -x` to not to expose password/dsn in logs
       shell: bash -euo pipefail {0}
       run: |
-        res=$(curl \
+        project=$(curl \
           "https://${API_HOST}/api/v2/projects" \
-          -w "%{http_code}" \
+          --fail \
           --header "Accept: application/json" \
           --header "Content-Type: application/json" \
           --header "Authorization: Bearer ${API_KEY}" \
@@ -87,15 +83,6 @@ runs:
               \"settings\": ${PROJECT_SETTINGS}
             }
           }")
-        
-        code=${res: -3}
-        if [[ ${code} -ge 400 ]]; then
-          echo Request failed with error code ${code}
-          echo ${res::-3}
-          exit 1
-        else
-          project=${res::-3}
-        fi
 
         # Mask password
         echo "::add-mask::$(echo $project | jq --raw-output '.roles[] | select(.name != "web_access") | .password')"
@@ -139,22 +126,6 @@ runs:
             -H "Accept: application/json" -H "Content-Type: application/json" -H "Authorization: Bearer ${ADMIN_API_KEY}" \
             -d "{\"scheduling\": \"Essential\"}"
         fi
-        # XXX
-        # This is a workaround for the default endpoint settings, which currently do not allow some settings in the public API.
-        # https://github.com/neondatabase/cloud/issues/27108
-        if [[ -n ${DEFAULT_ENDPOINT_SETTINGS} && ${DEFAULT_ENDPOINT_SETTINGS} != "{}" ]] ; then
-          PROJECT_DATA=$(curl -X GET \
-              "https://${API_HOST}/regions/${REGION_ID}/api/v1/admin/projects/${project_id}" \
-              -H "Accept: application/json" -H "Content-Type: application/json" -H "Authorization: Bearer ${ADMIN_API_KEY}" \
-              -d "{\"scheduling\": \"Essential\"}"
-          )
-          NEW_DEFAULT_ENDPOINT_SETTINGS=$(echo ${PROJECT_DATA} | jq -rc ".project.default_endpoint_settings + ${DEFAULT_ENDPOINT_SETTINGS}")
-          curl -X POST --fail \
-                "https://${API_HOST}/regions/${REGION_ID}/api/v1/admin/projects/${project_id}/default_endpoint_settings" \
-                -H "Accept: application/json" -H "Content-Type: application/json" -H "Authorization: Bearer ${ADMIN_API_KEY}" \
-                --data "${NEW_DEFAULT_ENDPOINT_SETTINGS}"
-        fi
-        
 
       env:
         API_HOST: ${{ inputs.api_host }}
@@ -171,4 +142,3 @@ runs:
         PSQL: ${{ inputs.psql_path }}
         LD_LIBRARY_PATH: ${{ inputs.libpq_lib_path }}
         PROJECT_SETTINGS: ${{ inputs.project_settings }}
-        DEFAULT_ENDPOINT_SETTINGS: ${{ inputs.default_endpoint_settings }}

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

@@ -28,16 +28,6 @@ on:
         required: false
         default: 'disabled'
         type: string
-      test-selection:
-        description: 'specification of selected test(s) to run'
-        required: false
-        default: ''
-        type: string
-      test-run-count:
-        description: 'number of runs to perform for selected tests'
-        required: false
-        default: 1
-        type: number
 
 defaults:
   run:
@@ -285,7 +275,7 @@ jobs:
                 for io_mode in buffered direct direct-rw ; do
                   NEON_PAGESERVER_UNIT_TEST_GET_VECTORED_CONCURRENT_IO=$get_vectored_concurrent_io \
                   NEON_PAGESERVER_UNIT_TEST_VIRTUAL_FILE_IOENGINE=$io_engine \
-                  NEON_PAGESERVER_UNIT_TEST_VIRTUAL_FILE_IO_MODE=$io_mode \
+                  NEON_PAGESERVER_UNIT_TEST_VIRTUAL_FILE_IOMODE=$io_mode \
                   ${cov_prefix} \
                   cargo nextest run $CARGO_FLAGS $CARGO_FEATURES  -E 'package(pageserver)'
               done
@@ -391,22 +381,21 @@ jobs:
           run_with_real_s3: true
           real_s3_bucket: neon-github-ci-tests
           real_s3_region: eu-central-1
-          rerun_failed: ${{ inputs.test-run-count == 1 }}
+          rerun_failed: true
           pg_version: ${{ matrix.pg_version }}
           sanitizers: ${{ inputs.sanitizers }}
           aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
           # `--session-timeout` is equal to (timeout-minutes - 10 minutes) * 60 seconds.
           # Attempt to stop tests gracefully to generate test reports
           # until they are forcibly stopped by the stricter `timeout-minutes` limit.
-          extra_params: --session-timeout=${{ inputs.sanitizers != 'enabled' && 3000 || 10200 }} --count=${{ inputs.test-run-count }}
-                        ${{ inputs.test-selection != '' && format('-k "{0}"', inputs.test-selection) || '' }}
+          extra_params: --session-timeout=${{ inputs.sanitizers != 'enabled' && 3000 || 10200 }}
         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
           PAGESERVER_GET_VECTORED_CONCURRENT_IO: sidecar-task
-          PAGESERVER_VIRTUAL_FILE_IO_MODE: direct-rw
+          PAGESERVER_VIRTUAL_FILE_IO_MODE: direct
           USE_LFC: ${{ matrix.lfc_state == 'with-lfc' && 'true' || 'false' }}
 
       # Temporary disable this step until we figure out why it's so flaky

.github/workflows/build-macos.yml

@@ -34,10 +34,11 @@ permissions:
 jobs:
   build-pgxn:
     if: |
-      inputs.pg_versions != '[]' || inputs.rebuild_everything ||
-      contains(github.event.pull_request.labels.*.name, 'run-extra-build-macos')  ||
-      contains(github.event.pull_request.labels.*.name, 'run-extra-build-*') ||
-      github.ref_name == 'main'
+      (inputs.pg_versions != '[]' || inputs.rebuild_everything) && (
+        contains(github.event.pull_request.labels.*.name, 'run-extra-build-macos')  ||
+        contains(github.event.pull_request.labels.*.name, 'run-extra-build-*') ||
+        github.ref_name == 'main'
+      )
     timeout-minutes: 30
     runs-on: macos-15
     strategy:
@@ -62,8 +63,13 @@ jobs:
 
       - name: Cache postgres ${{ matrix.postgres-version }} build
         id: cache_pg
-        uses: actions/cache@5a3ec84eff668545956fd18022155c47e93e2684 # v4.2.3
+        uses: tespkg/actions-cache@b7bf5fcc2f98a52ac6080eb0fd282c2f752074b1  # v1.8.0
         with:
+          endpoint: ${{ vars.HETZNER_CACHE_REGION }}.${{ vars.HETZNER_CACHE_ENDPOINT }}
+          bucket: ${{ vars.HETZNER_CACHE_BUCKET }}
+          accessKey: ${{ secrets.HETZNER_CACHE_ACCESS_KEY }}
+          secretKey: ${{ secrets.HETZNER_CACHE_SECRET_KEY }}
+          use-fallback: false
           path: pg_install/${{ matrix.postgres-version }}
           key: v1-${{ runner.os }}-${{ runner.arch }}-${{ env.BUILD_TYPE }}-pg-${{ matrix.postgres-version }}-${{ steps.pg_rev.outputs.pg_rev }}-${{ hashFiles('Makefile') }}
 
@@ -99,21 +105,13 @@ jobs:
         run: |
           make postgres-headers-${{ matrix.postgres-version }} -j$(sysctl -n hw.ncpu)
 
-      - name: Upload "pg_install/${{ matrix.postgres-version }}" artifact
-        uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4.6.2
-        with:
-          name: pg_install--${{ matrix.postgres-version }}
-          path: pg_install/${{ matrix.postgres-version }}
-          # The artifact is supposed to be used by the next job in the same workflow,
-          # so there’s no need to store it for too long.
-          retention-days: 1
-
   build-walproposer-lib:
     if: |
-      inputs.pg_versions != '[]' || inputs.rebuild_everything ||
-      contains(github.event.pull_request.labels.*.name, 'run-extra-build-macos')  ||
-      contains(github.event.pull_request.labels.*.name, 'run-extra-build-*') ||
-      github.ref_name == 'main'
+      (inputs.pg_versions != '[]' || inputs.rebuild_everything) && (
+        contains(github.event.pull_request.labels.*.name, 'run-extra-build-macos')  ||
+        contains(github.event.pull_request.labels.*.name, 'run-extra-build-*') ||
+        github.ref_name == 'main'
+      )
     timeout-minutes: 30
     runs-on: macos-15
     needs: [build-pgxn]
@@ -134,16 +132,27 @@ jobs:
         id: pg_rev
         run: echo pg_rev=$(git rev-parse HEAD:vendor/postgres-v17) | tee -a "${GITHUB_OUTPUT}"
 
-      - name: Download "pg_install/v17" artifact
-        uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
+      - name: Cache postgres v17 build
+        id: cache_pg
+        uses: tespkg/actions-cache@b7bf5fcc2f98a52ac6080eb0fd282c2f752074b1  # v1.8.0
         with:
-          name: pg_install--v17
+          endpoint: ${{ vars.HETZNER_CACHE_REGION }}.${{ vars.HETZNER_CACHE_ENDPOINT }}
+          bucket: ${{ vars.HETZNER_CACHE_BUCKET }}
+          accessKey: ${{ secrets.HETZNER_CACHE_ACCESS_KEY }}
+          secretKey: ${{ secrets.HETZNER_CACHE_SECRET_KEY }}
+          use-fallback: false
           path: pg_install/v17
+          key: v1-${{ runner.os }}-${{ runner.arch }}-${{ env.BUILD_TYPE }}-pg-v17-${{ steps.pg_rev.outputs.pg_rev }}-${{ hashFiles('Makefile') }}
 
       - name: Cache walproposer-lib
         id: cache_walproposer_lib
-        uses: actions/cache@5a3ec84eff668545956fd18022155c47e93e2684 # v4.2.3
+        uses: tespkg/actions-cache@b7bf5fcc2f98a52ac6080eb0fd282c2f752074b1  # v1.8.0
         with:
+          endpoint: ${{ vars.HETZNER_CACHE_REGION }}.${{ vars.HETZNER_CACHE_ENDPOINT }}
+          bucket: ${{ vars.HETZNER_CACHE_BUCKET }}
+          accessKey: ${{ secrets.HETZNER_CACHE_ACCESS_KEY }}
+          secretKey: ${{ secrets.HETZNER_CACHE_SECRET_KEY }}
+          use-fallback: false
           path: pg_install/build/walproposer-lib
           key: v1-${{ runner.os }}-${{ runner.arch }}-${{ env.BUILD_TYPE }}-walproposer_lib-v17-${{ steps.pg_rev.outputs.pg_rev }}-${{ hashFiles('Makefile') }}
 
@@ -169,21 +178,13 @@ jobs:
         run:
           make walproposer-lib -j$(sysctl -n hw.ncpu)
 
-      - name: Upload "pg_install/build/walproposer-lib" artifact
-        uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4.6.2
-        with:
-          name: pg_install--build--walproposer-lib
-          path: pg_install/build/walproposer-lib
-          # The artifact is supposed to be used by the next job in the same workflow,
-          # so there’s no need to store it for too long.
-          retention-days: 1
-
   cargo-build:
     if: |
-      inputs.pg_versions != '[]' || inputs.rebuild_rust_code || inputs.rebuild_everything ||
-      contains(github.event.pull_request.labels.*.name, 'run-extra-build-macos') ||
-      contains(github.event.pull_request.labels.*.name, 'run-extra-build-*') ||
-      github.ref_name == 'main'
+      (inputs.pg_versions != '[]' || inputs.rebuild_rust_code || inputs.rebuild_everything) && (
+        contains(github.event.pull_request.labels.*.name, 'run-extra-build-macos')  ||
+        contains(github.event.pull_request.labels.*.name, 'run-extra-build-*') ||
+        github.ref_name == 'main'
+      )
     timeout-minutes: 30
     runs-on: macos-15
     needs: [build-pgxn, build-walproposer-lib]
@@ -202,45 +203,72 @@ jobs:
         with:
           submodules: true
 
-      - name: Download "pg_install/v14" artifact
-        uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
+      - name: Set pg v14 for caching
+        id: pg_rev_v14
+        run: echo pg_rev=$(git rev-parse HEAD:vendor/postgres-v14) | tee -a "${GITHUB_OUTPUT}"
+      - name: Set pg v15 for caching
+        id: pg_rev_v15
+        run: echo pg_rev=$(git rev-parse HEAD:vendor/postgres-v15) | tee -a "${GITHUB_OUTPUT}"
+      - name: Set pg v16 for caching
+        id: pg_rev_v16
+        run: echo pg_rev=$(git rev-parse HEAD:vendor/postgres-v16) | tee -a "${GITHUB_OUTPUT}"
+      - name: Set pg v17 for caching
+        id: pg_rev_v17
+        run: echo pg_rev=$(git rev-parse HEAD:vendor/postgres-v17) | tee -a "${GITHUB_OUTPUT}"
+
+      - name: Cache postgres v14 build
+        id: cache_pg
+        uses: tespkg/actions-cache@b7bf5fcc2f98a52ac6080eb0fd282c2f752074b1  # v1.8.0
         with:
-          name: pg_install--v14
+          endpoint: ${{ vars.HETZNER_CACHE_REGION }}.${{ vars.HETZNER_CACHE_ENDPOINT }}
+          bucket: ${{ vars.HETZNER_CACHE_BUCKET }}
+          accessKey: ${{ secrets.HETZNER_CACHE_ACCESS_KEY }}
+          secretKey: ${{ secrets.HETZNER_CACHE_SECRET_KEY }}
+          use-fallback: false
           path: pg_install/v14
-
-      - name: Download "pg_install/v15" artifact
-        uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
+          key: v1-${{ runner.os }}-${{ runner.arch }}-${{ env.BUILD_TYPE }}-pg-v14-${{ steps.pg_rev_v14.outputs.pg_rev }}-${{ hashFiles('Makefile') }}
+      - name: Cache postgres v15 build
+        id: cache_pg_v15
+        uses: tespkg/actions-cache@b7bf5fcc2f98a52ac6080eb0fd282c2f752074b1  # v1.8.0
         with:
-          name: pg_install--v15
+          endpoint: ${{ vars.HETZNER_CACHE_REGION }}.${{ vars.HETZNER_CACHE_ENDPOINT }}
+          bucket: ${{ vars.HETZNER_CACHE_BUCKET }}
+          accessKey: ${{ secrets.HETZNER_CACHE_ACCESS_KEY }}
+          secretKey: ${{ secrets.HETZNER_CACHE_SECRET_KEY }}
+          use-fallback: false
           path: pg_install/v15
-
-      - name: Download "pg_install/v16" artifact
-        uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
+          key: v1-${{ runner.os }}-${{ runner.arch }}-${{ env.BUILD_TYPE }}-pg-v15-${{ steps.pg_rev_v15.outputs.pg_rev }}-${{ hashFiles('Makefile') }}
+      - name: Cache postgres v16 build
+        id: cache_pg_v16
+        uses: tespkg/actions-cache@b7bf5fcc2f98a52ac6080eb0fd282c2f752074b1  # v1.8.0
         with:
-          name: pg_install--v16
+          endpoint: ${{ vars.HETZNER_CACHE_REGION }}.${{ vars.HETZNER_CACHE_ENDPOINT }}
+          bucket: ${{ vars.HETZNER_CACHE_BUCKET }}
+          accessKey: ${{ secrets.HETZNER_CACHE_ACCESS_KEY }}
+          secretKey: ${{ secrets.HETZNER_CACHE_SECRET_KEY }}
+          use-fallback: false
           path: pg_install/v16
-
-      - name: Download "pg_install/v17" artifact
-        uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
+          key: v1-${{ runner.os }}-${{ runner.arch }}-${{ env.BUILD_TYPE }}-pg-v16-${{ steps.pg_rev_v16.outputs.pg_rev }}-${{ hashFiles('Makefile') }}
+      - name: Cache postgres v17 build
+        id: cache_pg_v17
+        uses: tespkg/actions-cache@b7bf5fcc2f98a52ac6080eb0fd282c2f752074b1  # v1.8.0
         with:
-          name: pg_install--v17
+          endpoint: ${{ vars.HETZNER_CACHE_REGION }}.${{ vars.HETZNER_CACHE_ENDPOINT }}
+          bucket: ${{ vars.HETZNER_CACHE_BUCKET }}
+          accessKey: ${{ secrets.HETZNER_CACHE_ACCESS_KEY }}
+          secretKey: ${{ secrets.HETZNER_CACHE_SECRET_KEY }}
+          use-fallback: false
           path: pg_install/v17
+          key: v1-${{ runner.os }}-${{ runner.arch }}-${{ env.BUILD_TYPE }}-pg-v17-${{ steps.pg_rev_v17.outputs.pg_rev }}-${{ hashFiles('Makefile') }}
 
-      - name: Download "pg_install/build/walproposer-lib" artifact
-        uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
-        with:
-          name: pg_install--build--walproposer-lib
-          path: pg_install/build/walproposer-lib
-
-      # `actions/download-artifact` doesn't preserve permissions:
-      # https://github.com/actions/download-artifact?tab=readme-ov-file#permission-loss
-      - name: Make pg_install/v*/bin/* executable
-        run: |
-          chmod +x pg_install/v*/bin/*
-
-      - name: Cache cargo deps
-        uses: actions/cache@5a3ec84eff668545956fd18022155c47e93e2684 # v4.2.3
+      - name: Cache cargo deps (only for v17)
+        uses: tespkg/actions-cache@b7bf5fcc2f98a52ac6080eb0fd282c2f752074b1  # v1.8.0
         with:
+          endpoint: ${{ vars.HETZNER_CACHE_REGION }}.${{ vars.HETZNER_CACHE_ENDPOINT }}
+          bucket: ${{ vars.HETZNER_CACHE_BUCKET }}
+          accessKey: ${{ secrets.HETZNER_CACHE_ACCESS_KEY }}
+          secretKey: ${{ secrets.HETZNER_CACHE_SECRET_KEY }}
+          use-fallback: false
           path: |
             ~/.cargo/registry
             !~/.cargo/registry/src
@@ -248,6 +276,18 @@ jobs:
             target
           key: v1-${{ runner.os }}-${{ runner.arch }}-cargo-${{ hashFiles('./Cargo.lock') }}-${{ hashFiles('./rust-toolchain.toml') }}-rust
 
+      - name: Cache walproposer-lib
+        id: cache_walproposer_lib
+        uses: tespkg/actions-cache@b7bf5fcc2f98a52ac6080eb0fd282c2f752074b1  # v1.8.0
+        with:
+          endpoint: ${{ vars.HETZNER_CACHE_REGION }}.${{ vars.HETZNER_CACHE_ENDPOINT }}
+          bucket: ${{ vars.HETZNER_CACHE_BUCKET }}
+          accessKey: ${{ secrets.HETZNER_CACHE_ACCESS_KEY }}
+          secretKey: ${{ secrets.HETZNER_CACHE_SECRET_KEY }}
+          use-fallback: false
+          path: pg_install/build/walproposer-lib
+          key: v1-${{ runner.os }}-${{ runner.arch }}-${{ env.BUILD_TYPE }}-walproposer_lib-v17-${{ steps.pg_rev_v17.outputs.pg_rev }}-${{ hashFiles('Makefile') }}
+
       - name: Install build dependencies
         run: |
           brew install flex bison openssl protobuf icu4c
@@ -257,8 +297,8 @@ jobs:
           echo 'LDFLAGS=-L/usr/local/opt/openssl@3/lib' >> $GITHUB_ENV
           echo 'CPPFLAGS=-I/usr/local/opt/openssl@3/include' >> $GITHUB_ENV
 
-      - name: Run cargo build
+      - name: Run cargo build (only for v17)
         run: cargo build --all --release -j$(sysctl -n hw.ncpu)
 
-      - name: Check that no warnings are produced
+      - name: Check that no warnings are produced (only for v17)
         run: ./run_clippy.sh

.github/workflows/build_and_run_selected_test.yml

@@ -1,120 +0,0 @@
-name: Build and Run Selected Test
-
-on:
-  workflow_dispatch:
-    inputs:
-      test-selection:
-        description: 'Specification of selected test(s), as accepted by pytest -k'
-        required: true
-        type: string
-      run-count:
-        description: 'Number of test runs to perform'
-        required: true
-        type: number
-      archs:
-        description: 'Archs to run tests on, e. g.: ["x64", "arm64"]'
-        default: '["x64"]'
-        required: true
-        type: string
-      build-types:
-        description: 'Build types to run tests on, e. g.: ["debug", "release"]'
-        default: '["release"]'
-        required: true
-        type: string
-      pg-versions:
-        description: 'Postgres versions to use for testing,  e.g,: [{"pg_version":"v16"}, {"pg_version":"v17"}])'
-        default: '[{"pg_version":"v17"}]'
-        required: true
-        type: string
-
-defaults:
-  run:
-    shell: bash -euxo pipefail {0}
-
-env:
-  RUST_BACKTRACE: 1
-  COPT: '-Werror'
-
-jobs:
-  meta:
-    uses: ./.github/workflows/_meta.yml
-    with:
-      github-event-name: ${{ github.event_name }}
-      github-event-json: ${{ toJSON(github.event) }}
-
-  build-and-test-locally:
-    needs: [ meta ]
-    strategy:
-      fail-fast: false
-      matrix:
-        arch: ${{ fromJson(inputs.archs) }}
-        build-type: ${{ fromJson(inputs.build-types) }}
-    uses: ./.github/workflows/_build-and-test-locally.yml
-    with:
-      arch: ${{ matrix.arch }}
-      build-tools-image: ghcr.io/neondatabase/build-tools:pinned-bookworm
-      build-tag: ${{ needs.meta.outputs.build-tag }}
-      build-type: ${{ matrix.build-type }}
-      test-cfg: ${{ inputs.pg-versions }}
-      test-selection: ${{ inputs.test-selection }}
-      test-run-count: ${{ fromJson(inputs.run-count) }}
-    secrets: inherit
-
-  create-test-report:
-    needs: [ build-and-test-locally ]
-    if: ${{ !cancelled() }}
-    permissions:
-      id-token: write # aws-actions/configure-aws-credentials
-      statuses: write
-      contents: write
-      pull-requests: write
-    outputs:
-      report-url: ${{ steps.create-allure-report.outputs.report-url }}
-
-    runs-on: [ self-hosted, small ]
-    container:
-      image: ghcr.io/neondatabase/build-tools:pinned-bookworm
-      credentials:
-        username: ${{ github.actor }}
-        password: ${{ secrets.GITHUB_TOKEN }}
-      options: --init
-
-    steps:
-      - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@4d991eb9b905ef189e4c376166672c3f2f230481 # v2.11.0
-        with:
-          egress-policy: audit
-
-      - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
-
-      - name: Create Allure report
-        if: ${{ !cancelled() }}
-        id: create-allure-report
-        uses: ./.github/actions/allure-report-generate
-        with:
-          store-test-results-into-db: true
-          aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
-        env:
-          REGRESS_TEST_RESULT_CONNSTR_NEW: ${{ secrets.REGRESS_TEST_RESULT_CONNSTR_DEV }}
-
-      - uses: actions/github-script@v7
-        if: ${{ !cancelled() }}
-        with:
-          # Retry script for 5XX server errors: https://github.com/actions/github-script#retries
-          retries: 5
-          script: |
-            const report = {
-              reportUrl:     "${{ steps.create-allure-report.outputs.report-url }}",
-              reportJsonUrl: "${{ steps.create-allure-report.outputs.report-json-url }}",
-            }
-
-            const coverage = {}
-
-            const script = require("./scripts/comment-test-report.js")
-            await script({
-              github,
-              context,
-              fetch,
-              report,
-              coverage,
-            })

.github/workflows/build_and_test.yml

@@ -324,7 +324,7 @@ jobs:
           TEST_RESULT_CONNSTR: "${{ secrets.REGRESS_TEST_RESULT_CONNSTR_NEW }}"
           PAGESERVER_VIRTUAL_FILE_IO_ENGINE: tokio-epoll-uring
           PAGESERVER_GET_VECTORED_CONCURRENT_IO: sidecar-task
-          PAGESERVER_VIRTUAL_FILE_IO_MODE: direct-rw
+          PAGESERVER_VIRTUAL_FILE_IO_MODE: direct
           SYNC_BETWEEN_TESTS: true
       # XXX: no coverage data handling here, since benchmarks are run on release builds,
       # while coverage is currently collected for the debug ones
@@ -1238,7 +1238,7 @@ jobs:
         env:
           GH_TOKEN: ${{ secrets.CI_ACCESS_TOKEN }}
         run: |
-          TIMEOUT=5400 # 90 minutes, usually it takes ~2-3 minutes, but if runners are busy, it might take longer
+          TIMEOUT=1800 # 30 minutes, usually it takes ~2-3 minutes, but if runners are busy, it might take longer
           INTERVAL=15 # try each N seconds
 
           last_status="" # a variable to carry the last status of the "build-and-upload-extensions" context

.github/workflows/check-permissions.yml

@@ -19,7 +19,7 @@ jobs:
     runs-on: ubuntu-22.04
     steps:
     - name: Harden the runner (Audit all outbound calls)
-      uses: step-security/harden-runner@0634a2670c59f64b4a01f0f96f84700a4088b9f0 # v2.12.0
+      uses: step-security/harden-runner@v2
       with:
         egress-policy: audit
 

.github/workflows/cleanup-caches-by-a-branch.yml

@@ -12,7 +12,7 @@ jobs:
     runs-on: ubuntu-22.04
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@0634a2670c59f64b4a01f0f96f84700a4088b9f0 # v2.12.0
+        uses: step-security/harden-runner@v2
         with:
           egress-policy: audit
 

.github/workflows/cloud-extensions.yml

@@ -1,112 +0,0 @@
-name: Cloud Extensions Test
-on:
-  schedule:
-    # * is a special character in YAML so you have to quote this string
-    #          ┌───────────── minute (0 - 59)
-    #          │ ┌───────────── hour (0 - 23)
-    #          │ │ ┌───────────── day of the month (1 - 31)
-    #          │ │ │ ┌───────────── month (1 - 12 or JAN-DEC)
-    #          │ │ │ │ ┌───────────── day of the week (0 - 6 or SUN-SAT)
-    - cron:  '45 1 * * *' # run once a day, timezone is utc
-  workflow_dispatch: # adds ability to run this manually
-    inputs:
-      region_id:
-        description: 'Project region id. If not set, the default region will be used'
-        required: false
-        default: 'aws-us-east-2'
-
-defaults:
-  run:
-    shell: bash -euxo pipefail {0}
-
-permissions:
-  id-token: write # aws-actions/configure-aws-credentials
-  statuses: write
-  contents: write
-
-jobs:
-  regress:
-    env:
-      POSTGRES_DISTRIB_DIR: /tmp/neon/pg_install
-      TEST_OUTPUT: /tmp/test_output
-      BUILD_TYPE: remote
-    strategy:
-      fail-fast: false
-      matrix:
-        pg-version: [16, 17]
-
-    runs-on: [ self-hosted, small ]
-    container:
-      # We use the neon-test-extensions image here as it contains the source code for the extensions.
-      image: ghcr.io/neondatabase/neon-test-extensions-v${{ matrix.pg-version }}:latest
-      credentials:
-        username: ${{ github.actor }}
-        password: ${{ secrets.GITHUB_TOKEN }}
-      options: --init
-
-    steps:
-      - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@4d991eb9b905ef189e4c376166672c3f2f230481 # v2.11.0
-        with:
-          egress-policy: audit
-
-      - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4.2.2
-
-      - name: Evaluate the settings
-        id: project-settings
-        run: |
-          if [[ $((${{ matrix.pg-version }})) -lt 17 ]]; then
-            ULID=ulid
-          else
-            ULID=pgx_ulid
-          fi
-          LIBS=timescaledb:rag_bge_small_en_v15,rag_jina_reranker_v1_tiny_en:$ULID
-          settings=$(jq -c -n --arg libs $LIBS '{preload_libraries:{use_defaults:false,enabled_libraries:($libs| split(":"))}}')
-          echo settings=$settings >> $GITHUB_OUTPUT
-          
-      - name: Create Neon Project
-        id: create-neon-project
-        uses: ./.github/actions/neon-project-create
-        with:
-          region_id: ${{ inputs.region_id }}
-          postgres_version: ${{ matrix.pg-version }}
-          project_settings: ${{ steps.project-settings.outputs.settings }}
-          # We need these settings to get the expected output results.
-          # We cannot use the environment variables e.g. PGTZ due to
-          # https://github.com/neondatabase/neon/issues/1287
-          default_endpoint_settings: >
-            {
-              "pg_settings": {
-                "DateStyle": "Postgres,MDY",
-                "TimeZone": "America/Los_Angeles",
-                "compute_query_id": "off",
-                "neon.allow_unstable_extensions": "on"
-              }
-            }
-          api_key: ${{ secrets.NEON_STAGING_API_KEY }}
-          admin_api_key: ${{ secrets.NEON_STAGING_ADMIN_API_KEY }}
-
-      - name: Run the regression tests
-        run: /run-tests.sh -r /ext-src
-        env:
-          BENCHMARK_CONNSTR: ${{ steps.create-neon-project.outputs.dsn }}
-          SKIP: "pg_hint_plan-src,pg_repack-src,pg_cron-src,plpgsql_check-src"
-
-      - 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: Post to a Slack channel
-        if: ${{ github.event.schedule && failure() }}
-        uses: slackapi/slack-github-action@fcfb566f8b0aab22203f066d80ca1d7e4b5d05b3 # v1.27.1
-        with:
-          channel-id: ${{ vars.SLACK_ON_CALL_QA_STAGING_STREAM }}
-          slack-message: |
-            Periodic extensions test on staging: ${{ job.status }}
-            <${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}|GitHub Run>
-        env:
-          SLACK_BOT_TOKEN: ${{ secrets.SLACK_BOT_TOKEN }}
-

.github/workflows/fast-forward.yml

@@ -14,7 +14,7 @@ jobs:
 
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@0634a2670c59f64b4a01f0f96f84700a4088b9f0 # v2.12.0
+        uses: step-security/harden-runner@v2
         with:
           egress-policy: audit
 

.github/workflows/label-for-external-users.yml

@@ -28,7 +28,7 @@ jobs:
 
     steps:
     - name: Harden the runner (Audit all outbound calls)
-      uses: step-security/harden-runner@0634a2670c59f64b4a01f0f96f84700a4088b9f0 # v2.12.0
+      uses: step-security/harden-runner@v2
       with:
         egress-policy: audit
 
@@ -75,7 +75,7 @@ jobs:
 
     steps:
     - name: Harden the runner (Audit all outbound calls)
-      uses: step-security/harden-runner@0634a2670c59f64b4a01f0f96f84700a4088b9f0 # v2.12.0
+      uses: step-security/harden-runner@v2
       with:
         egress-policy: audit
 

.github/workflows/neon_extra_builds.yml

@@ -69,6 +69,10 @@ jobs:
 
   check-macos-build:
     needs: [ check-permissions, files-changed ]
+    if: |
+      contains(github.event.pull_request.labels.*.name, 'run-extra-build-macos')  ||
+      contains(github.event.pull_request.labels.*.name, 'run-extra-build-*') ||
+      github.ref_name == 'main'
     uses: ./.github/workflows/build-macos.yml
     with:
       pg_versions: ${{ needs.files-changed.outputs.postgres_changes }}

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

@@ -41,7 +41,7 @@ jobs:
 
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@0634a2670c59f64b4a01f0f96f84700a4088b9f0 # v2.12.0
+        uses: step-security/harden-runner@v2
         with:
           egress-policy: audit
 

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

@@ -35,7 +35,7 @@ jobs:
 
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@0634a2670c59f64b4a01f0f96f84700a4088b9f0 # v2.12.0
+        uses: step-security/harden-runner@v2
         with:
           egress-policy: audit
 
@@ -73,7 +73,7 @@ jobs:
         }}
     steps:
       - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@0634a2670c59f64b4a01f0f96f84700a4088b9f0 # v2.12.0
+        uses: step-security/harden-runner@v2
         with:
           egress-policy: audit
 

Cargo.lock

@@ -40,7 +40,7 @@ dependencies = [
  "getrandom 0.2.11",
  "once_cell",
  "version_check",
- "zerocopy 0.7.31",
+ "zerocopy",
 ]
 
 [[package]]
@@ -1323,6 +1323,7 @@ dependencies = [
  "serde_json",
  "serde_with",
  "signal-hook",
+ "spki 0.7.3",
  "tar",
  "thiserror 1.0.69",
  "tokio",
@@ -4301,7 +4302,6 @@ dependencies = [
  "remote_storage",
  "reqwest",
  "rpds",
- "rstest",
  "rustls 0.23.18",
  "scopeguard",
  "send-future",
@@ -4415,9 +4415,9 @@ dependencies = [
 
 [[package]]
 name = "papaya"
-version = "0.2.1"
+version = "0.2.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "6827e3fc394523c21d4464d02c0bb1c19966ea4a58a9844ad6d746214179d2bc"
+checksum = "aab21828b6b5952fdadd6c377728ffae53ec3a21b2febc47319ab65741f7e2fd"
 dependencies = [
  "equivalent",
  "seize",
@@ -5204,7 +5204,7 @@ dependencies = [
  "walkdir",
  "workspace_hack",
  "x509-cert",
- "zerocopy 0.8.24",
+ "zerocopy",
 ]
 
 [[package]]
@@ -5594,7 +5594,7 @@ dependencies = [
  "wasm-bindgen-futures",
  "wasm-streams",
  "web-sys",
- "webpki-roots",
+ "webpki-roots 0.26.1",
  "winreg",
 ]
 
@@ -6195,13 +6195,13 @@ checksum = "224e328af6e080cddbab3c770b1cf50f0351ba0577091ef2410c3951d835ff87"
 
 [[package]]
 name = "sentry"
-version = "0.37.0"
+version = "0.32.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "255914a8e53822abd946e2ce8baa41d4cded6b8e938913b7f7b9da5b7ab44335"
+checksum = "00421ed8fa0c995f07cde48ba6c89e80f2b312f74ff637326f392fbfd23abe02"
 dependencies = [
  "httpdate",
  "reqwest",
- "rustls 0.23.18",
+ "rustls 0.21.12",
  "sentry-backtrace",
  "sentry-contexts",
  "sentry-core",
@@ -6209,14 +6209,14 @@ dependencies = [
  "sentry-tracing",
  "tokio",
  "ureq",
- "webpki-roots",
+ "webpki-roots 0.25.2",
 ]
 
 [[package]]
 name = "sentry-backtrace"
-version = "0.37.0"
+version = "0.32.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "00293cd332a859961f24fd69258f7e92af736feaeb91020cff84dac4188a4302"
+checksum = "a79194074f34b0cbe5dd33896e5928bbc6ab63a889bd9df2264af5acb186921e"
 dependencies = [
  "backtrace",
  "once_cell",
@@ -6226,9 +6226,9 @@ dependencies = [
 
 [[package]]
 name = "sentry-contexts"
-version = "0.37.0"
+version = "0.32.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "961990f9caa76476c481de130ada05614cd7f5aa70fb57c2142f0e09ad3fb2aa"
+checksum = "eba8870c5dba2bfd9db25c75574a11429f6b95957b0a78ac02e2970dd7a5249a"
 dependencies = [
  "hostname",
  "libc",
@@ -6240,9 +6240,9 @@ dependencies = [
 
 [[package]]
 name = "sentry-core"
-version = "0.37.0"
+version = "0.32.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "1a6409d845707d82415c800290a5d63be5e3df3c2e417b0997c60531dfbd35ef"
+checksum = "46a75011ea1c0d5c46e9e57df03ce81f5c7f0a9e199086334a1f9c0a541e0826"
 dependencies = [
  "once_cell",
  "rand 0.8.5",
@@ -6253,9 +6253,9 @@ dependencies = [
 
 [[package]]
 name = "sentry-panic"
-version = "0.37.0"
+version = "0.32.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "609b1a12340495ce17baeec9e08ff8ed423c337c1a84dffae36a178c783623f3"
+checksum = "2eaa3ecfa3c8750c78dcfd4637cfa2598b95b52897ed184b4dc77fcf7d95060d"
 dependencies = [
  "sentry-backtrace",
  "sentry-core",
@@ -6263,9 +6263,9 @@ dependencies = [
 
 [[package]]
 name = "sentry-tracing"
-version = "0.37.0"
+version = "0.32.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "49f4e86402d5c50239dc7d8fd3f6d5e048221d5fcb4e026d8d50ab57fe4644cb"
+checksum = "f715932bf369a61b7256687c6f0554141b7ce097287e30e3f7ed6e9de82498fe"
 dependencies = [
  "sentry-backtrace",
  "sentry-core",
@@ -6275,9 +6275,9 @@ dependencies = [
 
 [[package]]
 name = "sentry-types"
-version = "0.37.0"
+version = "0.32.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "3d3f117b8755dbede8260952de2aeb029e20f432e72634e8969af34324591631"
+checksum = "4519c900ce734f7a0eb7aba0869dfb225a7af8820634a7dd51449e3b093cfb7c"
 dependencies = [
  "debugid",
  "hex",
@@ -6616,14 +6616,12 @@ dependencies = [
  "anyhow",
  "async-stream",
  "bytes",
- "camino",
  "clap",
  "const_format",
  "futures",
  "futures-core",
  "futures-util",
  "http-body-util",
- "http-utils",
  "humantime",
  "hyper 1.4.1",
  "hyper-util",
@@ -6633,7 +6631,6 @@ dependencies = [
  "prost 0.13.3",
  "rustls 0.23.18",
  "tokio",
- "tokio-rustls 0.26.0",
  "tonic",
  "tonic-build",
  "tracing",
@@ -6714,6 +6711,8 @@ version = "0.1.0"
 dependencies = [
  "anyhow",
  "async-stream",
+ "aws-config",
+ "aws-sdk-s3",
  "camino",
  "chrono",
  "clap",
@@ -7802,7 +7801,7 @@ dependencies = [
  "rustls 0.23.18",
  "rustls-pki-types",
  "url",
- "webpki-roots",
+ "webpki-roots 0.26.1",
 ]
 
 [[package]]
@@ -8170,6 +8169,12 @@ dependencies = [
  "wasm-bindgen",
 ]
 
+[[package]]
+name = "webpki-roots"
+version = "0.25.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "14247bb57be4f377dfb94c72830b8ce8fc6beac03cf4bf7b9732eadd414123fc"
+
 [[package]]
 name = "webpki-roots"
 version = "0.26.1"
@@ -8477,8 +8482,6 @@ dependencies = [
  "regex-syntax 0.8.2",
  "reqwest",
  "rustls 0.23.18",
- "rustls-pki-types",
- "rustls-webpki 0.102.8",
  "scopeguard",
  "sec1 0.7.3",
  "serde",
@@ -8507,6 +8510,7 @@ dependencies = [
  "tracing-log",
  "url",
  "uuid",
+ "zerocopy",
  "zeroize",
  "zstd",
  "zstd-safe",
@@ -8610,16 +8614,8 @@ version = "0.7.31"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "1c4061bedbb353041c12f413700357bec76df2c7e2ca8e4df8bac24c6bf68e3d"
 dependencies = [
- "zerocopy-derive 0.7.31",
-]
-
-[[package]]
-name = "zerocopy"
-version = "0.8.24"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "2586fea28e186957ef732a5f8b3be2da217d65c5969d4b1e17f973ebbe876879"
-dependencies = [
- "zerocopy-derive 0.8.24",
+ "byteorder",
+ "zerocopy-derive",
 ]
 
 [[package]]
@@ -8633,17 +8629,6 @@ dependencies = [
  "syn 2.0.100",
 ]
 
-[[package]]
-name = "zerocopy-derive"
-version = "0.8.24"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "a996a8f63c5c4448cd959ac1bab0aaa3306ccfd060472f85943ee0750f0169be"
-dependencies = [
- "proc-macro2",
- "quote",
- "syn 2.0.100",
-]
-
 [[package]]
 name = "zerofrom"
 version = "0.1.5"

Cargo.toml

@@ -164,7 +164,7 @@ scopeguard = "1.1"
 sysinfo = "0.29.2"
 sd-notify = "0.4.1"
 send-future = "0.1.0"
-sentry = { version = "0.37", default-features = false, features = ["backtrace", "contexts", "panic", "rustls", "reqwest" ] }
+sentry = { version = "0.32", default-features = false, features = ["backtrace", "contexts", "panic", "rustls", "reqwest" ] }
 serde = { version = "1.0", features = ["derive"] }
 serde_json = "1"
 serde_path_to_error = "0.1"
@@ -220,7 +220,7 @@ uuid = { version = "1.6.1", features = ["v4", "v7", "serde"] }
 walkdir = "2.3.2"
 rustls-native-certs = "0.8"
 whoami = "1.5.1"
-zerocopy = { version = "0.8", features = ["derive", "simd"] }
+zerocopy = { version = "0.7", features = ["derive"] }
 json-structural-diff = { version = "0.2.0" }
 x509-cert = { version = "0.2.5" }
 

build-tools.Dockerfile

@@ -173,7 +173,7 @@ RUN curl -fsSL "https://github.com/protocolbuffers/protobuf/releases/download/v$
     && rm -rf protoc.zip protoc
 
 # s5cmd
-ENV S5CMD_VERSION=2.3.0
+ENV S5CMD_VERSION=2.2.2
 RUN curl -sL "https://github.com/peak/s5cmd/releases/download/v${S5CMD_VERSION}/s5cmd_${S5CMD_VERSION}_Linux-$(uname -m | sed 's/x86_64/64bit/g' | sed 's/aarch64/arm64/g').tar.gz" | tar zxvf - s5cmd \
     && chmod +x s5cmd \
     && mv s5cmd /usr/local/bin/s5cmd
@@ -206,7 +206,7 @@ RUN curl "https://awscli.amazonaws.com/awscli-exe-linux-$(uname -m).zip" -o "aws
     && rm awscliv2.zip
 
 # Mold: A Modern Linker
-ENV MOLD_VERSION=v2.37.1
+ENV MOLD_VERSION=v2.34.1
 RUN set -e \
     && git clone https://github.com/rui314/mold.git \
     && mkdir mold/build \
@@ -268,7 +268,7 @@ WORKDIR /home/nonroot
 RUN echo -e "--retry-connrefused\n--connect-timeout 15\n--retry 5\n--max-time 300\n" > /home/nonroot/.curlrc
 
 # Python
-ENV PYTHON_VERSION=3.11.12 \
+ENV PYTHON_VERSION=3.11.10 \
     PYENV_ROOT=/home/nonroot/.pyenv \
     PATH=/home/nonroot/.pyenv/shims:/home/nonroot/.pyenv/bin:/home/nonroot/.poetry/bin:$PATH
 RUN set -e \
@@ -296,12 +296,12 @@ ENV RUSTC_VERSION=1.86.0
 ENV RUSTUP_HOME="/home/nonroot/.rustup"
 ENV PATH="/home/nonroot/.cargo/bin:${PATH}"
 ARG RUSTFILT_VERSION=0.2.1
-ARG CARGO_HAKARI_VERSION=0.9.36
-ARG CARGO_DENY_VERSION=0.18.2
-ARG CARGO_HACK_VERSION=0.6.36
-ARG CARGO_NEXTEST_VERSION=0.9.94
+ARG CARGO_HAKARI_VERSION=0.9.33
+ARG CARGO_DENY_VERSION=0.16.2
+ARG CARGO_HACK_VERSION=0.6.33
+ARG CARGO_NEXTEST_VERSION=0.9.85
 ARG CARGO_CHEF_VERSION=0.1.71
-ARG CARGO_DIESEL_CLI_VERSION=2.2.9
+ARG CARGO_DIESEL_CLI_VERSION=2.2.6
 RUN curl -sSO https://static.rust-lang.org/rustup/dist/$(uname -m)-unknown-linux-gnu/rustup-init && whoami && \
 	chmod +x rustup-init && \
 	./rustup-init -y --default-toolchain ${RUSTC_VERSION} && \

compute/compute-node.Dockerfile

@@ -1800,8 +1800,8 @@ COPY compute/patches/pg_repack.patch /ext-src
 RUN cd /ext-src/pg_repack-src && patch -p1 </ext-src/pg_repack.patch && rm -f /ext-src/pg_repack.patch
 
 COPY --chmod=755 docker-compose/run-tests.sh /run-tests.sh
-RUN apt-get update && apt-get install -y libtap-parser-sourcehandler-pgtap-perl jq \
-   && apt clean && rm -rf /ext-src/*.tar.gz /ext-src/*.patch /var/lib/apt/lists/*
+RUN apt-get update && apt-get install -y libtap-parser-sourcehandler-pgtap-perl\
+   && apt clean && rm -rf /ext-src/*.tar.gz /var/lib/apt/lists/*
 ENV PATH=/usr/local/pgsql/bin:$PATH
 ENV PGHOST=compute
 ENV PGPORT=55433

compute_tools/Cargo.toml

@@ -44,6 +44,7 @@ serde.workspace = true
 serde_with.workspace = true
 serde_json.workspace = true
 signal-hook.workspace = true
+spki = { version = "0.7.3", features = ["std"] }
 tar.workspace = true
 tower.workspace = true
 tower-http.workspace = true

compute_tools/src/bin/compute_ctl.rs

@@ -57,24 +57,13 @@ use tracing::{error, info};
 use url::Url;
 use utils::failpoint_support;
 
-// Compatibility hack: if the control plane specified any remote-ext-config
-// use the default value for extension storage proxy gateway.
-// Remove this once the control plane is updated to pass the gateway URL
-fn parse_remote_ext_config(arg: &str) -> Result<String> {
-    if arg.starts_with("http") {
-        Ok(arg.trim_end_matches('/').to_string())
-    } else {
-        Ok("http://pg-ext-s3-gateway".to_string())
-    }
-}
-
 #[derive(Parser)]
 #[command(rename_all = "kebab-case")]
 struct Cli {
     #[arg(short = 'b', long, default_value = "postgres", env = "POSTGRES_PATH")]
     pub pgbin: String,
 
-    #[arg(short = 'r', long, value_parser = parse_remote_ext_config)]
+    #[arg(short = 'r', long)]
     pub remote_ext_config: Option<String>,
 
     /// The port to bind the external listening HTTP server to. Clients running

compute_tools/src/metrics.rs

@@ -1,8 +1,8 @@
-use metrics::core::{AtomicF64, AtomicU64, Collector, GenericCounter, GenericGauge};
+use metrics::core::{AtomicF64, Collector, GenericGauge};
 use metrics::proto::MetricFamily;
 use metrics::{
-    IntCounterVec, IntGaugeVec, UIntGaugeVec, register_gauge, register_int_counter,
-    register_int_counter_vec, register_int_gauge_vec, register_uint_gauge_vec,
+    IntCounterVec, IntGaugeVec, UIntGaugeVec, register_gauge, register_int_counter_vec,
+    register_int_gauge_vec, register_uint_gauge_vec,
 };
 use once_cell::sync::Lazy;
 
@@ -81,22 +81,6 @@ pub(crate) static COMPUTE_CTL_UP: Lazy<IntGaugeVec> = Lazy::new(|| {
     .expect("failed to define a metric")
 });
 
-pub(crate) static PG_CURR_DOWNTIME_MS: Lazy<GenericGauge<AtomicF64>> = Lazy::new(|| {
-    register_gauge!(
-        "compute_pg_current_downtime_ms",
-        "Non-cumulative duration of Postgres downtime in ms; resets after successful check",
-    )
-    .expect("failed to define a metric")
-});
-
-pub(crate) static PG_TOTAL_DOWNTIME_MS: Lazy<GenericCounter<AtomicU64>> = Lazy::new(|| {
-    register_int_counter!(
-        "compute_pg_downtime_ms_total",
-        "Cumulative duration of Postgres downtime in ms",
-    )
-    .expect("failed to define a metric")
-});
-
 pub fn collect() -> Vec<MetricFamily> {
     let mut metrics = COMPUTE_CTL_UP.collect();
     metrics.extend(INSTALLED_EXTENSIONS.collect());
@@ -104,7 +88,5 @@ pub fn collect() -> Vec<MetricFamily> {
     metrics.extend(REMOTE_EXT_REQUESTS_TOTAL.collect());
     metrics.extend(DB_MIGRATION_FAILED.collect());
     metrics.extend(AUDIT_LOG_DIR_SIZE.collect());
-    metrics.extend(PG_CURR_DOWNTIME_MS.collect());
-    metrics.extend(PG_TOTAL_DOWNTIME_MS.collect());
     metrics
 }

compute_tools/src/monitor.rs

@@ -6,294 +6,197 @@ use chrono::{DateTime, Utc};
 use compute_api::responses::ComputeStatus;
 use compute_api::spec::ComputeFeature;
 use postgres::{Client, NoTls};
-use tracing::{Level, error, info, instrument, span};
+use tracing::{debug, error, info, warn};
 
 use crate::compute::ComputeNode;
-use crate::metrics::{PG_CURR_DOWNTIME_MS, PG_TOTAL_DOWNTIME_MS};
 
 const MONITOR_CHECK_INTERVAL: Duration = Duration::from_millis(500);
 
-struct ComputeMonitor {
-    compute: Arc<ComputeNode>,
+// Spin in a loop and figure out the last activity time in the Postgres.
+// Then update it in the shared state. This function never errors out.
+// NB: the only expected panic is at `Mutex` unwrap(), all other errors
+// should be handled gracefully.
+fn watch_compute_activity(compute: &ComputeNode) {
+    // Suppose that `connstr` doesn't change
+    let connstr = compute.params.connstr.clone();
+    let conf = compute.get_conn_conf(Some("compute_ctl:activity_monitor"));
 
-    /// The moment when Postgres had some activity,
-    /// that should prevent compute from being suspended.
-    last_active: Option<DateTime<Utc>>,
+    // During startup and configuration we connect to every Postgres database,
+    // but we don't want to count this as some user activity. So wait until
+    // the compute fully started before monitoring activity.
+    wait_for_postgres_start(compute);
 
-    /// The moment when we last tried to check Postgres.
-    last_checked: DateTime<Utc>,
-    /// The last moment we did a successful Postgres check.
-    last_up: DateTime<Utc>,
+    // Define `client` outside of the loop to reuse existing connection if it's active.
+    let mut client = conf.connect(NoTls);
 
-    /// Only used for internal statistics change tracking
-    /// between monitor runs and can be outdated.
-    active_time: Option<f64>,
-    /// Only used for internal statistics change tracking
-    /// between monitor runs and can be outdated.
-    sessions: Option<i64>,
+    let mut sleep = false;
+    let mut prev_active_time: Option<f64> = None;
+    let mut prev_sessions: Option<i64> = None;
 
-    /// Use experimental statistics-based activity monitor. It's no longer
-    /// 'experimental' per se, as it's enabled for everyone, but we still
-    /// keep the flag as an option to turn it off in some cases if it will
-    /// misbehave.
-    experimental: bool,
-}
-
-impl ComputeMonitor {
-    fn report_down(&self) {
-        let now = Utc::now();
-
-        // Calculate and report current downtime
-        // (since the last time Postgres was up)
-        let downtime = now.signed_duration_since(self.last_up);
-        PG_CURR_DOWNTIME_MS.set(downtime.num_milliseconds() as f64);
-
-        // Calculate and update total downtime
-        // (cumulative duration of Postgres downtime in ms)
-        let inc = now
-            .signed_duration_since(self.last_checked)
-            .num_milliseconds();
-        PG_TOTAL_DOWNTIME_MS.inc_by(inc as u64);
+    if compute.has_feature(ComputeFeature::ActivityMonitorExperimental) {
+        info!("starting experimental activity monitor for {}", connstr);
+    } else {
+        info!("starting activity monitor for {}", connstr);
     }
 
-    fn report_up(&mut self) {
-        self.last_up = Utc::now();
-        PG_CURR_DOWNTIME_MS.set(0.0);
-    }
-
-    fn downtime_info(&self) -> String {
-        format!(
-            "total_ms: {}, current_ms: {}, last_up: {}",
-            PG_TOTAL_DOWNTIME_MS.get(),
-            PG_CURR_DOWNTIME_MS.get(),
-            self.last_up
-        )
-    }
-
-    /// Spin in a loop and figure out the last activity time in the Postgres.
-    /// Then update it in the shared state. This function never errors out.
-    /// NB: the only expected panic is at `Mutex` unwrap(), all other errors
-    /// should be handled gracefully.
-    #[instrument(skip_all)]
-    pub fn run(&mut self) {
-        // Suppose that `connstr` doesn't change
-        let connstr = self.compute.params.connstr.clone();
-        let conf = self
-            .compute
-            .get_conn_conf(Some("compute_ctl:compute_monitor"));
-
-        // During startup and configuration we connect to every Postgres database,
-        // but we don't want to count this as some user activity. So wait until
-        // the compute fully started before monitoring activity.
-        wait_for_postgres_start(&self.compute);
-
-        // Define `client` outside of the loop to reuse existing connection if it's active.
-        let mut client = conf.connect(NoTls);
-
-        info!("starting compute monitor for {}", connstr);
-
-        loop {
-            match &mut client {
-                Ok(cli) => {
-                    if cli.is_closed() {
-                        info!(
-                            downtime_info = self.downtime_info(),
-                            "connection to Postgres is closed, trying to reconnect"
-                        );
-                        self.report_down();
-
-                        // Connection is closed, reconnect and try again.
-                        client = conf.connect(NoTls);
-                    } else {
-                        match self.check(cli) {
-                            Ok(_) => {
-                                self.report_up();
-                                self.compute.update_last_active(self.last_active);
-                            }
-                            Err(e) => {
-                                // Although we have many places where we can return errors in `check()`,
-                                // normally it shouldn't happen. I.e., we will likely return error if
-                                // connection got broken, query timed out, Postgres returned invalid data, etc.
-                                // In all such cases it's suspicious, so let's report this as downtime.
-                                self.report_down();
-                                error!(
-                                    downtime_info = self.downtime_info(),
-                                    "could not check Postgres: {}", e
-                                );
-
-                                // Reconnect to Postgres just in case. During tests, I noticed
-                                // that queries in `check()` can fail with `connection closed`,
-                                // but `cli.is_closed()` above doesn't detect it. Even if old
-                                // connection is still alive, it will be dropped when we reassign
-                                // `client` to a new connection.
-                                client = conf.connect(NoTls);
-                            }
-                        }
-                    }
-                }
-                Err(e) => {
-                    info!(
-                        downtime_info = self.downtime_info(),
-                        "could not connect to Postgres: {}, retrying", e
-                    );
-                    self.report_down();
-
-                    // Establish a new connection and try again.
-                    client = conf.connect(NoTls);
-                }
-            }
-
-            // Reset the `last_checked` timestamp and sleep before the next iteration.
-            self.last_checked = Utc::now();
+    loop {
+        // We use `continue` a lot, so it's more convenient to sleep at the top of the loop.
+        // But skip the first sleep, so we can connect to Postgres immediately.
+        if sleep {
+            // Should be outside of the mutex lock to allow others to read while we sleep.
             thread::sleep(MONITOR_CHECK_INTERVAL);
+        } else {
+            sleep = true;
         }
-    }
 
-    #[instrument(skip_all)]
-    fn check(&mut self, cli: &mut Client) -> anyhow::Result<()> {
-        // This is new logic, only enable if the feature flag is set.
-        // TODO: remove this once we are sure that it works OR drop it altogether.
-        if self.experimental {
-            // Check if the total active time or sessions across all databases has changed.
-            // If it did, it means that user executed some queries. In theory, it can even go down if
-            // some databases were dropped, but it's still user activity.
-            match get_database_stats(cli) {
-                Ok((active_time, sessions)) => {
-                    let mut detected_activity = false;
+        match &mut client {
+            Ok(cli) => {
+                if cli.is_closed() {
+                    info!("connection to Postgres is closed, trying to reconnect");
 
-                    if let Some(prev_active_time) = self.active_time {
-                        if active_time != prev_active_time {
-                            detected_activity = true;
+                    // Connection is closed, reconnect and try again.
+                    client = conf.connect(NoTls);
+                    continue;
+                }
+
+                // This is a new logic, only enable if the feature flag is set.
+                // TODO: remove this once we are sure that it works OR drop it altogether.
+                if compute.has_feature(ComputeFeature::ActivityMonitorExperimental) {
+                    // First, check if the total active time or sessions across all databases has changed.
+                    // If it did, it means that user executed some queries. In theory, it can even go down if
+                    // some databases were dropped, but it's still a user activity.
+                    match get_database_stats(cli) {
+                        Ok((active_time, sessions)) => {
+                            let mut detected_activity = false;
+
+                            prev_active_time = match prev_active_time {
+                                Some(prev_active_time) => {
+                                    if active_time != prev_active_time {
+                                        detected_activity = true;
+                                    }
+                                    Some(active_time)
+                                }
+                                None => Some(active_time),
+                            };
+                            prev_sessions = match prev_sessions {
+                                Some(prev_sessions) => {
+                                    if sessions != prev_sessions {
+                                        detected_activity = true;
+                                    }
+                                    Some(sessions)
+                                }
+                                None => Some(sessions),
+                            };
+
+                            if detected_activity {
+                                // Update the last active time and continue, we don't need to
+                                // check backends state change.
+                                compute.update_last_active(Some(Utc::now()));
+                                continue;
+                            }
+                        }
+                        Err(e) => {
+                            error!("could not get database statistics: {}", e);
+                            continue;
                         }
                     }
-                    self.active_time = Some(active_time);
+                }
 
-                    if let Some(prev_sessions) = self.sessions {
-                        if sessions != prev_sessions {
-                            detected_activity = true;
+                // Second, if database statistics is the same, check all backends state change,
+                // maybe there is some with more recent activity. `get_backends_state_change()`
+                // can return None or stale timestamp, so it's `compute.update_last_active()`
+                // responsibility to check if the new timestamp is more recent than the current one.
+                // This helps us to discover new sessions, that did nothing yet.
+                match get_backends_state_change(cli) {
+                    Ok(last_active) => {
+                        compute.update_last_active(last_active);
+                    }
+                    Err(e) => {
+                        error!("could not get backends state change: {}", e);
+                    }
+                }
+
+                // Finally, if there are existing (logical) walsenders, do not suspend.
+                //
+                // walproposer doesn't currently show up in pg_stat_replication,
+                // but protect if it will be
+                let ws_count_query = "select count(*) from pg_stat_replication where application_name != 'walproposer';";
+                match cli.query_one(ws_count_query, &[]) {
+                    Ok(r) => match r.try_get::<&str, i64>("count") {
+                        Ok(num_ws) => {
+                            if num_ws > 0 {
+                                compute.update_last_active(Some(Utc::now()));
+                                continue;
+                            }
                         }
-                    }
-                    self.sessions = Some(sessions);
-
-                    if detected_activity {
-                        // Update the last active time and continue, we don't need to
-                        // check backends state change.
-                        self.last_active = Some(Utc::now());
-                        return Ok(());
+                        Err(e) => {
+                            warn!("failed to parse walsenders count: {:?}", e);
+                            continue;
+                        }
+                    },
+                    Err(e) => {
+                        warn!("failed to get list of walsenders: {:?}", e);
+                        continue;
                     }
                 }
-                Err(e) => {
-                    return Err(anyhow::anyhow!("could not get database statistics: {}", e));
+                //
+                // Don't suspend compute if there is an active logical replication subscription
+                //
+                // `where pid is not null` – to filter out read only computes and subscription on branches
+                //
+                let logical_subscriptions_query =
+                    "select count(*) from pg_stat_subscription where pid is not null;";
+                match cli.query_one(logical_subscriptions_query, &[]) {
+                    Ok(row) => match row.try_get::<&str, i64>("count") {
+                        Ok(num_subscribers) => {
+                            if num_subscribers > 0 {
+                                compute.update_last_active(Some(Utc::now()));
+                                continue;
+                            }
+                        }
+                        Err(e) => {
+                            warn!("failed to parse `pg_stat_subscription` count: {:?}", e);
+                            continue;
+                        }
+                    },
+                    Err(e) => {
+                        warn!(
+                            "failed to get list of active logical replication subscriptions: {:?}",
+                            e
+                        );
+                        continue;
+                    }
+                }
+                //
+                // Do not suspend compute if autovacuum is running
+                //
+                let autovacuum_count_query = "select count(*) from pg_stat_activity where backend_type = 'autovacuum worker'";
+                match cli.query_one(autovacuum_count_query, &[]) {
+                    Ok(r) => match r.try_get::<&str, i64>("count") {
+                        Ok(num_workers) => {
+                            if num_workers > 0 {
+                                compute.update_last_active(Some(Utc::now()));
+                                continue;
+                            }
+                        }
+                        Err(e) => {
+                            warn!("failed to parse autovacuum workers count: {:?}", e);
+                            continue;
+                        }
+                    },
+                    Err(e) => {
+                        warn!("failed to get list of autovacuum workers: {:?}", e);
+                        continue;
+                    }
                 }
             }
-        }
-
-        // If database statistics are the same, check all backends for state changes.
-        // Maybe there are some with more recent activity. `get_backends_state_change()`
-        // can return None or stale timestamp, so it's `compute.update_last_active()`
-        // responsibility to check if the new timestamp is more recent than the current one.
-        // This helps us to discover new sessions that have not done anything yet.
-        match get_backends_state_change(cli) {
-            Ok(last_active) => match (last_active, self.last_active) {
-                (Some(last_active), Some(prev_last_active)) => {
-                    if last_active > prev_last_active {
-                        self.last_active = Some(last_active);
-                        return Ok(());
-                    }
-                }
-                (Some(last_active), None) => {
-                    self.last_active = Some(last_active);
-                    return Ok(());
-                }
-                _ => {}
-            },
             Err(e) => {
-                return Err(anyhow::anyhow!(
-                    "could not get backends state change: {}",
-                    e
-                ));
+                debug!("could not connect to Postgres: {}, retrying", e);
+
+                // Establish a new connection and try again.
+                client = conf.connect(NoTls);
             }
         }
-
-        // If there are existing (logical) walsenders, do not suspend.
-        //
-        // N.B. walproposer doesn't currently show up in pg_stat_replication,
-        // but protect if it will.
-        const WS_COUNT_QUERY: &str =
-            "select count(*) from pg_stat_replication where application_name != 'walproposer';";
-        match cli.query_one(WS_COUNT_QUERY, &[]) {
-            Ok(r) => match r.try_get::<&str, i64>("count") {
-                Ok(num_ws) => {
-                    if num_ws > 0 {
-                        self.last_active = Some(Utc::now());
-                        return Ok(());
-                    }
-                }
-                Err(e) => {
-                    let err: anyhow::Error = e.into();
-                    return Err(err.context("failed to parse walsenders count"));
-                }
-            },
-            Err(e) => {
-                return Err(anyhow::anyhow!("failed to get list of walsenders: {}", e));
-            }
-        }
-
-        // Don't suspend compute if there is an active logical replication subscription
-        //
-        // `where pid is not null` – to filter out read only computes and subscription on branches
-        const LOGICAL_SUBSCRIPTIONS_QUERY: &str =
-            "select count(*) from pg_stat_subscription where pid is not null;";
-        match cli.query_one(LOGICAL_SUBSCRIPTIONS_QUERY, &[]) {
-            Ok(row) => match row.try_get::<&str, i64>("count") {
-                Ok(num_subscribers) => {
-                    if num_subscribers > 0 {
-                        self.last_active = Some(Utc::now());
-                        return Ok(());
-                    }
-                }
-                Err(e) => {
-                    return Err(anyhow::anyhow!(
-                        "failed to parse 'pg_stat_subscription' count: {}",
-                        e
-                    ));
-                }
-            },
-            Err(e) => {
-                return Err(anyhow::anyhow!(
-                    "failed to get list of active logical replication subscriptions: {}",
-                    e
-                ));
-            }
-        }
-
-        // Do not suspend compute if autovacuum is running
-        const AUTOVACUUM_COUNT_QUERY: &str =
-            "select count(*) from pg_stat_activity where backend_type = 'autovacuum worker'";
-        match cli.query_one(AUTOVACUUM_COUNT_QUERY, &[]) {
-            Ok(r) => match r.try_get::<&str, i64>("count") {
-                Ok(num_workers) => {
-                    if num_workers > 0 {
-                        self.last_active = Some(Utc::now());
-                        return Ok(());
-                    };
-                }
-                Err(e) => {
-                    return Err(anyhow::anyhow!(
-                        "failed to parse autovacuum workers count: {}",
-                        e
-                    ));
-                }
-            },
-            Err(e) => {
-                return Err(anyhow::anyhow!(
-                    "failed to get list of autovacuum workers: {}",
-                    e
-                ));
-            }
-        }
-
-        Ok(())
     }
 }
 
@@ -412,24 +315,9 @@ fn get_backends_state_change(cli: &mut Client) -> anyhow::Result<Option<DateTime
 /// Launch a separate compute monitor thread and return its `JoinHandle`.
 pub fn launch_monitor(compute: &Arc<ComputeNode>) -> thread::JoinHandle<()> {
     let compute = Arc::clone(compute);
-    let experimental = compute.has_feature(ComputeFeature::ActivityMonitorExperimental);
-    let now = Utc::now();
-    let mut monitor = ComputeMonitor {
-        compute,
-        last_active: None,
-        last_checked: now,
-        last_up: now,
-        active_time: None,
-        sessions: None,
-        experimental,
-    };
 
-    let span = span!(Level::INFO, "compute_monitor");
     thread::Builder::new()
         .name("compute-monitor".into())
-        .spawn(move || {
-            let _enter = span.enter();
-            monitor.run();
-        })
+        .spawn(move || watch_compute_activity(&compute))
         .expect("cannot launch compute monitor thread")
 }

compute_tools/src/tls.rs

@@ -3,6 +3,7 @@ use std::{io::Write, os::unix::fs::OpenOptionsExt, path::Path, time::Duration};
 use anyhow::{Context, Result, bail};
 use compute_api::responses::TlsConfig;
 use ring::digest;
+use spki::der::{Decode, PemReader};
 use x509_cert::Certificate;
 
 #[derive(Clone, Copy)]
@@ -51,7 +52,7 @@ pub fn update_key_path_blocking(pg_data: &Path, tls_config: &TlsConfig) {
         match try_update_key_path_blocking(pg_data, tls_config) {
             Ok(()) => break,
             Err(e) => {
-                tracing::error!(error = ?e, "could not create key file");
+                tracing::error!("could not create key file {e:?}");
                 std::thread::sleep(Duration::from_secs(1))
             }
         }
@@ -91,14 +92,8 @@ fn try_update_key_path_blocking(pg_data: &Path, tls_config: &TlsConfig) -> Resul
 fn verify_key_cert(key: &str, cert: &str) -> Result<()> {
     use x509_cert::der::oid::db::rfc5912::ECDSA_WITH_SHA_256;
 
-    let certs = Certificate::load_pem_chain(cert.as_bytes())
-        .context("decoding PEM encoded certificates")?;
-
-    // First certificate is our server-cert,
-    // all the rest of the certs are the CA cert chain.
-    let Some(cert) = certs.first() else {
-        bail!("no certificates found");
-    };
+    let cert = Certificate::decode(&mut PemReader::new(cert.as_bytes()).context("pem reader")?)
+        .context("decode cert")?;
 
     match cert.signature_algorithm.oid {
         ECDSA_WITH_SHA_256 => {
@@ -120,82 +115,3 @@ fn verify_key_cert(key: &str, cert: &str) -> Result<()> {
 
     Ok(())
 }
-
-#[cfg(test)]
-mod tests {
-    use super::verify_key_cert;
-
-    /// Real certificate chain file, generated by cert-manager in dev.
-    /// The server auth certificate has expired since 2025-04-24T15:41:35Z.
-    const CERT: &str = "
------BEGIN CERTIFICATE-----
-MIICCDCCAa+gAwIBAgIQKhLomFcNULbZA/bPdGzaSzAKBggqhkjOPQQDAjBEMQsw
-CQYDVQQGEwJVUzESMBAGA1UEChMJTmVvbiBJbmMuMSEwHwYDVQQDExhOZW9uIEs4
-cyBJbnRlcm1lZGlhdGUgQ0EwHhcNMjUwNDIzMTU0MTM1WhcNMjUwNDI0MTU0MTM1
-WjBBMT8wPQYDVQQDEzZjb21wdXRlLXdpc3B5LWdyYXNzLXcwY21laWp3LmRlZmF1
-bHQuc3ZjLmNsdXN0ZXIubG9jYWwwWTATBgcqhkjOPQIBBggqhkjOPQMBBwNCAATF
-QCcG2m/EVHAiZtSsYgVnHgoTjUL/Jtwfdrpvz2t0bVRZmBmSKhlo53uPV9Y5eKFG
-AmR54p9/gT2eO3xU7vAgo4GFMIGCMA4GA1UdDwEB/wQEAwIFoDAMBgNVHRMBAf8E
-AjAAMB8GA1UdIwQYMBaAFFR2JAhXkeiNQNEixTvAYIwxUu3QMEEGA1UdEQQ6MDiC
-NmNvbXB1dGUtd2lzcHktZ3Jhc3MtdzBjbWVpancuZGVmYXVsdC5zdmMuY2x1c3Rl
-ci5sb2NhbDAKBggqhkjOPQQDAgNHADBEAiBLG22wKG8XS9e9RxBT+kmUx/kIThcP
-DIpp7jx0PrFcdQIgEMTdnXpx5Cv/Z0NIEDxtMHUD7G0vuRPfztki36JuakM=
------END CERTIFICATE-----
------BEGIN CERTIFICATE-----
-MIICFzCCAb6gAwIBAgIUbbX98N2Ip6lWAONRk8dU9hSz+YIwCgYIKoZIzj0EAwIw
-RDELMAkGA1UEBhMCVVMxEjAQBgNVBAoTCU5lb24gSW5jLjEhMB8GA1UEAxMYTmVv
-biBBV1MgSW50ZXJtZWRpYXRlIENBMB4XDTI1MDQyMjE1MTAxMFoXDTI1MDcyMTE1
-MTAxMFowRDELMAkGA1UEBhMCVVMxEjAQBgNVBAoTCU5lb24gSW5jLjEhMB8GA1UE
-AxMYTmVvbiBLOHMgSW50ZXJtZWRpYXRlIENBMFkwEwYHKoZIzj0CAQYIKoZIzj0D
-AQcDQgAE5++m5owqNI4BPMTVNIUQH0qvU7pYhdpHGVGhdj/Lgars6ROvE6uSNQV4
-SAmJN5HBzj5/6kLQaTPWpXW7EHXjK6OBjTCBijAOBgNVHQ8BAf8EBAMCAQYwEgYD
-VR0TAQH/BAgwBgEB/wIBADAdBgNVHQ4EFgQUVHYkCFeR6I1A0SLFO8BgjDFS7dAw
-HwYDVR0jBBgwFoAUgHfNXfyKtHO0V9qoLOWCjkNiaI8wJAYDVR0eAQH/BBowGKAW
-MBSCEi5zdmMuY2x1c3Rlci5sb2NhbDAKBggqhkjOPQQDAgNHADBEAiBObVFFdXaL
-QpOXmN60dYUNnQRwjKreFduEkQgOdOlssgIgVAdJJQFgvlrvEOBhY8j5WyeKRwUN
-k/ALs6KpgaFBCGY=
------END CERTIFICATE-----
------BEGIN CERTIFICATE-----
-MIIB4jCCAYegAwIBAgIUFlxWFn/11yoGdmD+6gf+yQMToS0wCgYIKoZIzj0EAwIw
-ODELMAkGA1UEBhMCVVMxEjAQBgNVBAoTCU5lb24gSW5jLjEVMBMGA1UEAxMMTmVv
-biBSb290IENBMB4XDTI1MDQwMzA3MTUyMloXDTI2MDQwMzA3MTUyMlowRDELMAkG
-A1UEBhMCVVMxEjAQBgNVBAoTCU5lb24gSW5jLjEhMB8GA1UEAxMYTmVvbiBBV1Mg
-SW50ZXJtZWRpYXRlIENBMFkwEwYHKoZIzj0CAQYIKoZIzj0DAQcDQgAEqonG/IQ6
-ZxtEtOUTkkoNopPieXDO5CBKUkNFTGeJEB7OxRlSpYJgsBpaYIaD6Vc4sVk3thIF
-p+pLw52idQOIN6NjMGEwDgYDVR0PAQH/BAQDAgEGMA8GA1UdEwEB/wQFMAMBAf8w
-HQYDVR0OBBYEFIB3zV38irRztFfaqCzlgo5DYmiPMB8GA1UdIwQYMBaAFKh7M4/G
-FHvr/ORDQZt4bMLlJvHCMAoGCCqGSM49BAMCA0kAMEYCIQCbS4x7QPslONzBYbjC
-UQaQ0QLDW4CJHvQ4u4gbWFG87wIhAJMsHQHjP9qTT27Q65zQCR7O8QeLAfha1jrH
-Ag/LsxSr
------END CERTIFICATE-----
-";
-
-    /// The key corresponding to [`CERT`]
-    const KEY: &str = "
------BEGIN EC PRIVATE KEY-----
-MHcCAQEEIDnAnrqmIJjndCLWP1iIO5X3X63Aia48TGpGuMXwvm6IoAoGCCqGSM49
-AwEHoUQDQgAExUAnBtpvxFRwImbUrGIFZx4KE41C/ybcH3a6b89rdG1UWZgZkioZ
-aOd7j1fWOXihRgJkeeKff4E9njt8VO7wIA==
------END EC PRIVATE KEY-----
-";
-
-    /// An incorrect key.
-    const INCORRECT_KEY: &str = "
------BEGIN EC PRIVATE KEY-----
-MHcCAQEEIL6WqqBDyvM0HWz7Ir5M5+jhFWB7IzOClGn26OPrzHCXoAoGCCqGSM49
-AwEHoUQDQgAE7XVvdOy5lfwtNKb+gJEUtnG+DrnnXLY5LsHDeGQKV9PTRcEMeCrG
-YZzHyML4P6Sr4yi2ts+4B9i47uvAG8+XwQ==
------END EC PRIVATE KEY-----
-";
-
-    #[test]
-    fn certificate_verification() {
-        verify_key_cert(KEY, CERT).unwrap();
-    }
-
-    #[test]
-    #[should_panic(expected = "private key file does not match certificate")]
-    fn certificate_verification_fail() {
-        verify_key_cert(INCORRECT_KEY, CERT).unwrap();
-    }
-}

control_plane/src/bin/neon_local.rs

@@ -17,10 +17,8 @@ use std::time::Duration;
 use anyhow::{Context, Result, anyhow, bail};
 use clap::Parser;
 use compute_api::spec::ComputeMode;
-use control_plane::broker::StorageBroker;
 use control_plane::endpoint::ComputeControlPlane;
 use control_plane::endpoint_storage::{ENDPOINT_STORAGE_DEFAULT_PORT, EndpointStorage};
-use control_plane::local_env;
 use control_plane::local_env::{
     EndpointStorageConf, InitForceMode, LocalEnv, NeonBroker, NeonLocalInitConf,
     NeonLocalInitPageserverConf, SafekeeperConf,
@@ -30,6 +28,7 @@ use control_plane::safekeeper::SafekeeperNode;
 use control_plane::storage_controller::{
     NeonStorageControllerStartArgs, NeonStorageControllerStopArgs, StorageController,
 };
+use control_plane::{broker, local_env};
 use nix::fcntl::{FlockArg, flock};
 use pageserver_api::config::{
     DEFAULT_HTTP_LISTEN_PORT as DEFAULT_PAGESERVER_HTTP_PORT,
@@ -989,8 +988,7 @@ fn handle_init(args: &InitCmdArgs) -> anyhow::Result<LocalEnv> {
         NeonLocalInitConf {
             control_plane_api: Some(DEFAULT_PAGESERVER_CONTROL_PLANE_API.parse().unwrap()),
             broker: NeonBroker {
-                listen_addr: Some(DEFAULT_BROKER_ADDR.parse().unwrap()),
-                listen_https_addr: None,
+                listen_addr: DEFAULT_BROKER_ADDR.parse().unwrap(),
             },
             safekeepers: vec![SafekeeperConf {
                 id: DEFAULT_SAFEKEEPER_ID,
@@ -1779,8 +1777,7 @@ async fn handle_endpoint_storage(
 async fn handle_storage_broker(subcmd: &StorageBrokerCmd, env: &local_env::LocalEnv) -> Result<()> {
     match subcmd {
         StorageBrokerCmd::Start(args) => {
-            let storage_broker = StorageBroker::from_env(env);
-            if let Err(e) = storage_broker.start(&args.start_timeout).await {
+            if let Err(e) = broker::start_broker_process(env, &args.start_timeout).await {
                 eprintln!("broker start failed: {e}");
                 exit(1);
             }
@@ -1788,8 +1785,7 @@ async fn handle_storage_broker(subcmd: &StorageBrokerCmd, env: &local_env::Local
 
         StorageBrokerCmd::Stop(_args) => {
             // FIXME: stop_mode unused
-            let storage_broker = StorageBroker::from_env(env);
-            if let Err(e) = storage_broker.stop() {
+            if let Err(e) = broker::stop_broker_process(env) {
                 eprintln!("broker stop failed: {e}");
                 exit(1);
             }
@@ -1839,11 +1835,8 @@ async fn handle_start_all_impl(
     #[allow(clippy::redundant_closure_call)]
     (|| {
         js.spawn(async move {
-            let storage_broker = StorageBroker::from_env(env);
-            storage_broker
-                .start(&retry_timeout)
-                .await
-                .map_err(|e| e.context("start storage_broker"))
+            let retry_timeout = retry_timeout;
+            broker::start_broker_process(env, &retry_timeout).await
         });
 
         js.spawn(async move {
@@ -1998,8 +1991,7 @@ async fn try_stop_all(env: &local_env::LocalEnv, immediate: bool) {
         }
     }
 
-    let storage_broker = StorageBroker::from_env(env);
-    if let Err(e) = storage_broker.stop() {
+    if let Err(e) = broker::stop_broker_process(env) {
         eprintln!("neon broker stop failed: {e:#}");
     }
 

control_plane/src/broker.rs

@@ -3,86 +3,60 @@
 //! In the local test environment, the storage broker stores its data directly in
 //!
 //! ```text
-//!   .neon/storage_broker
+//!   .neon
 //! ```
 use std::time::Duration;
 
 use anyhow::Context;
 use camino::Utf8PathBuf;
 
-use crate::{background_process, local_env::LocalEnv};
+use crate::{background_process, local_env};
 
-pub struct StorageBroker {
-    env: LocalEnv,
+pub async fn start_broker_process(
+    env: &local_env::LocalEnv,
+    retry_timeout: &Duration,
+) -> anyhow::Result<()> {
+    let broker = &env.broker;
+    let listen_addr = &broker.listen_addr;
+
+    print!("Starting neon broker at {}", listen_addr);
+
+    let args = [format!("--listen-addr={listen_addr}")];
+
+    let client = reqwest::Client::new();
+    background_process::start_process(
+        "storage_broker",
+        &env.base_data_dir,
+        &env.storage_broker_bin(),
+        args,
+        [],
+        background_process::InitialPidFile::Create(storage_broker_pid_file_path(env)),
+        retry_timeout,
+        || async {
+            let url = broker.client_url();
+            let status_url = url.join("status").with_context(|| {
+                format!("Failed to append /status path to broker endpoint {url}")
+            })?;
+            let request = client
+                .get(status_url)
+                .build()
+                .with_context(|| format!("Failed to construct request to broker endpoint {url}"))?;
+            match client.execute(request).await {
+                Ok(resp) => Ok(resp.status().is_success()),
+                Err(_) => Ok(false),
+            }
+        },
+    )
+    .await
+    .context("Failed to spawn storage_broker subprocess")?;
+    Ok(())
 }
 
-impl StorageBroker {
-    /// Create a new `StorageBroker` instance from the environment.
-    pub fn from_env(env: &LocalEnv) -> Self {
-        Self { env: env.clone() }
-    }
-
-    pub fn initialize(&self) -> anyhow::Result<()> {
-        if self.env.generate_local_ssl_certs {
-            self.env.generate_ssl_cert(
-                &self.env.storage_broker_data_dir().join("server.crt"),
-                &self.env.storage_broker_data_dir().join("server.key"),
-            )?;
-        }
-        Ok(())
-    }
-
-    /// Start the storage broker process.
-    pub async fn start(&self, retry_timeout: &Duration) -> anyhow::Result<()> {
-        let broker = &self.env.broker;
-
-        print!("Starting neon broker at {}", broker.client_url());
-
-        let mut args = Vec::new();
-
-        if let Some(addr) = &broker.listen_addr {
-            args.push(format!("--listen-addr={addr}"));
-        }
-        if let Some(addr) = &broker.listen_https_addr {
-            args.push(format!("--listen-https-addr={addr}"));
-        }
-
-        let client = self.env.create_http_client();
-        background_process::start_process(
-            "storage_broker",
-            &self.env.storage_broker_data_dir(),
-            &self.env.storage_broker_bin(),
-            args,
-            [],
-            background_process::InitialPidFile::Create(self.pid_file_path()),
-            retry_timeout,
-            || async {
-                let url = broker.client_url();
-                let status_url = url.join("status").with_context(|| {
-                    format!("Failed to append /status path to broker endpoint {url}")
-                })?;
-                let request = client.get(status_url).build().with_context(|| {
-                    format!("Failed to construct request to broker endpoint {url}")
-                })?;
-                match client.execute(request).await {
-                    Ok(resp) => Ok(resp.status().is_success()),
-                    Err(_) => Ok(false),
-                }
-            },
-        )
-        .await
-        .context("Failed to spawn storage_broker subprocess")?;
-        Ok(())
-    }
-
-    /// Stop the storage broker process.
-    pub fn stop(&self) -> anyhow::Result<()> {
-        background_process::stop_process(true, "storage_broker", &self.pid_file_path())
-    }
-
-    /// Get the path to the PID file for the storage broker.
-    fn pid_file_path(&self) -> Utf8PathBuf {
-        Utf8PathBuf::from_path_buf(self.env.base_data_dir.join("storage_broker.pid"))
-            .expect("non-Unicode path")
-    }
+pub fn stop_broker_process(env: &local_env::LocalEnv) -> anyhow::Result<()> {
+    background_process::stop_process(true, "storage_broker", &storage_broker_pid_file_path(env))
+}
+
+fn storage_broker_pid_file_path(env: &local_env::LocalEnv) -> Utf8PathBuf {
+    Utf8PathBuf::from_path_buf(env.base_data_dir.join("storage_broker.pid"))
+        .expect("non-Unicode path")
 }

control_plane/src/local_env.rs

@@ -4,7 +4,7 @@
 //! script which will use local paths.
 
 use std::collections::HashMap;
-use std::net::SocketAddr;
+use std::net::{IpAddr, Ipv4Addr, SocketAddr};
 use std::path::{Path, PathBuf};
 use std::process::{Command, Stdio};
 use std::time::Duration;
@@ -14,12 +14,11 @@ use anyhow::{Context, bail};
 use clap::ValueEnum;
 use pem::Pem;
 use postgres_backend::AuthType;
-use reqwest::{Certificate, Url};
+use reqwest::Url;
 use serde::{Deserialize, Serialize};
 use utils::auth::encode_from_key_file;
 use utils::id::{NodeId, TenantId, TenantTimelineId, TimelineId};
 
-use crate::broker::StorageBroker;
 use crate::endpoint_storage::{ENDPOINT_STORAGE_REMOTE_STORAGE_DIR, EndpointStorage};
 use crate::pageserver::{PAGESERVER_REMOTE_STORAGE_DIR, PageServerNode};
 use crate::safekeeper::SafekeeperNode;
@@ -158,16 +157,11 @@ pub struct EndpointStorageConf {
 }
 
 /// Broker config for cluster internal communication.
-#[derive(Serialize, Deserialize, PartialEq, Eq, Clone, Debug, Default)]
+#[derive(Serialize, Deserialize, PartialEq, Eq, Clone, Debug)]
 #[serde(default)]
 pub struct NeonBroker {
-    /// Broker listen HTTP address for storage nodes coordination, e.g. '127.0.0.1:50051'.
-    /// At least one of listen_addr or listen_https_addr must be set.
-    pub listen_addr: Option<SocketAddr>,
-    /// Broker listen HTTPS address for storage nodes coordination, e.g. '127.0.0.1:50051'.
-    /// At least one of listen_addr or listen_https_addr must be set.
-    /// listen_https_addr is preferred over listen_addr in neon_local.
-    pub listen_https_addr: Option<SocketAddr>,
+    /// Broker listen address for storage nodes coordination, e.g. '127.0.0.1:50051'.
+    pub listen_addr: SocketAddr,
 }
 
 /// A part of storage controller's config the neon_local knows about.
@@ -241,19 +235,18 @@ impl Default for NeonStorageControllerConf {
     }
 }
 
+// Dummy Default impl to satisfy Deserialize derive.
+impl Default for NeonBroker {
+    fn default() -> Self {
+        NeonBroker {
+            listen_addr: SocketAddr::new(IpAddr::V4(Ipv4Addr::new(0, 0, 0, 0)), 0),
+        }
+    }
+}
+
 impl NeonBroker {
     pub fn client_url(&self) -> Url {
-        let url = if let Some(addr) = self.listen_https_addr {
-            format!("https://{}", addr)
-        } else {
-            format!(
-                "http://{}",
-                self.listen_addr
-                    .expect("at least one address should be set")
-            )
-        };
-
-        Url::parse(&url).expect("failed to construct url")
+        Url::parse(&format!("http://{}", self.listen_addr)).expect("failed to construct url")
     }
 }
 
@@ -448,10 +441,6 @@ impl LocalEnv {
         self.base_data_dir.join("endpoints")
     }
 
-    pub fn storage_broker_data_dir(&self) -> PathBuf {
-        self.base_data_dir.join("storage_broker")
-    }
-
     pub fn pageserver_data_dir(&self, pageserver_id: NodeId) -> PathBuf {
         self.base_data_dir
             .join(format!("pageserver_{pageserver_id}"))
@@ -514,23 +503,6 @@ impl LocalEnv {
         )
     }
 
-    /// Creates HTTP client with local SSL CA certificates.
-    pub fn create_http_client(&self) -> reqwest::Client {
-        let ssl_ca_certs = self.ssl_ca_cert_path().map(|ssl_ca_file| {
-            let buf = std::fs::read(ssl_ca_file).expect("SSL CA file should exist");
-            Certificate::from_pem_bundle(&buf).expect("SSL CA file should be valid")
-        });
-
-        let mut http_client = reqwest::Client::builder();
-        for ssl_ca_cert in ssl_ca_certs.unwrap_or_default() {
-            http_client = http_client.add_root_certificate(ssl_ca_cert);
-        }
-
-        http_client
-            .build()
-            .expect("HTTP client should construct with no error")
-    }
-
     /// Inspect the base data directory and extract the instance id and instance directory path
     /// for all storage controller instances
     pub async fn storage_controller_instances(&self) -> std::io::Result<Vec<(u8, PathBuf)>> {
@@ -939,12 +911,6 @@ impl LocalEnv {
         // create endpoints dir
         fs::create_dir_all(env.endpoints_path())?;
 
-        // create storage broker dir
-        fs::create_dir_all(env.storage_broker_data_dir())?;
-        StorageBroker::from_env(&env)
-            .initialize()
-            .context("storage broker init failed")?;
-
         // create safekeeper dirs
         for safekeeper in &env.safekeepers {
             fs::create_dir_all(SafekeeperNode::datadir_path_by_id(&env, safekeeper.id))?;

control_plane/src/pageserver.rs

@@ -21,6 +21,7 @@ use pageserver_api::shard::TenantShardId;
 use pageserver_client::mgmt_api;
 use postgres_backend::AuthType;
 use postgres_connection::{PgConnectionConfig, parse_host_port};
+use reqwest::Certificate;
 use utils::auth::{Claims, Scope};
 use utils::id::{NodeId, TenantId, TimelineId};
 use utils::lsn::Lsn;
@@ -50,6 +51,19 @@ impl PageServerNode {
             parse_host_port(&conf.listen_pg_addr).expect("Unable to parse listen_pg_addr");
         let port = port.unwrap_or(5432);
 
+        let ssl_ca_certs = env.ssl_ca_cert_path().map(|ssl_ca_file| {
+            let buf = std::fs::read(ssl_ca_file).expect("SSL root CA file should exist");
+            Certificate::from_pem_bundle(&buf).expect("SSL CA file should be valid")
+        });
+
+        let mut http_client = reqwest::Client::builder();
+        for ssl_ca_cert in ssl_ca_certs.unwrap_or_default() {
+            http_client = http_client.add_root_certificate(ssl_ca_cert);
+        }
+        let http_client = http_client
+            .build()
+            .expect("Client constructs with no errors");
+
         let endpoint = if env.storage_controller.use_https_pageserver_api {
             format!(
                 "https://{}",
@@ -66,7 +80,7 @@ impl PageServerNode {
             conf: conf.clone(),
             env: env.clone(),
             http_client: mgmt_api::Client::new(
-                env.create_http_client(),
+                http_client,
                 endpoint,
                 {
                     match conf.http_auth_type {

control_plane/src/safekeeper.rs

@@ -87,7 +87,7 @@ impl SafekeeperNode {
             conf: conf.clone(),
             pg_connection_config: Self::safekeeper_connection_config(&listen_addr, conf.pg_port),
             env: env.clone(),
-            http_client: env.create_http_client(),
+            http_client: reqwest::Client::new(),
             http_base_url: format!("http://{}:{}/v1", listen_addr, conf.http_port),
             listen_addr,
         }

control_plane/src/storage_controller.rs

@@ -20,7 +20,7 @@ use pageserver_api::shard::TenantShardId;
 use pageserver_client::mgmt_api::ResponseErrorMessageExt;
 use pem::Pem;
 use postgres_backend::AuthType;
-use reqwest::Method;
+use reqwest::{Certificate, Method};
 use serde::de::DeserializeOwned;
 use serde::{Deserialize, Serialize};
 use tokio::process::Command;
@@ -153,11 +153,24 @@ impl StorageController {
             }
         };
 
+        let ssl_ca_certs = env.ssl_ca_cert_path().map(|ssl_ca_file| {
+            let buf = std::fs::read(ssl_ca_file).expect("SSL CA file should exist");
+            Certificate::from_pem_bundle(&buf).expect("SSL CA file should be valid")
+        });
+
+        let mut http_client = reqwest::Client::builder();
+        for ssl_ca_cert in ssl_ca_certs.unwrap_or_default() {
+            http_client = http_client.add_root_certificate(ssl_ca_cert);
+        }
+        let http_client = http_client
+            .build()
+            .expect("HTTP client should construct with no error");
+
         Self {
             env: env.clone(),
             private_key,
             public_key,
-            client: env.create_http_client(),
+            client: http_client,
             config: env.storage_controller.clone(),
             listen_port: OnceLock::default(),
         }

deny.toml

@@ -45,7 +45,9 @@ allow = [
     "ISC",
     "MIT",
     "MPL-2.0",
+    "OpenSSL",
     "Unicode-3.0",
+    "Zlib",
 ]
 confidence-threshold = 0.8
 exceptions = [
@@ -54,6 +56,14 @@ exceptions = [
     { allow = ["Zlib"], name = "const_format", version = "*" },
 ]
 
+[[licenses.clarify]]
+name = "ring"
+version = "*"
+expression = "MIT AND ISC AND OpenSSL"
+license-files = [
+    { path = "LICENSE", hash = 0xbd0eed23 }
+]
+
 [licenses.private]
 ignore = true
 registries = []
@@ -106,11 +116,7 @@ name = "openssl"
 unknown-registry = "warn"
 unknown-git = "warn"
 allow-registry = ["https://github.com/rust-lang/crates.io-index"]
-allow-git = [
-    # Crate pinned to commit in origin repo due to opentelemetry version.
-    # TODO: Remove this once crate is fetched from crates.io again.
-    "https://github.com/mattiapenati/tower-otel",
-]
+allow-git = []
 
 [sources.allow-org]
 github = [

docker-compose/docker_compose_test.sh

@@ -9,20 +9,21 @@
 # to verify custom image builds (e.g pre-published ones).
 #
 # A test script for postgres extensions
-# Currently supports only v16+
+# Currently supports only v16
 #
 set -eux -o pipefail
 
-export COMPOSE_FILE='docker-compose.yml'
-export COMPOSE_PROFILES=test-extensions
-cd "$(dirname "${0}")"
+COMPOSE_FILE='docker-compose.yml'
+cd $(dirname $0)
+COMPUTE_CONTAINER_NAME=docker-compose-compute-1
+TEST_CONTAINER_NAME=docker-compose-neon-test-extensions-1
 PSQL_OPTION="-h localhost -U cloud_admin -p 55433 -d postgres"
 
-function cleanup() {
+cleanup() {
     echo "show container information"
     docker ps
     echo "stop containers..."
-    docker compose down
+    docker compose --profile test-extensions -f $COMPOSE_FILE down
 }
 
 for pg_version in ${TEST_VERSION_ONLY-14 15 16 17}; do
@@ -30,55 +31,55 @@ for pg_version in ${TEST_VERSION_ONLY-14 15 16 17}; do
     echo "clean up containers if exists"
     cleanup
     PG_TEST_VERSION=$((pg_version < 16 ? 16 : pg_version))
-    PG_VERSION=${pg_version} PG_TEST_VERSION=${PG_TEST_VERSION} docker compose up --quiet-pull --build -d
+    PG_VERSION=$pg_version PG_TEST_VERSION=$PG_TEST_VERSION docker compose --profile test-extensions -f $COMPOSE_FILE up --quiet-pull --build -d
 
     echo "wait until the compute is ready. timeout after 60s. "
     cnt=0
     while sleep 3; do
         # check timeout
-        (( cnt += 3 ))
-        if [[ ${cnt} -gt 60 ]]; then
+        cnt=`expr $cnt + 3`
+        if [ $cnt -gt 60 ]; then
             echo "timeout before the compute is ready."
             exit 1
         fi
-        if docker compose logs "compute_is_ready" | grep -q "accepting connections"; then
+        if docker compose --profile test-extensions -f $COMPOSE_FILE logs "compute_is_ready" | grep -q "accepting connections"; then
             echo "OK. The compute is ready to connect."
             echo "execute simple queries."
-            docker compose exec compute /bin/bash -c "psql ${PSQL_OPTION} -c 'SELECT 1'"
+            docker exec $COMPUTE_CONTAINER_NAME /bin/bash -c "psql $PSQL_OPTION"
             break
         fi
     done
 
-    if [[ ${pg_version} -ge 16 ]]; then
+    if [ $pg_version -ge 16 ]; then
         # This is required for the pg_hint_plan test, to prevent flaky log message causing the test to fail
         # It cannot be moved to Dockerfile now because the database directory is created after the start of the container
         echo Adding dummy config
-        docker compose exec compute touch /var/db/postgres/compute/compute_ctl_temp_override.conf
+        docker exec $COMPUTE_CONTAINER_NAME touch /var/db/postgres/compute/compute_ctl_temp_override.conf
         # The following block copies the files for the pg_hintplan test to the compute node for the extension test in an isolated docker-compose environment
         TMPDIR=$(mktemp -d)
-        docker compose cp neon-test-extensions:/ext-src/pg_hint_plan-src/data "${TMPDIR}/data"
-        docker compose cp "${TMPDIR}/data" compute:/ext-src/pg_hint_plan-src/
-        rm -rf "${TMPDIR}"
+        docker cp $TEST_CONTAINER_NAME:/ext-src/pg_hint_plan-src/data $TMPDIR/data
+        docker cp $TMPDIR/data $COMPUTE_CONTAINER_NAME:/ext-src/pg_hint_plan-src/
+        rm -rf $TMPDIR
         # The following block does the same for the contrib/file_fdw test
         TMPDIR=$(mktemp -d)
-        docker compose cp neon-test-extensions:/postgres/contrib/file_fdw/data "${TMPDIR}/data"
-        docker compose cp "${TMPDIR}/data" compute:/postgres/contrib/file_fdw/data
-        rm -rf "${TMPDIR}"
+        docker cp $TEST_CONTAINER_NAME:/postgres/contrib/file_fdw/data $TMPDIR/data
+        docker cp $TMPDIR/data $COMPUTE_CONTAINER_NAME:/postgres/contrib/file_fdw/data
+        rm -rf $TMPDIR
         # Apply patches
-        docker compose exec -T neon-test-extensions bash -c "(cd /postgres && patch -p1)" <"../compute/patches/contrib_pg${pg_version}.patch"
+        cat ../compute/patches/contrib_pg${pg_version}.patch | docker exec -i $TEST_CONTAINER_NAME bash -c "(cd /postgres && patch -p1)"
         # We are running tests now
         rm -f testout.txt testout_contrib.txt
-        docker compose exec -e USE_PGXS=1 -e SKIP=timescaledb-src,rdkit-src,postgis-src,pg_jsonschema-src,kq_imcx-src,wal2json_2_5-src,rag_jina_reranker_v1_tiny_en-src,rag_bge_small_en_v15-src \
-        neon-test-extensions /run-tests.sh /ext-src | tee testout.txt && EXT_SUCCESS=1 || EXT_SUCCESS=0
-        docker compose exec -e SKIP=start-scripts,postgres_fdw,ltree_plpython,jsonb_plpython,jsonb_plperl,hstore_plpython,hstore_plperl,dblink,bool_plperl \
-        neon-test-extensions /run-tests.sh /postgres/contrib | tee testout_contrib.txt && CONTRIB_SUCCESS=1 || CONTRIB_SUCCESS=0
-        if [[ ${EXT_SUCCESS} -eq 0 || ${CONTRIB_SUCCESS} -eq 0 ]]; then
+        docker exec -e USE_PGXS=1 -e SKIP=timescaledb-src,rdkit-src,postgis-src,pg_jsonschema-src,kq_imcx-src,wal2json_2_5-src,rag_jina_reranker_v1_tiny_en-src,rag_bge_small_en_v15-src \
+        $TEST_CONTAINER_NAME /run-tests.sh /ext-src | tee testout.txt && EXT_SUCCESS=1 || EXT_SUCCESS=0
+        docker exec -e SKIP=start-scripts,postgres_fdw,ltree_plpython,jsonb_plpython,jsonb_plperl,hstore_plpython,hstore_plperl,dblink,bool_plperl \
+        $TEST_CONTAINER_NAME /run-tests.sh /postgres/contrib | tee testout_contrib.txt && CONTRIB_SUCCESS=1 || CONTRIB_SUCCESS=0
+        if [ $EXT_SUCCESS -eq 0 ] || [ $CONTRIB_SUCCESS -eq 0 ]; then
             CONTRIB_FAILED=
             FAILED=
-            [[ ${EXT_SUCCESS} -eq 0 ]] && FAILED=$(tail -1 testout.txt | awk '{for(i=1;i<=NF;i++){print "/ext-src/"$i;}}')
-            [[ ${CONTRIB_SUCCESS} -eq 0 ]] && CONTRIB_FAILED=$(tail -1 testout_contrib.txt | awk '{for(i=0;i<=NF;i++){print "/postgres/contrib/"$i;}}')
-            for d in ${FAILED} ${CONTRIB_FAILED}; do
-                docker compose exec neon-test-extensions bash -c 'for file in $(find '"${d}"' -name regression.diffs -o -name regression.out); do cat ${file}; done' || [[ ${?} -eq 1 ]]
+            [ $EXT_SUCCESS -eq 0 ] && FAILED=$(tail -1 testout.txt | awk '{for(i=1;i<=NF;i++){print "/ext-src/"$i;}}')
+            [ $CONTRIB_SUCCESS -eq 0 ] && CONTRIB_FAILED=$(tail -1 testout_contrib.txt | awk '{for(i=0;i<=NF;i++){print "/postgres/contrib/"$i;}}')
+            for d in $FAILED $CONTRIB_FAILED; do
+                docker exec $TEST_CONTAINER_NAME bash -c 'for file in $(find '"$d"' -name regression.diffs -o -name regression.out); do cat $file; done' || [ $? -eq 1 ]
             done
         exit 1
         fi

docker-compose/ext-src/README.md

@@ -1,99 +0,0 @@
-# PostgreSQL Extensions for Testing
-
-This directory contains PostgreSQL extensions used primarily for:
-1. Testing extension upgrades between different Compute versions
-2. Running regression tests with regular users (mostly for cloud instances)
-
-## Directory Structure
-
-Each extension directory follows a standard structure:
-
-- `extension-name-src/` - Directory containing test files for the extension
-  - `test-upgrade.sh` - Script for testing upgrade scenarios
-  - `regular-test.sh` - Script for testing with regular users
-  - Additional test files depending on the extension
-
-## Available Extensions
-
-This directory includes the following extensions:
-
-- `hll-src` - HyperLogLog, a fixed-size data structure for approximating cardinality
-- `hypopg-src` - Extension to create hypothetical indexes
-- `ip4r-src` - IPv4/v6 and subnet data types
-- `pg_cron-src` - Run periodic jobs in PostgreSQL
-- `pg_graphql-src` - GraphQL support for PostgreSQL
-- `pg_hint_plan-src` - Execution plan hints
-- `pg_ivm-src` - Incremental view maintenance
-- `pg_jsonschema-src` - JSON Schema validation
-- `pg_repack-src` - Reorganize tables with minimal locks
-- `pg_roaringbitmap-src` - Roaring bitmap implementation
-- `pg_semver-src` - Semantic version data type
-- `pg_session_jwt-src` - JWT authentication for PostgreSQL
-- `pg_tiktoken-src` - OpenAI Tiktoken tokenizer
-- `pg_uuidv7-src` - UUIDv7 implementation for PostgreSQL
-- `pgjwt-src` - JWT tokens for PostgreSQL
-- `pgrag-src` - Retrieval Augmented Generation for PostgreSQL
-- `pgtap-src` - Unit testing framework for PostgreSQL
-- `pgvector-src` - Vector similarity search
-- `pgx_ulid-src` - ULID data type
-- `plv8-src` - JavaScript language for PostgreSQL stored procedures
-- `postgresql-unit-src` - SI units for PostgreSQL
-- `prefix-src` - Prefix matching for strings
-- `rag_bge_small_en_v15-src` - BGE embedding model for RAG
-- `rag_jina_reranker_v1_tiny_en-src` - Jina reranker model for RAG
-- `rum-src` - RUM access method for text search
-
-## Usage
-
-### Extension Upgrade Testing
-
-The extensions in this directory are used by the `test-upgrade.sh` script to test upgrading extensions between different versions of Neon Compute nodes. The script:
-
-1. Creates a database with extensions installed on an old Compute version
-2. Creates timelines for each extension
-3. Switches to a new Compute version and tests the upgrade process
-4. Verifies extension functionality after upgrade
-
-### Regular User Testing
-
-For testing with regular users (particularly for cloud instances), each extension directory typically contains a `regular-test.sh` script that:
-
-1. Drops the database if it exists
-2. Creates a fresh test database
-3. Installs the extension
-4. Runs regression tests
-
-A note about pg_regress: Since pg_regress attempts to set `lc_messages` for the database by default, which is forbidden for regular users, we create databases manually and use the `--use-existing` option to bypass this limitation.
-
-### CI Workflows
-
-Two main workflows use these extensions:
-
-1. **Cloud Extensions Test** - Tests extensions on Neon cloud projects
-2. **Force Test Upgrading of Extension** - Tests upgrading extensions between different Compute versions
-
-These workflows are integrated into the build-and-test pipeline through shell scripts:
-
-- `docker_compose_test.sh` - Tests extensions in a Docker Compose environment
-       
-- `test_extensions_upgrade.sh` - Tests extension upgrades between different Compute versions
-
-## Adding New Extensions
-
-To add a new extension for testing:
-
-1. Create a directory named `extension-name-src` in this directory
-2. Add at minimum:
-   - `regular-test.sh` for testing with regular users
-   - If `regular-test.sh` doesn't exist, the system will look for `neon-test.sh`
-   - If neither exists, it will try to run `make installcheck`
-   - `test-upgrade.sh` is only needed if you want to test upgrade scenarios
-3. Update the list of extensions in the `test_extensions_upgrade.sh` script if needed for upgrade testing
-
-### Patching Extension Sources
-
-If you need to patch the extension sources:
-
-1. Place the patch file in the extension's directory
-2. Apply the patch in the appropriate script (`test-upgrade.sh`, `neon-test.sh`, `regular-test.sh`, or `Makefile`)
-3. The patch will be applied during the testing process

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

@@ -1,7 +0,0 @@
-#!/bin/sh
-set -ex
-cd "$(dirname ${0})"
-PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
-dropdb --if-exists contrib_regression
-createdb contrib_regression
-${PG_REGRESS} --use-existing --inputdir=./ --bindir='/usr/local/pgsql/bin'    --dbname=contrib_regression setup add_agg agg_oob auto_sparse card_op cast_shape copy_binary cumulative_add_cardinality_correction cumulative_add_comprehensive_promotion cumulative_add_sparse_edge cumulative_add_sparse_random cumulative_add_sparse_step cumulative_union_comprehensive cumulative_union_explicit_explicit cumulative_union_explicit_promotion cumulative_union_probabilistic_probabilistic cumulative_union_sparse_full_representation cumulative_union_sparse_promotion cumulative_union_sparse_sparse disable_hashagg equal explicit_thresh hash hash_any meta_func murmur_bigint murmur_bytea nosparse notequal scalar_oob storedproc transaction typmod typmod_insert union_op

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

@@ -1,7 +0,0 @@
-#!/bin/sh
-set -ex
-cd "$(dirname ${0})"
-dropdb --if-exists contrib_regression
-createdb contrib_regression
-PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
-${PG_REGRESS} --inputdir=./ --bindir='/usr/local/pgsql/bin' --use-existing --inputdir=test --dbname=contrib_regression hypopg hypo_brin hypo_index_part hypo_include hypo_hash hypo_hide_index

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

@@ -1,7 +0,0 @@
-#!/bin/sh
-set -ex
-cd "$(dirname ${0})"
-dropdb --if-exist contrib_regression
-createdb contrib_regression
-PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
-${PG_REGRESS} --use-existing --inputdir=./ --bindir='/usr/local/pgsql/bin'    --dbname=contrib_regression ip4r ip4r-softerr ip4r-v11

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

@@ -1,7 +0,0 @@
-#!/bin/sh
-set -ex
-cd "$(dirname ${0})"
-dropdb --if-exist contrib_regression
-createdb contrib_regression
-PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
-${PG_REGRESS} --use-existing --inputdir=./ --bindir='/usr/local/pgsql/bin'    --dbname=contrib_regression pg_cron-test

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

@@ -1,23 +0,0 @@
-#!/bin/bash
-set -ex
-cd "$(dirname "${0}")"
-PGXS="$(dirname "$(pg_config --pgxs)" )"
-REGRESS="${PGXS}/../test/regress/pg_regress"
-TESTDIR="test"
-TESTS=$(ls "${TESTDIR}/sql" | sort )
-TESTS=${TESTS//\.sql/}
-TESTS=${TESTS/empty_mutations/}
-TESTS=${TESTS/function_return_row_is_selectable/}
-TESTS=${TESTS/issue_300/}
-TESTS=${TESTS/permissions_connection_column/}
-TESTS=${TESTS/permissions_functions/}
-TESTS=${TESTS/permissions_node_column/}
-TESTS=${TESTS/permissions_table_level/}
-TESTS=${TESTS/permissions_types/}
-TESTS=${TESTS/row_level_security/}
-TESTS=${TESTS/sqli_connection/}
-dropdb --if-exist contrib_regression
-createdb contrib_regression
-psql -v ON_ERROR_STOP=1 -f test/fixtures.sql -d contrib_regression
-${REGRESS} --use-existing --dbname=contrib_regression --inputdir=${TESTDIR} ${TESTS}
-

docker-compose/ext-src/pg_hint_plan-src/regular-test.sh

@@ -1,7 +0,0 @@
-#!/bin/sh
-set -ex
-cd "$(dirname ${0})"
-dropdb --if-exist contrib_regression
-createdb contrib_regression
-PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
-${PG_REGRESS} --use-existing  --inputdir=./ --bindir='/usr/local/pgsql/bin'    --encoding=UTF8 --dbname=contrib_regression init base_plan pg_hint_plan ut-init ut-A ut-S ut-J ut-L ut-G ut-R ut-fdw ut-W ut-T ut-fini hints_anywhere plpgsql oldextversions

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

@@ -1,9 +0,0 @@
-#!/bin/sh
-set -ex
-dropdb --if-exist contrib_regression
-createdb contrib_regression
-cd "$(dirname ${0})"
-PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
-patch -p1 <regular.patch
-${PG_REGRESS} --use-existing --inputdir=./ --bindir='/usr/local/pgsql/bin' --dbname=contrib_regression pg_ivm create_immv refresh_immv
-patch -R -p1 <regular.patch

docker-compose/ext-src/pg_ivm-src/regular.patch

@@ -1,309 +0,0 @@
-diff --git a/expected/pg_ivm.out b/expected/pg_ivm.out
-index e8798ee..4081680 100644
---- a/expected/pg_ivm.out
-+++ b/expected/pg_ivm.out
-@@ -1363,61 +1363,6 @@ SELECT * FROM mv ORDER BY i;
-    |   2 |   4 |                 2 |                 2 |             2
- (1 row)
- 
--ROLLBACK;
---- IMMV containing user defined type
--BEGIN;
--CREATE TYPE mytype;
--CREATE FUNCTION mytype_in(cstring)
-- RETURNS mytype AS 'int4in'
-- LANGUAGE INTERNAL STRICT IMMUTABLE;
--NOTICE:  return type mytype is only a shell
--CREATE FUNCTION mytype_out(mytype)
-- RETURNS cstring AS 'int4out'
-- LANGUAGE INTERNAL STRICT IMMUTABLE;
--NOTICE:  argument type mytype is only a shell
--CREATE TYPE mytype (
-- LIKE = int4,
-- INPUT = mytype_in,
-- OUTPUT = mytype_out
--);
--CREATE FUNCTION mytype_eq(mytype, mytype)
-- RETURNS bool AS 'int4eq'
-- LANGUAGE INTERNAL STRICT IMMUTABLE;
--CREATE FUNCTION mytype_lt(mytype, mytype)
-- RETURNS bool AS 'int4lt'
-- LANGUAGE INTERNAL STRICT IMMUTABLE;
--CREATE FUNCTION mytype_cmp(mytype, mytype)
-- RETURNS integer AS 'btint4cmp'
-- LANGUAGE INTERNAL STRICT IMMUTABLE;
--CREATE OPERATOR = (
-- leftarg = mytype, rightarg = mytype,
-- procedure = mytype_eq);
--CREATE OPERATOR < (
-- leftarg = mytype, rightarg = mytype,
-- procedure = mytype_lt);
--CREATE OPERATOR CLASS mytype_ops
-- DEFAULT FOR TYPE mytype USING btree AS
-- OPERATOR        1       <,
-- OPERATOR        3       = ,
-- FUNCTION		1		mytype_cmp(mytype,mytype);
--CREATE TABLE t_mytype (x mytype);
--SELECT create_immv('mv_mytype',
-- 'SELECT * FROM t_mytype');
--NOTICE:  could not create an index on immv "mv_mytype" automatically
--DETAIL:  This target list does not have all the primary key columns, or this view does not contain GROUP BY or DISTINCT clause.
--HINT:  Create an index on the immv for efficient incremental maintenance.
-- create_immv 
---------------
--           0
--(1 row)
--
--INSERT INTO t_mytype VALUES ('1'::mytype);
--SELECT * FROM mv_mytype;
-- x 
-----
-- 1
--(1 row)
--
- ROLLBACK;
- -- outer join is not supported
- SELECT create_immv('mv(a,b)',
-@@ -1510,112 +1455,6 @@ SELECT create_immv('mv_ivm_only_values1', 'values(1)');
- ERROR:  VALUES is not supported on incrementally maintainable materialized view
- SELECT create_immv('mv_ivm_only_values2',  'SELECT * FROM (values(1)) AS tmp');
- ERROR:  VALUES is not supported on incrementally maintainable materialized view
---- views containing base tables with Row Level Security
--DROP USER IF EXISTS ivm_admin;
--NOTICE:  role "ivm_admin" does not exist, skipping
--DROP USER IF EXISTS ivm_user;
--NOTICE:  role "ivm_user" does not exist, skipping
--CREATE USER ivm_admin;
--CREATE USER ivm_user;
----- create a table with RLS
--SET SESSION AUTHORIZATION ivm_admin;
--CREATE TABLE rls_tbl(id int, data text, owner name);
--INSERT INTO rls_tbl VALUES
--  (1,'foo','ivm_user'),
--  (2,'bar','postgres');
--CREATE TABLE num_tbl(id int, num text);
--INSERT INTO num_tbl VALUES
--  (1,'one'),
--  (2,'two'),
--  (3,'three'),
--  (4,'four'),
--  (5,'five'),
--  (6,'six');
----- Users can access only their own rows
--CREATE POLICY rls_tbl_policy ON rls_tbl FOR SELECT TO PUBLIC USING(owner = current_user);
--ALTER TABLE rls_tbl ENABLE ROW LEVEL SECURITY;
--GRANT ALL on rls_tbl TO PUBLIC;
--GRANT ALL on num_tbl TO PUBLIC;
----- create a view owned by ivm_user
--SET SESSION AUTHORIZATION ivm_user;
--SELECT create_immv('ivm_rls', 'SELECT * FROM rls_tbl');
--NOTICE:  could not create an index on immv "ivm_rls" automatically
--DETAIL:  This target list does not have all the primary key columns, or this view does not contain GROUP BY or DISTINCT clause.
--HINT:  Create an index on the immv for efficient incremental maintenance.
-- create_immv 
---------------
--           1
--(1 row)
--
--SELECT id, data, owner FROM ivm_rls ORDER BY 1,2,3;
-- id | data |  owner   
------+------+----------
--  1 | foo  | ivm_user
--(1 row)
--
--RESET SESSION AUTHORIZATION;
----- inserts rows owned by different users
--INSERT INTO rls_tbl VALUES
--  (3,'baz','ivm_user'),
--  (4,'qux','postgres');
--SELECT id, data, owner FROM ivm_rls ORDER BY 1,2,3;
-- id | data |  owner   
------+------+----------
--  1 | foo  | ivm_user
--  3 | baz  | ivm_user
--(2 rows)
--
----- combination of diffent kinds of commands
--WITH
-- i AS (INSERT INTO rls_tbl VALUES(5,'quux','postgres'), (6,'corge','ivm_user')),
-- u AS (UPDATE rls_tbl SET owner = 'postgres' WHERE id = 1),
-- u2 AS (UPDATE rls_tbl SET owner = 'ivm_user' WHERE id = 2)
--SELECT;
----
--(1 row)
--
--SELECT id, data, owner FROM ivm_rls ORDER BY 1,2,3;
-- id | data  |  owner   
------+-------+----------
--  2 | bar   | ivm_user
--  3 | baz   | ivm_user
--  6 | corge | ivm_user
--(3 rows)
--
-----
--SET SESSION AUTHORIZATION ivm_user;
--SELECT create_immv('ivm_rls2', 'SELECT * FROM rls_tbl JOIN num_tbl USING(id)');
--NOTICE:  could not create an index on immv "ivm_rls2" automatically
--DETAIL:  This target list does not have all the primary key columns, or this view does not contain GROUP BY or DISTINCT clause.
--HINT:  Create an index on the immv for efficient incremental maintenance.
-- create_immv 
---------------
--           3
--(1 row)
--
--RESET SESSION AUTHORIZATION;
--WITH
-- x AS (UPDATE rls_tbl SET data = data || '_2' where id in (3,4)),
-- y AS (UPDATE num_tbl SET num = num || '_2' where id in (3,4))
--SELECT;
----
--(1 row)
--
--SELECT * FROM ivm_rls2 ORDER BY 1,2,3;
-- id | data  |  owner   |   num   
------+-------+----------+---------
--  2 | bar   | ivm_user | two
--  3 | baz_2 | ivm_user | three_2
--  6 | corge | ivm_user | six
--(3 rows)
--
--DROP TABLE rls_tbl CASCADE;
--NOTICE:  drop cascades to 2 other objects
--DETAIL:  drop cascades to table ivm_rls
--drop cascades to table ivm_rls2
--DROP TABLE num_tbl CASCADE;
--DROP USER ivm_user;
--DROP USER ivm_admin;
- -- automatic index creation
- BEGIN;
- CREATE TABLE base_a (i int primary key, j int);
-diff --git a/sql/pg_ivm.sql b/sql/pg_ivm.sql
-index d3c1a01..203213d 100644
---- a/sql/pg_ivm.sql
-+++ b/sql/pg_ivm.sql
-@@ -454,53 +454,6 @@ DELETE FROM base_t WHERE v = 5;
- SELECT * FROM mv ORDER BY i;
- ROLLBACK;
- 
---- IMMV containing user defined type
--BEGIN;
--
--CREATE TYPE mytype;
--CREATE FUNCTION mytype_in(cstring)
-- RETURNS mytype AS 'int4in'
-- LANGUAGE INTERNAL STRICT IMMUTABLE;
--CREATE FUNCTION mytype_out(mytype)
-- RETURNS cstring AS 'int4out'
-- LANGUAGE INTERNAL STRICT IMMUTABLE;
--CREATE TYPE mytype (
-- LIKE = int4,
-- INPUT = mytype_in,
-- OUTPUT = mytype_out
--);
--
--CREATE FUNCTION mytype_eq(mytype, mytype)
-- RETURNS bool AS 'int4eq'
-- LANGUAGE INTERNAL STRICT IMMUTABLE;
--CREATE FUNCTION mytype_lt(mytype, mytype)
-- RETURNS bool AS 'int4lt'
-- LANGUAGE INTERNAL STRICT IMMUTABLE;
--CREATE FUNCTION mytype_cmp(mytype, mytype)
-- RETURNS integer AS 'btint4cmp'
-- LANGUAGE INTERNAL STRICT IMMUTABLE;
--
--CREATE OPERATOR = (
-- leftarg = mytype, rightarg = mytype,
-- procedure = mytype_eq);
--CREATE OPERATOR < (
-- leftarg = mytype, rightarg = mytype,
-- procedure = mytype_lt);
--
--CREATE OPERATOR CLASS mytype_ops
-- DEFAULT FOR TYPE mytype USING btree AS
-- OPERATOR        1       <,
-- OPERATOR        3       = ,
-- FUNCTION		1		mytype_cmp(mytype,mytype);
--
--CREATE TABLE t_mytype (x mytype);
--SELECT create_immv('mv_mytype',
-- 'SELECT * FROM t_mytype');
--INSERT INTO t_mytype VALUES ('1'::mytype);
--SELECT * FROM mv_mytype;
--
--ROLLBACK;
--
- -- outer join is not supported
- SELECT create_immv('mv(a,b)',
-     'SELECT a.i, b.i FROM mv_base_a a LEFT JOIN mv_base_b b ON a.i=b.i');
-@@ -579,71 +532,6 @@ SELECT create_immv('mv_ivm31', 'SELECT sum(i)/sum(j) FROM mv_base_a');
- SELECT create_immv('mv_ivm_only_values1', 'values(1)');
- SELECT create_immv('mv_ivm_only_values2',  'SELECT * FROM (values(1)) AS tmp');
- 
--
---- views containing base tables with Row Level Security
--DROP USER IF EXISTS ivm_admin;
--DROP USER IF EXISTS ivm_user;
--CREATE USER ivm_admin;
--CREATE USER ivm_user;
--
----- create a table with RLS
--SET SESSION AUTHORIZATION ivm_admin;
--CREATE TABLE rls_tbl(id int, data text, owner name);
--INSERT INTO rls_tbl VALUES
--  (1,'foo','ivm_user'),
--  (2,'bar','postgres');
--CREATE TABLE num_tbl(id int, num text);
--INSERT INTO num_tbl VALUES
--  (1,'one'),
--  (2,'two'),
--  (3,'three'),
--  (4,'four'),
--  (5,'five'),
--  (6,'six');
--
----- Users can access only their own rows
--CREATE POLICY rls_tbl_policy ON rls_tbl FOR SELECT TO PUBLIC USING(owner = current_user);
--ALTER TABLE rls_tbl ENABLE ROW LEVEL SECURITY;
--GRANT ALL on rls_tbl TO PUBLIC;
--GRANT ALL on num_tbl TO PUBLIC;
--
----- create a view owned by ivm_user
--SET SESSION AUTHORIZATION ivm_user;
--SELECT create_immv('ivm_rls', 'SELECT * FROM rls_tbl');
--SELECT id, data, owner FROM ivm_rls ORDER BY 1,2,3;
--RESET SESSION AUTHORIZATION;
--
----- inserts rows owned by different users
--INSERT INTO rls_tbl VALUES
--  (3,'baz','ivm_user'),
--  (4,'qux','postgres');
--SELECT id, data, owner FROM ivm_rls ORDER BY 1,2,3;
--
----- combination of diffent kinds of commands
--WITH
-- i AS (INSERT INTO rls_tbl VALUES(5,'quux','postgres'), (6,'corge','ivm_user')),
-- u AS (UPDATE rls_tbl SET owner = 'postgres' WHERE id = 1),
-- u2 AS (UPDATE rls_tbl SET owner = 'ivm_user' WHERE id = 2)
--SELECT;
--SELECT id, data, owner FROM ivm_rls ORDER BY 1,2,3;
--
-----
--SET SESSION AUTHORIZATION ivm_user;
--SELECT create_immv('ivm_rls2', 'SELECT * FROM rls_tbl JOIN num_tbl USING(id)');
--RESET SESSION AUTHORIZATION;
--
--WITH
-- x AS (UPDATE rls_tbl SET data = data || '_2' where id in (3,4)),
-- y AS (UPDATE num_tbl SET num = num || '_2' where id in (3,4))
--SELECT;
--SELECT * FROM ivm_rls2 ORDER BY 1,2,3;
--
--DROP TABLE rls_tbl CASCADE;
--DROP TABLE num_tbl CASCADE;
--
--DROP USER ivm_user;
--DROP USER ivm_admin;
--
- -- automatic index creation
- BEGIN;
- CREATE TABLE base_a (i int primary key, j int);

docker-compose/ext-src/pg_jsonschema-src/Makefile

@@ -1,13 +1,8 @@
 EXTENSION = pg_jsonschema
 DATA = pg_jsonschema--1.0.sql
 REGRESS = jsonschema_valid_api  jsonschema_edge_cases
+REGRESS_OPTS = --load-extension=pg_jsonschema
 
 PG_CONFIG ?= pg_config
 PGXS := $(shell $(PG_CONFIG) --pgxs)
-PG_REGRESS := $(dir $(PGXS))../../src/test/regress/pg_regress
-.PHONY installcheck:
-installcheck:
-	dropdb --if-exists contrib_regression
-	createdb contrib_regression
-	psql -d contrib_regression -c "CREATE EXTENSION $(EXTENSION)"
-	$(PG_REGRESS) --use-existing --dbname=contrib_regression $(REGRESS)
+include $(PGXS)

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

@@ -1,7 +0,0 @@
-#!/bin/sh
-set -ex
-cd "$(dirname ${0})"
-dropdb --if-exist contrib_regression
-createdb contrib_regression
-PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
-${PG_REGRESS} --use-existing --inputdir=./ --bindir='/usr/local/pgsql/bin'    --dbname=contrib_regression roaringbitmap

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

@@ -1,12 +0,0 @@
-#!/bin/bash
-set -ex
-# For v16 it's required to create a type which is impossible without superuser access
-# do not run this test so far
-if [[ "${PG_VERSION}" = v16 ]]; then
-  exit 0
-fi
-cd "$(dirname ${0})"
-dropdb --if-exist contrib_regression
-createdb contrib_regression
-PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
-${PG_REGRESS} --use-existing --inputdir=./ --bindir='/usr/local/pgsql/bin'    --inputdir=test --dbname=contrib_regression base corpus

docker-compose/ext-src/pg_session_jwt-src/Makefile

@@ -6,10 +6,4 @@ export PGOPTIONS = -c pg_session_jwt.jwk={"crv":"Ed25519","kty":"OKP","x":"R_Abz
 
 PG_CONFIG ?= pg_config
 PGXS := $(shell $(PG_CONFIG) --pgxs)
-PG_REGRESS := $(dir $(PGXS))../../src/test/regress/pg_regress
-.PHONY installcheck:
-installcheck:
-	dropdb --if-exists contrib_regression
-	createdb contrib_regression
-	psql -d contrib_regression -c "CREATE EXTENSION $(EXTENSION)"
-	$(PG_REGRESS) --use-existing --dbname=contrib_regression $(REGRESS)
+include $(PGXS)

docker-compose/ext-src/pg_tiktoken-src/Makefile

@@ -5,6 +5,4 @@ REGRESS = pg_tiktoken
 installcheck: regression-test
 
 regression-test:
-	dropdb --if-exists contrib_regression
-	createdb contrib_regression
-	$(PG_REGRESS) --inputdir=. --outputdir=. --use-existing --dbname=contrib_regression $(REGRESS)
+	$(PG_REGRESS) --inputdir=. --outputdir=. --dbname=contrib_regression $(REGRESS)

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

@@ -1,7 +0,0 @@
-#!/bin/sh
-set -ex
-cd "$(dirname "${0}")"
-dropdb --if-exist contrib_regression
-createdb contrib_regression
-PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
-${PG_REGRESS} --use-existing --inputdir=./ --bindir='/usr/local/pgsql/bin'    --inputdir=test --dbname=contrib_regression 001_setup 002_uuid_generate_v7 003_uuid_v7_to_timestamptz 004_uuid_timestamptz_to_v7 005_uuid_v7_to_timestamp 006_uuid_timestamp_to_v7

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

@@ -1,6 +1,4 @@
 #!/bin/bash
 set -ex
 cd "$(dirname "${0}")"
-dropdb --if-exists contrib_regression
-createdb contrib_regression
-pg_prove -d contrib_regression test.sql
+pg_prove test.sql

docker-compose/ext-src/pgrag-src/regular-test.sh

@@ -1,8 +0,0 @@
-#!/bin/sh
-set -ex
-cd "$(dirname "${0}")"
-dropdb --if-exist contrib_regression
-createdb contrib_regression
-psql -d contrib_regression -c "CREATE EXTENSION vector" -c "CREATE EXTENSION rag"
-PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
-${PG_REGRESS} --inputdir=./ --bindir='/usr/local/pgsql/bin'    --use-existing --load-extension=vector --load-extension=rag --dbname=contrib_regression basic_functions text_processing api_keys chunking_functions document_processing embedding_api_functions voyageai_functions

docker-compose/ext-src/pgtap-src/regular-test.sh

@@ -1,10 +0,0 @@
-#!/bin/sh
-set -ex
-cd "$(dirname ${0})"
-make installcheck || true
-dropdb --if-exist contrib_regression
-createdb contrib_regression
-PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
-sed -i '/hastap/d' test/build/run.sch
-sed -Ei 's/\b(aretap|enumtap|ownership|privs|usergroup)\b//g' test/build/run.sch
-${PG_REGRESS} --use-existing --dbname=contrib_regression --inputdir=./ --bindir='/usr/local/pgsql/bin'    --inputdir=test --max-connections=879 --schedule test/schedule/main.sch   --schedule test/build/run.sch

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

@@ -1,8 +0,0 @@
-#!/bin/sh
-set -ex
-cd "$(dirname ${0})"
-dropdb --if-exist contrib_regression
-createdb contrib_regression
-psql -d contrib_regression -c "CREATE EXTENSION vector"
-PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
-${PG_REGRESS} --inputdir=./ --bindir='/usr/local/pgsql/bin' --inputdir=test --use-existing --dbname=contrib_regression bit btree cast copy halfvec hnsw_bit hnsw_halfvec hnsw_sparsevec hnsw_vector ivfflat_bit ivfflat_halfvec ivfflat_vector sparsevec vector_type

docker-compose/ext-src/pgx_ulid-src/Makefile

@@ -4,21 +4,13 @@ PGFILEDESC = "pgx_ulid - ULID type for PostgreSQL"
 
 PG_CONFIG ?= pg_config
 PGXS := $(shell $(PG_CONFIG) --pgxs)
-PG_REGRESS = $(dir $(PGXS))/../../src/test/regress/pg_regress
 PG_MAJOR_VERSION := $(word 2, $(subst ., , $(shell $(PG_CONFIG) --version)))
 ifeq ($(shell test $(PG_MAJOR_VERSION) -lt 17; echo $$?),0)
+  REGRESS_OPTS = --load-extension=ulid
   REGRESS = 00_ulid_generation 01_ulid_conversions 03_ulid_errors
-  EXTNAME = ulid
 else
+  REGRESS_OPTS = --load-extension=pgx_ulid
   REGRESS = 00_ulid_generation 01_ulid_conversions 02_ulid_conversions 03_ulid_errors
-  EXTNAME = pgx_ulid
 endif
 
-.PHONY: installcheck
-installcheck: regression-test
-
-regression-test:
-	dropdb --if-exists contrib_regression
-	createdb contrib_regression
-	psql -d contrib_regression -c "CREATE EXTENSION $(EXTNAME)"
-	$(PG_REGRESS) --inputdir=. --outputdir=. --use-existing --dbname=contrib_regression $(REGRESS)
+include $(PGXS)

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

@@ -1,12 +0,0 @@
-#!/bin/bash
-set -ex
-cd "$(dirname ${0})"
-dropdb --if-exist contrib_regression
-createdb contrib_regression
-PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
-REGRESS="$(make -n installcheck | awk '{print substr($0,index($0,"init-extension"));}')"
-REGRESS="${REGRESS/startup_perms/}"
-REGRESS="${REGRESS/startup /}"
-REGRESS="${REGRESS/find_function_perms/}"
-REGRESS="${REGRESS/guc/}"
-${PG_REGRESS} --inputdir=./ --bindir='/usr/local/pgsql/bin'  --use-existing --dbname=contrib_regression ${REGRESS}

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

@@ -1,7 +0,0 @@
-#!/bin/sh
-set -ex
-cd "$(dirname ${0})"
-dropdb --if-exist contrib_regression
-createdb contrib_regression
-PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
-${PG_REGRESS} --inputdir=./ --bindir='/usr/local/pgsql/bin' --use-existing --dbname=contrib_regression extension tables unit binary unicode prefix units time temperature functions language_functions round derived compare aggregate iec custom crosstab convert

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

@@ -1,7 +0,0 @@
-#!/bin/sh
-set -ex
-cd "$(dirname ${0})"
-dropdb --if-exist contrib_regression
-createdb contrib_regression
-PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
-${PG_REGRESS} --use-existing --inputdir=./ --bindir='/usr/local/pgsql/bin'    --dbname=contrib_regression create_extension prefix falcon explain queries

docker-compose/ext-src/rag_bge_small_en_v15-src/Makefile

@@ -3,13 +3,8 @@ MODULE_big = rag_bge_small_en_v15
 OBJS = $(patsubst %.rs,%.o,$(wildcard src/*.rs))
 
 REGRESS = basic_functions embedding_functions basic_functions_enhanced embedding_functions_enhanced
+REGRESS_OPTS = --load-extension=vector --load-extension=rag_bge_small_en_v15
 
 PG_CONFIG = pg_config
 PGXS := $(shell $(PG_CONFIG) --pgxs)
-PG_REGRESS := $(dir $(PGXS))../../src/test/regress/pg_regress
-.PHONY installcheck:
-installcheck:
-	dropdb --if-exists contrib_regression
-	createdb contrib_regression
-	psql -d contrib_regression -c "CREATE EXTENSION vector" -c "CREATE EXTENSION rag_bge_small_en_v15"
-	$(PG_REGRESS) --use-existing --dbname=contrib_regression $(REGRESS)
+include $(PGXS)

docker-compose/ext-src/rag_jina_reranker_v1_tiny_en-src/Makefile

@@ -3,13 +3,8 @@ MODULE_big = rag_jina_reranker_v1_tiny_en
 OBJS = $(patsubst %.rs,%.o,$(wildcard src/*.rs))
 
 REGRESS = reranking_functions reranking_functions_enhanced
+REGRESS_OPTS = --load-extension=vector --load-extension=rag_jina_reranker_v1_tiny_en
 
 PG_CONFIG = pg_config
 PGXS := $(shell $(PG_CONFIG) --pgxs)
-PG_REGRESS := $(dir $(PGXS))../../src/test/regress/pg_regress
-.PHONY installcheck:
-installcheck:
-	dropdb --if-exists contrib_regression
-	createdb contrib_regression
-	psql -d contrib_regression -c "CREATE EXTENSION vector" -c "CREATE EXTENSION rag_jina_reranker_v1_tiny_en"
-	$(PG_REGRESS) --use-existing --dbname=contrib_regression $(REGRESS)
+include $(PGXS)

docker-compose/ext-src/rag_jina_reranker_v1_tiny_en-src/expected/reranking_functions.out

@@ -1,27 +1,25 @@
 -- Reranking function tests
-SELECT ROUND(rag_jina_reranker_v1_tiny_en.rerank_distance('the cat sat on the mat', 'the baboon played with the balloon')::NUMERIC,4);
- round  
---------
- 0.8989
-(1 row)
-
-SELECT ARRAY(SELECT ROUND(x::NUMERIC,4) FROM unnest(rag_jina_reranker_v1_tiny_en.rerank_distance('the cat sat on the mat',
-    ARRAY['the baboon played with the balloon', 'the tanks fired at the buildings'])) AS x);
-      array      
+SELECT rag_jina_reranker_v1_tiny_en.rerank_distance('the cat sat on the mat', 'the baboon played with the balloon');
+ rerank_distance 
 -----------------
- {0.8989,1.3018}
+       0.8989152
 (1 row)
 
-SELECT ROUND(rag_jina_reranker_v1_tiny_en.rerank_score('the cat sat on the mat', 'the baboon played with the balloon')::NUMERIC,4);
-  round  
----------
- -0.8989
+SELECT rag_jina_reranker_v1_tiny_en.rerank_distance('the cat sat on the mat', ARRAY['the baboon played with the balloon', 'the tanks fired at the buildings']);
+    rerank_distance    
+-----------------------
+ {0.8989152,1.3018152}
 (1 row)
 
-SELECT ARRAY(SELECT ROUND(x::NUMERIC,4) FROM unnest(rag_jina_reranker_v1_tiny_en.rerank_score('the cat sat on the mat',
-    ARRAY['the baboon played with the balloon', 'the tanks fired at the buildings'])) as x);
-       array       
--------------------
- {-0.8989,-1.3018}
+SELECT rag_jina_reranker_v1_tiny_en.rerank_score('the cat sat on the mat', 'the baboon played with the balloon');
+ rerank_score 
+--------------
+   -0.8989152
+(1 row)
+
+SELECT rag_jina_reranker_v1_tiny_en.rerank_score('the cat sat on the mat', ARRAY['the baboon played with the balloon', 'the tanks fired at the buildings']);
+      rerank_score       
+-------------------------
+ {-0.8989152,-1.3018152}
 (1 row)
 

docker-compose/ext-src/rag_jina_reranker_v1_tiny_en-src/expected/reranking_functions_enhanced.out

@@ -1,41 +1,41 @@
 -- Reranking function tests - single passage
-SELECT ROUND(rag_jina_reranker_v1_tiny_en.rerank_distance('the cat sat on the mat', 'the baboon played with the balloon')::NUMERIC,4);
- round  
---------
- 0.8989
+SELECT rag_jina_reranker_v1_tiny_en.rerank_distance('the cat sat on the mat', 'the baboon played with the balloon');
+ rerank_distance 
+-----------------
+       0.8989152
 (1 row)
 
-SELECT ROUND(rag_jina_reranker_v1_tiny_en.rerank_distance('the cat sat on the mat', 'the tanks fired at the buildings')::NUMERIC,4);
- round  
---------
- 1.3018
+SELECT rag_jina_reranker_v1_tiny_en.rerank_distance('the cat sat on the mat', 'the tanks fired at the buildings');
+ rerank_distance 
+-----------------
+       1.3018152
 (1 row)
 
-SELECT ROUND(rag_jina_reranker_v1_tiny_en.rerank_distance('query about cats', 'information about felines')::NUMERIC,4);
- round  
---------
- 1.3133
+SELECT rag_jina_reranker_v1_tiny_en.rerank_distance('query about cats', 'information about felines');
+ rerank_distance 
+-----------------
+       1.3133051
 (1 row)
 
-SELECT ROUND(rag_jina_reranker_v1_tiny_en.rerank_distance('', 'empty query test')::NUMERIC,4);
- round  
---------
- 0.7076
+SELECT rag_jina_reranker_v1_tiny_en.rerank_distance('', 'empty query test');
+ rerank_distance 
+-----------------
+       0.7075559
 (1 row)
 
 -- Reranking function tests - array of passages
-SELECT ARRAY(SELECT ROUND(x::NUMERIC,4) FROM unnest(rag_jina_reranker_v1_tiny_en.rerank_distance('the cat sat on the mat',
-    ARRAY['the baboon played with the balloon', 'the tanks fired at the buildings'])) AS x);
-      array      
------------------
- {0.8989,1.3018}
+SELECT rag_jina_reranker_v1_tiny_en.rerank_distance('the cat sat on the mat',
+    ARRAY['the baboon played with the balloon', 'the tanks fired at the buildings']);
+    rerank_distance    
+-----------------------
+ {0.8989152,1.3018152}
 (1 row)
 
-SELECT ARRAY(SELECT ROUND(x::NUMERIC,4) FROM unnest(rag_jina_reranker_v1_tiny_en.rerank_distance('query about programming',
-    ARRAY['Python is a programming language', 'Java is also a programming language', 'SQL is used for databases'])) AS x);
-         array          
-------------------------
- {0.1659,0.3348,0.1013}
+SELECT rag_jina_reranker_v1_tiny_en.rerank_distance('query about programming',
+    ARRAY['Python is a programming language', 'Java is also a programming language', 'SQL is used for databases']);
+          rerank_distance           
+------------------------------------
+ {0.16591403,0.33475375,0.10132827}
 (1 row)
 
 SELECT rag_jina_reranker_v1_tiny_en.rerank_distance('empty array test', ARRAY[]::text[]);
@@ -45,43 +45,43 @@ SELECT rag_jina_reranker_v1_tiny_en.rerank_distance('empty array test', ARRAY[]:
 (1 row)
 
 -- Reranking score function tests - single passage
-SELECT ROUND(rag_jina_reranker_v1_tiny_en.rerank_score('the cat sat on the mat', 'the baboon played with the balloon')::NUMERIC,4);
-  round  
----------
- -0.8989
+SELECT rag_jina_reranker_v1_tiny_en.rerank_score('the cat sat on the mat', 'the baboon played with the balloon');
+ rerank_score 
+--------------
+   -0.8989152
 (1 row)
 
-SELECT ROUND(rag_jina_reranker_v1_tiny_en.rerank_score('the cat sat on the mat', 'the tanks fired at the buildings')::NUMERIC,4);
-  round  
----------
- -1.3018
+SELECT rag_jina_reranker_v1_tiny_en.rerank_score('the cat sat on the mat', 'the tanks fired at the buildings');
+ rerank_score 
+--------------
+   -1.3018152
 (1 row)
 
-SELECT ROUND(rag_jina_reranker_v1_tiny_en.rerank_score('query about cats', 'information about felines')::NUMERIC,4);
-  round  
----------
- -1.3133
+SELECT rag_jina_reranker_v1_tiny_en.rerank_score('query about cats', 'information about felines');
+ rerank_score 
+--------------
+   -1.3133051
 (1 row)
 
-SELECT ROUND(rag_jina_reranker_v1_tiny_en.rerank_score('', 'empty query test')::NUMERIC,4);
-  round  
----------
- -0.7076
+SELECT rag_jina_reranker_v1_tiny_en.rerank_score('', 'empty query test');
+ rerank_score 
+--------------
+   -0.7075559
 (1 row)
 
 -- Reranking score function tests - array of passages
-SELECT ARRAY(SELECT ROUND(x::NUMERIC,4) FROM unnest(rag_jina_reranker_v1_tiny_en.rerank_score('the cat sat on the mat',
-    ARRAY['the baboon played with the balloon', 'the tanks fired at the buildings'])) AS x);
-       array       
--------------------
- {-0.8989,-1.3018}
+SELECT rag_jina_reranker_v1_tiny_en.rerank_score('the cat sat on the mat',
+    ARRAY['the baboon played with the balloon', 'the tanks fired at the buildings']);
+      rerank_score       
+-------------------------
+ {-0.8989152,-1.3018152}
 (1 row)
 
-SELECT ARRAY(SELECT ROUND(x::NUMERIC,4) FROM unnest(rag_jina_reranker_v1_tiny_en.rerank_score('query about programming',
-    ARRAY['Python is a programming language', 'Java is also a programming language', 'SQL is used for databases'])) AS x);
-           array           
----------------------------
- {-0.1659,-0.3348,-0.1013}
+SELECT rag_jina_reranker_v1_tiny_en.rerank_score('query about programming',
+    ARRAY['Python is a programming language', 'Java is also a programming language', 'SQL is used for databases']);
+             rerank_score              
+---------------------------------------
+ {-0.16591403,-0.33475375,-0.10132827}
 (1 row)
 
 SELECT rag_jina_reranker_v1_tiny_en.rerank_score('empty array test', ARRAY[]::text[]);

docker-compose/ext-src/rag_jina_reranker_v1_tiny_en-src/sql/reranking_functions.sql

@@ -1,10 +1,8 @@
 -- Reranking function tests
-SELECT ROUND(rag_jina_reranker_v1_tiny_en.rerank_distance('the cat sat on the mat', 'the baboon played with the balloon')::NUMERIC,4);
+SELECT rag_jina_reranker_v1_tiny_en.rerank_distance('the cat sat on the mat', 'the baboon played with the balloon');
 
-SELECT ARRAY(SELECT ROUND(x::NUMERIC,4) FROM unnest(rag_jina_reranker_v1_tiny_en.rerank_distance('the cat sat on the mat',
-    ARRAY['the baboon played with the balloon', 'the tanks fired at the buildings'])) AS x);
+SELECT rag_jina_reranker_v1_tiny_en.rerank_distance('the cat sat on the mat', ARRAY['the baboon played with the balloon', 'the tanks fired at the buildings']);
 
-SELECT ROUND(rag_jina_reranker_v1_tiny_en.rerank_score('the cat sat on the mat', 'the baboon played with the balloon')::NUMERIC,4);
+SELECT rag_jina_reranker_v1_tiny_en.rerank_score('the cat sat on the mat', 'the baboon played with the balloon');
 
-SELECT ARRAY(SELECT ROUND(x::NUMERIC,4) FROM unnest(rag_jina_reranker_v1_tiny_en.rerank_score('the cat sat on the mat',
-    ARRAY['the baboon played with the balloon', 'the tanks fired at the buildings'])) as x);
+SELECT rag_jina_reranker_v1_tiny_en.rerank_score('the cat sat on the mat', ARRAY['the baboon played with the balloon', 'the tanks fired at the buildings']);

docker-compose/ext-src/rag_jina_reranker_v1_tiny_en-src/sql/reranking_functions_enhanced.sql

@@ -1,35 +1,35 @@
 -- Reranking function tests - single passage
-SELECT ROUND(rag_jina_reranker_v1_tiny_en.rerank_distance('the cat sat on the mat', 'the baboon played with the balloon')::NUMERIC,4);
+SELECT rag_jina_reranker_v1_tiny_en.rerank_distance('the cat sat on the mat', 'the baboon played with the balloon');
 
-SELECT ROUND(rag_jina_reranker_v1_tiny_en.rerank_distance('the cat sat on the mat', 'the tanks fired at the buildings')::NUMERIC,4);
+SELECT rag_jina_reranker_v1_tiny_en.rerank_distance('the cat sat on the mat', 'the tanks fired at the buildings');
 
-SELECT ROUND(rag_jina_reranker_v1_tiny_en.rerank_distance('query about cats', 'information about felines')::NUMERIC,4);
+SELECT rag_jina_reranker_v1_tiny_en.rerank_distance('query about cats', 'information about felines');
 
-SELECT ROUND(rag_jina_reranker_v1_tiny_en.rerank_distance('', 'empty query test')::NUMERIC,4);
+SELECT rag_jina_reranker_v1_tiny_en.rerank_distance('', 'empty query test');
 
 -- Reranking function tests - array of passages
-SELECT ARRAY(SELECT ROUND(x::NUMERIC,4) FROM unnest(rag_jina_reranker_v1_tiny_en.rerank_distance('the cat sat on the mat',
-    ARRAY['the baboon played with the balloon', 'the tanks fired at the buildings'])) AS x);
+SELECT rag_jina_reranker_v1_tiny_en.rerank_distance('the cat sat on the mat',
+    ARRAY['the baboon played with the balloon', 'the tanks fired at the buildings']);
 
-SELECT ARRAY(SELECT ROUND(x::NUMERIC,4) FROM unnest(rag_jina_reranker_v1_tiny_en.rerank_distance('query about programming',
-    ARRAY['Python is a programming language', 'Java is also a programming language', 'SQL is used for databases'])) AS x);
+SELECT rag_jina_reranker_v1_tiny_en.rerank_distance('query about programming',
+    ARRAY['Python is a programming language', 'Java is also a programming language', 'SQL is used for databases']);
 
 SELECT rag_jina_reranker_v1_tiny_en.rerank_distance('empty array test', ARRAY[]::text[]);
 
 -- Reranking score function tests - single passage
-SELECT ROUND(rag_jina_reranker_v1_tiny_en.rerank_score('the cat sat on the mat', 'the baboon played with the balloon')::NUMERIC,4);
+SELECT rag_jina_reranker_v1_tiny_en.rerank_score('the cat sat on the mat', 'the baboon played with the balloon');
 
-SELECT ROUND(rag_jina_reranker_v1_tiny_en.rerank_score('the cat sat on the mat', 'the tanks fired at the buildings')::NUMERIC,4);
+SELECT rag_jina_reranker_v1_tiny_en.rerank_score('the cat sat on the mat', 'the tanks fired at the buildings');
 
-SELECT ROUND(rag_jina_reranker_v1_tiny_en.rerank_score('query about cats', 'information about felines')::NUMERIC,4);
+SELECT rag_jina_reranker_v1_tiny_en.rerank_score('query about cats', 'information about felines');
 
-SELECT ROUND(rag_jina_reranker_v1_tiny_en.rerank_score('', 'empty query test')::NUMERIC,4);
+SELECT rag_jina_reranker_v1_tiny_en.rerank_score('', 'empty query test');
 
 -- Reranking score function tests - array of passages
-SELECT ARRAY(SELECT ROUND(x::NUMERIC,4) FROM unnest(rag_jina_reranker_v1_tiny_en.rerank_score('the cat sat on the mat',
-    ARRAY['the baboon played with the balloon', 'the tanks fired at the buildings'])) AS x);
+SELECT rag_jina_reranker_v1_tiny_en.rerank_score('the cat sat on the mat',
+    ARRAY['the baboon played with the balloon', 'the tanks fired at the buildings']);
 
-SELECT ARRAY(SELECT ROUND(x::NUMERIC,4) FROM unnest(rag_jina_reranker_v1_tiny_en.rerank_score('query about programming',
-    ARRAY['Python is a programming language', 'Java is also a programming language', 'SQL is used for databases'])) AS x);
+SELECT rag_jina_reranker_v1_tiny_en.rerank_score('query about programming',
+    ARRAY['Python is a programming language', 'Java is also a programming language', 'SQL is used for databases']);
 
 SELECT rag_jina_reranker_v1_tiny_en.rerank_score('empty array test', ARRAY[]::text[]);

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

@@ -1,7 +0,0 @@
-#!/bin/sh
-set -ex
-cd "$(dirname ${0})"
-dropdb --if-exist contrib_regression
-createdb contrib_regression
-PG_REGRESS=$(dirname "$(pg_config --pgxs)")/../test/regress/pg_regress
-${PG_REGRESS} --inputdir=./ --bindir='/usr/local/pgsql/bin' --use-existing --dbname=contrib_regression rum rum_hash ruminv timestamp orderby orderby_hash altorder altorder_hash limits int2 int4 int8 float4 float8 money oid time timetz date interval macaddr inet cidr text varchar char bytea bit varbit numeric rum_weight expr array

docker-compose/run-tests.sh

@@ -1,42 +1,6 @@
 #!/bin/bash
 set -x
 
-if [[ -v BENCHMARK_CONNSTR ]]; then
-  uri_no_proto="${BENCHMARK_CONNSTR#postgres://}"
-  uri_no_proto="${uri_no_proto#postgresql://}"
-  if [[ $uri_no_proto == *\?* ]]; then
-    base="${uri_no_proto%%\?*}"       # before '?'
-  else
-    base="$uri_no_proto"
-  fi
-  if [[ $base =~ ^([^:]+):([^@]+)@([^:/]+):?([0-9]*)/(.+)$ ]]; then
-    export PGUSER="${BASH_REMATCH[1]}"
-    export PGPASSWORD="${BASH_REMATCH[2]}"
-    export PGHOST="${BASH_REMATCH[3]}"
-    export PGPORT="${BASH_REMATCH[4]:-5432}"
-    export PGDATABASE="${BASH_REMATCH[5]}"
-    echo export PGUSER="${BASH_REMATCH[1]}"
-    echo export PGPASSWORD="${BASH_REMATCH[2]}"
-    echo export PGHOST="${BASH_REMATCH[3]}"
-    echo export PGPORT="${BASH_REMATCH[4]:-5432}"
-    echo export PGDATABASE="${BASH_REMATCH[5]}"
-  else
-    echo "Invalid PostgreSQL base URI"
-    exit 1
-  fi
-fi
-REGULAR_USER=false
-while getopts r arg; do
-  case $arg in
-  r)
-    REGULAR_USER=true
-    shift $((OPTIND-1))
-    ;;
-  *) :
-    ;;
-  esac
-done
-
 extdir=${1}
 
 cd "${extdir}" || exit 2
@@ -48,11 +12,6 @@ for d in ${LIST}; do
       FAILED="${d} ${FAILED}"
       break
     fi
-    if [[ ${REGULAR_USER} = true ]] && [ -f "${d}"/regular-test.sh ]; then
-       "${d}/regular-test.sh" || FAILED="${d} ${FAILED}"
-       continue
-    fi
-
     if [ -f "${d}/neon-test.sh" ]; then
        "${d}/neon-test.sh" || FAILED="${d} ${FAILED}"
     else
@@ -60,8 +19,5 @@ for d in ${LIST}; do
     fi
 done
 [ -z "${FAILED}" ] && exit 0
-for d in ${FAILED}; do
-  cat "$(find $d -name regression.diffs)"
-done
 echo "${FAILED}"
 exit 1

docs/consumption_metrics.md

@@ -13,7 +13,7 @@ For design details see [the RFC](./rfcs/021-metering.md) and [the discussion on
 batch format is
 ```json
 
-{ "events" : [metric1, metric2, ...] }
+{ "events" : [metric1, metric2, ...]]}
 
 ```
 See metric format examples below.
@@ -49,13 +49,11 @@ Size of the remote storage (S3) directory.
 This is an absolute, per-tenant metric.
 
 - `timeline_logical_size`
-
-Logical size of the data in the timeline.
+Logical size of the data in the timeline
 This is an absolute, per-timeline metric.
 
 - `synthetic_storage_size`
-
-Size of all tenant's branches including WAL.
+Size of all tenant's branches including WAL
 This is the same metric that `tenant/{tenant_id}/size` endpoint returns.
 This is an absolute, per-tenant metric.
 
@@ -108,10 +106,10 @@ This is an incremental, per-endpoint metric.
 ```
 
 The metric is incremental, so the value is the difference between the current and the previous value.
-If there is no previous value, the value is the current value and the `start_time` equals `stop_time`.
+If there is no previous value, the value, the value is the current value and the `start_time` equals `stop_time`.
 
 ### TODO
 
 - [ ] Handle errors better: currently if one tenant fails to gather metrics, the whole iteration fails and metrics are not sent for any tenant.
 - [ ] Add retries
-- [ ] Tune the interval
+- [ ] Tune the interval

libs/pageserver_api/src/config.rs

@@ -181,7 +181,6 @@ pub struct ConfigToml {
     pub generate_unarchival_heatmap: Option<bool>,
     pub tracing: Option<Tracing>,
     pub enable_tls_page_service_api: bool,
-    pub dev_mode: bool,
 }
 
 #[derive(Debug, Clone, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
@@ -658,7 +657,6 @@ impl Default for ConfigToml {
             generate_unarchival_heatmap: None,
             tracing: None,
             enable_tls_page_service_api: false,
-            dev_mode: false,
         }
     }
 }

libs/pageserver_api/src/controller_api.rs

@@ -169,8 +169,6 @@ pub struct TenantDescribeResponseShard {
     pub is_pending_compute_notification: bool,
     /// A shard split is currently underway
     pub is_splitting: bool,
-    /// A timeline is being imported into this tenant
-    pub is_importing: bool,
 
     pub scheduling_policy: ShardSchedulingPolicy,
 

libs/pageserver_api/src/models.rs

@@ -320,35 +320,6 @@ pub struct TimelineCreateRequest {
     pub mode: TimelineCreateRequestMode,
 }
 
-impl TimelineCreateRequest {
-    pub fn mode_tag(&self) -> &'static str {
-        match &self.mode {
-            TimelineCreateRequestMode::Branch { .. } => "branch",
-            TimelineCreateRequestMode::ImportPgdata { .. } => "import",
-            TimelineCreateRequestMode::Bootstrap { .. } => "bootstrap",
-        }
-    }
-
-    pub fn is_import(&self) -> bool {
-        matches!(self.mode, TimelineCreateRequestMode::ImportPgdata { .. })
-    }
-}
-
-#[derive(Serialize, Deserialize, Debug, Clone, PartialEq, Eq)]
-pub enum ShardImportStatus {
-    InProgress,
-    Done,
-    Error(String),
-}
-impl ShardImportStatus {
-    pub fn is_terminal(&self) -> bool {
-        match self {
-            ShardImportStatus::InProgress => false,
-            ShardImportStatus::Done | ShardImportStatus::Error(_) => true,
-        }
-    }
-}
-
 /// Storage controller specific extensions to [`TimelineInfo`].
 #[derive(Serialize, Deserialize, Clone)]
 pub struct TimelineCreateResponseStorcon {
@@ -1803,8 +1774,6 @@ pub struct TopTenantShardsResponse {
 }
 
 pub mod virtual_file {
-    use std::sync::LazyLock;
-
     #[derive(
         Copy,
         Clone,
@@ -1842,33 +1811,35 @@ pub mod virtual_file {
     pub enum IoMode {
         /// Uses buffered IO.
         Buffered,
-        /// Uses direct IO for reads only.
+        /// Uses direct IO, error out if the operation fails.
         #[cfg(target_os = "linux")]
         Direct,
-        /// Use direct IO for reads and writes.
-        #[cfg(target_os = "linux")]
-        DirectRw,
     }
 
     impl IoMode {
         pub fn preferred() -> Self {
             // The default behavior when running Rust unit tests without any further
-            // flags is to use the newest behavior (DirectRw).
+            // flags is to use the newest behavior if available on the platform (Direct).
             // The CI uses the following environment variable to unit tests for all
             // different modes.
             // NB: the Python regression & perf tests have their own defaults management
             // that writes pageserver.toml; they do not use this variable.
             if cfg!(test) {
-                static CACHED: LazyLock<IoMode> = LazyLock::new(|| {
+                use once_cell::sync::Lazy;
+                static CACHED: Lazy<IoMode> = Lazy::new(|| {
                     utils::env::var_serde_json_string(
                         "NEON_PAGESERVER_UNIT_TEST_VIRTUAL_FILE_IO_MODE",
                     )
-                    .unwrap_or(
+                    .unwrap_or({
                         #[cfg(target_os = "linux")]
-                        IoMode::DirectRw,
+                        {
+                            IoMode::Direct
+                        }
                         #[cfg(not(target_os = "linux"))]
-                        IoMode::Buffered,
-                    )
+                        {
+                            IoMode::Buffered
+                        }
+                    })
                 });
                 *CACHED
             } else {
@@ -1885,8 +1856,6 @@ pub mod virtual_file {
                 v if v == (IoMode::Buffered as u8) => IoMode::Buffered,
                 #[cfg(target_os = "linux")]
                 v if v == (IoMode::Direct as u8) => IoMode::Direct,
-                #[cfg(target_os = "linux")]
-                v if v == (IoMode::DirectRw as u8) => IoMode::DirectRw,
                 x => return Err(x),
             })
         }

libs/pageserver_api/src/upcall_api.rs

@@ -4,10 +4,10 @@
 //! See docs/rfcs/025-generation-numbers.md
 
 use serde::{Deserialize, Serialize};
-use utils::id::{NodeId, TimelineId};
+use utils::id::NodeId;
 
 use crate::controller_api::NodeRegisterRequest;
-use crate::models::{LocationConfigMode, ShardImportStatus};
+use crate::models::LocationConfigMode;
 use crate::shard::TenantShardId;
 
 /// Upcall message sent by the pageserver to the configured `control_plane_api` on
@@ -62,10 +62,3 @@ pub struct ValidateResponseTenant {
     pub id: TenantShardId,
     pub valid: bool,
 }
-
-#[derive(Serialize, Deserialize)]
-pub struct PutTimelineImportStatusRequest {
-    pub tenant_shard_id: TenantShardId,
-    pub timeline_id: TimelineId,
-    pub status: ShardImportStatus,
-}

libs/remote_storage/src/azure_blob.rs

@@ -14,9 +14,8 @@ use anyhow::{Context, Result};
 use azure_core::request_options::{IfMatchCondition, MaxResults, Metadata, Range};
 use azure_core::{Continuable, HttpClient, RetryOptions, TransportOptions};
 use azure_storage::StorageCredentials;
+use azure_storage_blobs::blob::CopyStatus;
 use azure_storage_blobs::blob::operations::GetBlobBuilder;
-use azure_storage_blobs::blob::{Blob, CopyStatus};
-use azure_storage_blobs::container::operations::ListBlobsBuilder;
 use azure_storage_blobs::prelude::{ClientBuilder, ContainerClient};
 use bytes::Bytes;
 use futures::FutureExt;
@@ -254,15 +253,53 @@ impl AzureBlobStorage {
         download
     }
 
-    fn list_streaming_for_fn<T: Default + ListingCollector>(
+    async fn permit(
+        &self,
+        kind: RequestKind,
+        cancel: &CancellationToken,
+    ) -> Result<tokio::sync::SemaphorePermit<'_>, Cancelled> {
+        let acquire = self.concurrency_limiter.acquire(kind);
+
+        tokio::select! {
+            permit = acquire => Ok(permit.expect("never closed")),
+            _ = cancel.cancelled() => Err(Cancelled),
+        }
+    }
+
+    pub fn container_name(&self) -> &str {
+        &self.container_name
+    }
+}
+
+fn to_azure_metadata(metadata: StorageMetadata) -> Metadata {
+    let mut res = Metadata::new();
+    for (k, v) in metadata.0.into_iter() {
+        res.insert(k, v);
+    }
+    res
+}
+
+fn to_download_error(error: azure_core::Error) -> DownloadError {
+    if let Some(http_err) = error.as_http_error() {
+        match http_err.status() {
+            StatusCode::NotFound => DownloadError::NotFound,
+            StatusCode::NotModified => DownloadError::Unmodified,
+            StatusCode::BadRequest => DownloadError::BadInput(anyhow::Error::new(error)),
+            _ => DownloadError::Other(anyhow::Error::new(error)),
+        }
+    } else {
+        DownloadError::Other(error.into())
+    }
+}
+
+impl RemoteStorage for AzureBlobStorage {
+    fn list_streaming(
         &self,
         prefix: Option<&RemotePath>,
         mode: ListingMode,
         max_keys: Option<NonZeroU32>,
         cancel: &CancellationToken,
-        request_kind: RequestKind,
-        customize_builder: impl Fn(ListBlobsBuilder) -> ListBlobsBuilder,
-    ) -> impl Stream<Item = Result<T, DownloadError>> {
+    ) -> 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 s| {
@@ -274,7 +311,7 @@ impl AzureBlobStorage {
         });
 
         async_stream::stream! {
-            let _permit = self.permit(request_kind, cancel).await?;
+            let _permit = self.permit(RequestKind::List, cancel).await?;
 
             let mut builder = self.client.list_blobs();
 
@@ -290,8 +327,6 @@ impl AzureBlobStorage {
                 builder = builder.max_results(MaxResults::new(limit));
             }
 
-            builder = customize_builder(builder);
-
             let mut next_marker = None;
 
             let mut timeout_try_cnt = 1;
@@ -347,20 +382,26 @@ impl AzureBlobStorage {
                     break;
                 };
 
-                let mut res = T::default();
+                let mut res = Listing::default();
                 next_marker = entry.continuation();
                 let prefix_iter = entry
                     .blobs
                     .prefixes()
                     .map(|prefix| self.name_to_relative_path(&prefix.name));
-                res.add_prefixes(self, prefix_iter);
+                res.prefixes.extend(prefix_iter);
 
                 let blob_iter = entry
                     .blobs
-                    .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,
+                    }
+                );
 
                 for key in blob_iter {
-                    res.add_blob(self, key);
+                    res.keys.push(key);
 
                     if let Some(mut mk) = max_keys {
                         assert!(mk > 0);
@@ -382,128 +423,6 @@ impl AzureBlobStorage {
         }
     }
 
-    async fn permit(
-        &self,
-        kind: RequestKind,
-        cancel: &CancellationToken,
-    ) -> Result<tokio::sync::SemaphorePermit<'_>, Cancelled> {
-        let acquire = self.concurrency_limiter.acquire(kind);
-
-        tokio::select! {
-            permit = acquire => Ok(permit.expect("never closed")),
-            _ = cancel.cancelled() => Err(Cancelled),
-        }
-    }
-
-    pub fn container_name(&self) -> &str {
-        &self.container_name
-    }
-}
-
-trait ListingCollector {
-    fn add_prefixes(&mut self, abs: &AzureBlobStorage, prefix_it: impl Iterator<Item = RemotePath>);
-    fn add_blob(&mut self, abs: &AzureBlobStorage, blob: &Blob);
-}
-
-impl ListingCollector for Listing {
-    fn add_prefixes(
-        &mut self,
-        _abs: &AzureBlobStorage,
-        prefix_it: impl Iterator<Item = RemotePath>,
-    ) {
-        self.prefixes.extend(prefix_it);
-    }
-    fn add_blob(&mut self, abs: &AzureBlobStorage, blob: &Blob) {
-        self.keys.push(ListingObject {
-            key: abs.name_to_relative_path(&blob.name),
-            last_modified: blob.properties.last_modified.into(),
-            size: blob.properties.content_length,
-        });
-    }
-}
-
-impl ListingCollector for crate::VersionListing {
-    fn add_prefixes(
-        &mut self,
-        _abs: &AzureBlobStorage,
-        _prefix_it: impl Iterator<Item = RemotePath>,
-    ) {
-        // nothing
-    }
-    fn add_blob(&mut self, abs: &AzureBlobStorage, blob: &Blob) {
-        let id = crate::VersionId(blob.version_id.clone().expect("didn't find version ID"));
-        self.versions.push(crate::Version {
-            key: abs.name_to_relative_path(&blob.name),
-            last_modified: blob.properties.last_modified.into(),
-            kind: crate::VersionKind::Version(id),
-        });
-    }
-}
-
-fn to_azure_metadata(metadata: StorageMetadata) -> Metadata {
-    let mut res = Metadata::new();
-    for (k, v) in metadata.0.into_iter() {
-        res.insert(k, v);
-    }
-    res
-}
-
-fn to_download_error(error: azure_core::Error) -> DownloadError {
-    if let Some(http_err) = error.as_http_error() {
-        match http_err.status() {
-            StatusCode::NotFound => DownloadError::NotFound,
-            StatusCode::NotModified => DownloadError::Unmodified,
-            StatusCode::BadRequest => DownloadError::BadInput(anyhow::Error::new(error)),
-            _ => DownloadError::Other(anyhow::Error::new(error)),
-        }
-    } else {
-        DownloadError::Other(error.into())
-    }
-}
-
-impl RemoteStorage for AzureBlobStorage {
-    fn list_streaming(
-        &self,
-        prefix: Option<&RemotePath>,
-        mode: ListingMode,
-        max_keys: Option<NonZeroU32>,
-        cancel: &CancellationToken,
-    ) -> impl Stream<Item = Result<Listing, DownloadError>> {
-        let customize_builder = |builder| builder;
-        let kind = RequestKind::ListVersions;
-        self.list_streaming_for_fn(prefix, mode, max_keys, cancel, kind, customize_builder)
-    }
-
-    async fn list_versions(
-        &self,
-        prefix: Option<&RemotePath>,
-        mode: ListingMode,
-        max_keys: Option<NonZeroU32>,
-        cancel: &CancellationToken,
-    ) -> std::result::Result<crate::VersionListing, DownloadError> {
-        let customize_builder = |mut builder: ListBlobsBuilder| {
-            builder = builder.include_versions(true);
-            builder
-        };
-        let kind = RequestKind::ListVersions;
-
-        let mut stream = std::pin::pin!(self.list_streaming_for_fn(
-            prefix,
-            mode,
-            max_keys,
-            cancel,
-            kind,
-            customize_builder
-        ));
-        let mut combined: crate::VersionListing =
-            stream.next().await.expect("At least one item required")?;
-        while let Some(list) = stream.next().await {
-            let list = list?;
-            combined.versions.extend(list.versions.into_iter());
-        }
-        Ok(combined)
-    }
-
     async fn head_object(
         &self,
         key: &RemotePath,
@@ -613,12 +532,7 @@ impl RemoteStorage for AzureBlobStorage {
         let mut builder = blob_client.get();
 
         if let Some(ref etag) = opts.etag {
-            builder = builder.if_match(IfMatchCondition::NotMatch(etag.to_string()));
-        }
-
-        if let Some(ref version_id) = opts.version_id {
-            let version_id = azure_storage_blobs::prelude::VersionId::new(version_id.0.clone());
-            builder = builder.blob_versioning(version_id);
+            builder = builder.if_match(IfMatchCondition::NotMatch(etag.to_string()))
         }
 
         if let Some((start, end)) = opts.byte_range() {

libs/remote_storage/src/lib.rs

@@ -176,32 +176,6 @@ pub struct Listing {
     pub keys: Vec<ListingObject>,
 }
 
-#[derive(Default)]
-pub struct VersionListing {
-    pub versions: Vec<Version>,
-}
-
-pub struct Version {
-    pub key: RemotePath,
-    pub last_modified: SystemTime,
-    pub kind: VersionKind,
-}
-
-impl Version {
-    pub fn version_id(&self) -> Option<&VersionId> {
-        match &self.kind {
-            VersionKind::Version(id) => Some(id),
-            VersionKind::DeletionMarker => None,
-        }
-    }
-}
-
-#[derive(Debug)]
-pub enum VersionKind {
-    DeletionMarker,
-    Version(VersionId),
-}
-
 /// Options for downloads. The default value is a plain GET.
 pub struct DownloadOpts {
     /// If given, returns [`DownloadError::Unmodified`] if the object still has
@@ -212,8 +186,6 @@ pub struct DownloadOpts {
     /// The end of the byte range to download, or unbounded. Must be after the
     /// start bound.
     pub byte_end: Bound<u64>,
-    /// Optionally request a specific version of a key
-    pub version_id: Option<VersionId>,
     /// Indicate whether we're downloading something small or large: this indirectly controls
     /// timeouts: for something like an index/manifest/heatmap, we should time out faster than
     /// for layer files
@@ -225,16 +197,12 @@ pub enum DownloadKind {
     Small,
 }
 
-#[derive(Debug, Clone)]
-pub struct VersionId(pub String);
-
 impl Default for DownloadOpts {
     fn default() -> Self {
         Self {
             etag: Default::default(),
             byte_start: Bound::Unbounded,
             byte_end: Bound::Unbounded,
-            version_id: None,
             kind: DownloadKind::Large,
         }
     }
@@ -327,14 +295,6 @@ pub trait RemoteStorage: Send + Sync + 'static {
         Ok(combined)
     }
 
-    async fn list_versions(
-        &self,
-        prefix: Option<&RemotePath>,
-        mode: ListingMode,
-        max_keys: Option<NonZeroU32>,
-        cancel: &CancellationToken,
-    ) -> Result<VersionListing, DownloadError>;
-
     /// Obtain metadata information about an object.
     async fn head_object(
         &self,
@@ -515,22 +475,6 @@ impl<Other: RemoteStorage> GenericRemoteStorage<Arc<Other>> {
         }
     }
 
-    // See [`RemoteStorage::list_versions`].
-    pub async fn list_versions<'a>(
-        &'a self,
-        prefix: Option<&'a RemotePath>,
-        mode: ListingMode,
-        max_keys: Option<NonZeroU32>,
-        cancel: &'a CancellationToken,
-    ) -> Result<VersionListing, DownloadError> {
-        match self {
-            Self::LocalFs(s) => s.list_versions(prefix, mode, max_keys, cancel).await,
-            Self::AwsS3(s) => s.list_versions(prefix, mode, max_keys, cancel).await,
-            Self::AzureBlob(s) => s.list_versions(prefix, mode, max_keys, cancel).await,
-            Self::Unreliable(s) => s.list_versions(prefix, mode, max_keys, cancel).await,
-        }
-    }
-
     // See [`RemoteStorage::head_object`].
     pub async fn head_object(
         &self,
@@ -783,7 +727,6 @@ impl ConcurrencyLimiter {
             RequestKind::Copy => &self.write,
             RequestKind::TimeTravel => &self.write,
             RequestKind::Head => &self.read,
-            RequestKind::ListVersions => &self.read,
         }
     }
 

libs/remote_storage/src/local_fs.rs

@@ -445,16 +445,6 @@ impl RemoteStorage for LocalFs {
         }
     }
 
-    async fn list_versions(
-        &self,
-        _prefix: Option<&RemotePath>,
-        _mode: ListingMode,
-        _max_keys: Option<NonZeroU32>,
-        _cancel: &CancellationToken,
-    ) -> Result<crate::VersionListing, DownloadError> {
-        unimplemented!()
-    }
-
     async fn head_object(
         &self,
         key: &RemotePath,

libs/remote_storage/src/metrics.rs

@@ -14,7 +14,6 @@ pub(crate) enum RequestKind {
     Copy = 4,
     TimeTravel = 5,
     Head = 6,
-    ListVersions = 7,
 }
 
 use RequestKind::*;
@@ -30,7 +29,6 @@ impl RequestKind {
             Copy => "copy_object",
             TimeTravel => "time_travel_recover",
             Head => "head_object",
-            ListVersions => "list_versions",
         }
     }
     const fn as_index(&self) -> usize {
@@ -38,10 +36,7 @@ impl RequestKind {
     }
 }
 
-const REQUEST_KIND_LIST: &[RequestKind] =
-    &[Get, Put, Delete, List, Copy, TimeTravel, Head, ListVersions];
-
-const REQUEST_KIND_COUNT: usize = REQUEST_KIND_LIST.len();
+const REQUEST_KIND_COUNT: usize = 7;
 pub(crate) struct RequestTyped<C>([C; REQUEST_KIND_COUNT]);
 
 impl<C> RequestTyped<C> {
@@ -50,11 +45,12 @@ impl<C> RequestTyped<C> {
     }
 
     fn build_with(mut f: impl FnMut(RequestKind) -> C) -> Self {
-        let mut it = REQUEST_KIND_LIST.iter();
+        use RequestKind::*;
+        let mut it = [Get, Put, Delete, List, Copy, TimeTravel, Head].into_iter();
         let arr = std::array::from_fn::<C, REQUEST_KIND_COUNT, _>(|index| {
             let next = it.next().unwrap();
             assert_eq!(index, next.as_index());
-            f(*next)
+            f(next)
         });
 
         if let Some(next) = it.next() {

libs/remote_storage/src/s3_bucket.rs

@@ -21,8 +21,9 @@ use aws_sdk_s3::config::{AsyncSleep, IdentityCache, Region, SharedAsyncSleep};
 use aws_sdk_s3::error::SdkError;
 use aws_sdk_s3::operation::get_object::GetObjectError;
 use aws_sdk_s3::operation::head_object::HeadObjectError;
-use aws_sdk_s3::types::{Delete, ObjectIdentifier, StorageClass};
+use aws_sdk_s3::types::{Delete, DeleteMarkerEntry, ObjectIdentifier, ObjectVersion, StorageClass};
 use aws_smithy_async::rt::sleep::TokioSleep;
+use aws_smithy_types::DateTime;
 use aws_smithy_types::body::SdkBody;
 use aws_smithy_types::byte_stream::ByteStream;
 use aws_smithy_types::date_time::ConversionError;
@@ -45,7 +46,7 @@ use crate::support::PermitCarrying;
 use crate::{
     ConcurrencyLimiter, Download, DownloadError, DownloadOpts, Listing, ListingMode, ListingObject,
     MAX_KEYS_PER_DELETE_S3, REMOTE_STORAGE_PREFIX_SEPARATOR, RemotePath, RemoteStorage,
-    TimeTravelError, TimeoutOrCancel, Version, VersionId, VersionKind, VersionListing,
+    TimeTravelError, TimeoutOrCancel,
 };
 
 /// AWS S3 storage.
@@ -65,7 +66,6 @@ struct GetObjectRequest {
     key: String,
     etag: Option<String>,
     range: Option<String>,
-    version_id: Option<String>,
 }
 impl S3Bucket {
     /// Creates the S3 storage, errors if incorrect AWS S3 configuration provided.
@@ -251,7 +251,6 @@ impl S3Bucket {
             .get_object()
             .bucket(request.bucket)
             .key(request.key)
-            .set_version_id(request.version_id)
             .set_range(request.range);
 
         if let Some(etag) = request.etag {
@@ -406,124 +405,6 @@ impl S3Bucket {
         Ok(())
     }
 
-    async fn list_versions_with_permit(
-        &self,
-        _permit: &tokio::sync::SemaphorePermit<'_>,
-        prefix: Option<&RemotePath>,
-        mode: ListingMode,
-        max_keys: Option<NonZeroU32>,
-        cancel: &CancellationToken,
-    ) -> Result<crate::VersionListing, DownloadError> {
-        // get the passed prefix or if it is not set use prefix_in_bucket value
-        let prefix = prefix
-            .map(|p| self.relative_path_to_s3_object(p))
-            .or_else(|| self.prefix_in_bucket.clone());
-
-        let warn_threshold = 3;
-        let max_retries = 10;
-        let is_permanent = |e: &_| matches!(e, DownloadError::Cancelled);
-
-        let mut key_marker = None;
-        let mut version_id_marker = None;
-        let mut versions_and_deletes = Vec::new();
-
-        loop {
-            let response = backoff::retry(
-                || async {
-                    let mut request = self
-                        .client
-                        .list_object_versions()
-                        .bucket(self.bucket_name.clone())
-                        .set_prefix(prefix.clone())
-                        .set_key_marker(key_marker.clone())
-                        .set_version_id_marker(version_id_marker.clone());
-
-                    if let ListingMode::WithDelimiter = mode {
-                        request = request.delimiter(REMOTE_STORAGE_PREFIX_SEPARATOR.to_string());
-                    }
-
-                    let op = request.send();
-
-                    tokio::select! {
-                        res = op => res.map_err(|e| DownloadError::Other(e.into())),
-                        _ = cancel.cancelled() => Err(DownloadError::Cancelled),
-                    }
-                },
-                is_permanent,
-                warn_threshold,
-                max_retries,
-                "listing object versions",
-                cancel,
-            )
-            .await
-            .ok_or_else(|| DownloadError::Cancelled)
-            .and_then(|x| x)?;
-
-            tracing::trace!(
-                "  Got List response version_id_marker={:?}, key_marker={:?}",
-                response.version_id_marker,
-                response.key_marker
-            );
-            let versions = response
-                .versions
-                .unwrap_or_default()
-                .into_iter()
-                .map(|version| {
-                    let key = version.key.expect("response does not contain a key");
-                    let key = self.s3_object_to_relative_path(&key);
-                    let version_id = VersionId(version.version_id.expect("needing version id"));
-                    let last_modified =
-                        SystemTime::try_from(version.last_modified.expect("no last_modified"))?;
-                    Ok(Version {
-                        key,
-                        last_modified,
-                        kind: crate::VersionKind::Version(version_id),
-                    })
-                });
-            let deletes = response
-                .delete_markers
-                .unwrap_or_default()
-                .into_iter()
-                .map(|version| {
-                    let key = version.key.expect("response does not contain a key");
-                    let key = self.s3_object_to_relative_path(&key);
-                    let last_modified =
-                        SystemTime::try_from(version.last_modified.expect("no last_modified"))?;
-                    Ok(Version {
-                        key,
-                        last_modified,
-                        kind: crate::VersionKind::DeletionMarker,
-                    })
-                });
-            itertools::process_results(versions.chain(deletes), |n_vds| {
-                versions_and_deletes.extend(n_vds)
-            })
-            .map_err(DownloadError::Other)?;
-            fn none_if_empty(v: Option<String>) -> Option<String> {
-                v.filter(|v| !v.is_empty())
-            }
-            version_id_marker = none_if_empty(response.next_version_id_marker);
-            key_marker = none_if_empty(response.next_key_marker);
-            if version_id_marker.is_none() {
-                // The final response is not supposed to be truncated
-                if response.is_truncated.unwrap_or_default() {
-                    return Err(DownloadError::Other(anyhow::anyhow!(
-                        "Received truncated ListObjectVersions response for prefix={prefix:?}"
-                    )));
-                }
-                break;
-            }
-            if let Some(max_keys) = max_keys {
-                if versions_and_deletes.len() >= max_keys.get().try_into().unwrap() {
-                    return Err(DownloadError::Other(anyhow::anyhow!("too many versions")));
-                }
-            }
-        }
-        Ok(VersionListing {
-            versions: versions_and_deletes,
-        })
-    }
-
     pub fn bucket_name(&self) -> &str {
         &self.bucket_name
     }
@@ -740,19 +621,6 @@ impl RemoteStorage for S3Bucket {
         }
     }
 
-    async fn list_versions(
-        &self,
-        prefix: Option<&RemotePath>,
-        mode: ListingMode,
-        max_keys: Option<NonZeroU32>,
-        cancel: &CancellationToken,
-    ) -> Result<crate::VersionListing, DownloadError> {
-        let kind = RequestKind::ListVersions;
-        let permit = self.permit(kind, cancel).await?;
-        self.list_versions_with_permit(&permit, prefix, mode, max_keys, cancel)
-            .await
-    }
-
     async fn head_object(
         &self,
         key: &RemotePath,
@@ -933,7 +801,6 @@ impl RemoteStorage for S3Bucket {
                 key: self.relative_path_to_s3_object(from),
                 etag: opts.etag.as_ref().map(|e| e.to_string()),
                 range: opts.byte_range_header(),
-                version_id: opts.version_id.as_ref().map(|v| v.0.to_owned()),
             },
             cancel,
         )
@@ -978,25 +845,94 @@ impl RemoteStorage for S3Bucket {
         let kind = RequestKind::TimeTravel;
         let permit = self.permit(kind, cancel).await?;
 
+        let timestamp = DateTime::from(timestamp);
+        let done_if_after = DateTime::from(done_if_after);
+
         tracing::trace!("Target time: {timestamp:?}, done_if_after {done_if_after:?}");
 
-        // Limit the number of versions deletions, mostly so that we don't
-        // keep requesting forever if the list is too long, as we'd put the
-        // list in RAM.
-        // Building a list of 100k entries that reaches the limit roughly takes
-        // 40 seconds, and roughly corresponds to tenants of 2 TiB physical size.
-        const COMPLEXITY_LIMIT: Option<NonZeroU32> = NonZeroU32::new(100_000);
+        // get the passed prefix or if it is not set use prefix_in_bucket value
+        let prefix = prefix
+            .map(|p| self.relative_path_to_s3_object(p))
+            .or_else(|| self.prefix_in_bucket.clone());
 
-        let mode = ListingMode::NoDelimiter;
-        let version_listing = self
-            .list_versions_with_permit(&permit, prefix, mode, COMPLEXITY_LIMIT, cancel)
+        let warn_threshold = 3;
+        let max_retries = 10;
+        let is_permanent = |e: &_| matches!(e, TimeTravelError::Cancelled);
+
+        let mut key_marker = None;
+        let mut version_id_marker = None;
+        let mut versions_and_deletes = Vec::new();
+
+        loop {
+            let response = backoff::retry(
+                || async {
+                    let op = self
+                        .client
+                        .list_object_versions()
+                        .bucket(self.bucket_name.clone())
+                        .set_prefix(prefix.clone())
+                        .set_key_marker(key_marker.clone())
+                        .set_version_id_marker(version_id_marker.clone())
+                        .send();
+
+                    tokio::select! {
+                        res = op => res.map_err(|e| TimeTravelError::Other(e.into())),
+                        _ = cancel.cancelled() => Err(TimeTravelError::Cancelled),
+                    }
+                },
+                is_permanent,
+                warn_threshold,
+                max_retries,
+                "listing object versions for time_travel_recover",
+                cancel,
+            )
             .await
-            .map_err(|err| match err {
-                DownloadError::Other(e) => TimeTravelError::Other(e),
-                DownloadError::Cancelled => TimeTravelError::Cancelled,
-                other => TimeTravelError::Other(other.into()),
-            })?;
-        let versions_and_deletes = version_listing.versions;
+            .ok_or_else(|| TimeTravelError::Cancelled)
+            .and_then(|x| x)?;
+
+            tracing::trace!(
+                "  Got List response version_id_marker={:?}, key_marker={:?}",
+                response.version_id_marker,
+                response.key_marker
+            );
+            let versions = response
+                .versions
+                .unwrap_or_default()
+                .into_iter()
+                .map(VerOrDelete::from_version);
+            let deletes = response
+                .delete_markers
+                .unwrap_or_default()
+                .into_iter()
+                .map(VerOrDelete::from_delete_marker);
+            itertools::process_results(versions.chain(deletes), |n_vds| {
+                versions_and_deletes.extend(n_vds)
+            })
+            .map_err(TimeTravelError::Other)?;
+            fn none_if_empty(v: Option<String>) -> Option<String> {
+                v.filter(|v| !v.is_empty())
+            }
+            version_id_marker = none_if_empty(response.next_version_id_marker);
+            key_marker = none_if_empty(response.next_key_marker);
+            if version_id_marker.is_none() {
+                // The final response is not supposed to be truncated
+                if response.is_truncated.unwrap_or_default() {
+                    return Err(TimeTravelError::Other(anyhow::anyhow!(
+                        "Received truncated ListObjectVersions response for prefix={prefix:?}"
+                    )));
+                }
+                break;
+            }
+            // Limit the number of versions deletions, mostly so that we don't
+            // keep requesting forever if the list is too long, as we'd put the
+            // list in RAM.
+            // Building a list of 100k entries that reaches the limit roughly takes
+            // 40 seconds, and roughly corresponds to tenants of 2 TiB physical size.
+            const COMPLEXITY_LIMIT: usize = 100_000;
+            if versions_and_deletes.len() >= COMPLEXITY_LIMIT {
+                return Err(TimeTravelError::TooManyVersions);
+            }
+        }
 
         tracing::info!(
             "Built list for time travel with {} versions and deletions",
@@ -1012,26 +948,24 @@ impl RemoteStorage for S3Bucket {
         let mut vds_for_key = HashMap::<_, Vec<_>>::new();
 
         for vd in &versions_and_deletes {
-            let Version { key, .. } = &vd;
-            let version_id = vd.version_id().map(|v| v.0.as_str());
-            if version_id == Some("null") {
+            let VerOrDelete {
+                version_id, key, ..
+            } = &vd;
+            if version_id == "null" {
                 return Err(TimeTravelError::Other(anyhow!(
                     "Received ListVersions response for key={key} with version_id='null', \
                     indicating either disabled versioning, or legacy objects with null version id values"
                 )));
             }
-            tracing::trace!("Parsing version key={key} kind={:?}", vd.kind);
+            tracing::trace!(
+                "Parsing version key={key} version_id={version_id} kind={:?}",
+                vd.kind
+            );
 
             vds_for_key.entry(key).or_default().push(vd);
         }
-
-        let warn_threshold = 3;
-        let max_retries = 10;
-        let is_permanent = |e: &_| matches!(e, TimeTravelError::Cancelled);
-
         for (key, versions) in vds_for_key {
             let last_vd = versions.last().unwrap();
-            let key = self.relative_path_to_s3_object(key);
             if last_vd.last_modified > done_if_after {
                 tracing::trace!("Key {key} has version later than done_if_after, skipping");
                 continue;
@@ -1056,11 +990,11 @@ impl RemoteStorage for S3Bucket {
                 do_delete = true;
             } else {
                 match &versions[version_to_restore_to - 1] {
-                    Version {
-                        kind: VersionKind::Version(version_id),
+                    VerOrDelete {
+                        kind: VerOrDeleteKind::Version,
+                        version_id,
                         ..
                     } => {
-                        let version_id = &version_id.0;
                         tracing::trace!("Copying old version {version_id} for {key}...");
                         // Restore the state to the last version by copying
                         let source_id =
@@ -1072,7 +1006,7 @@ impl RemoteStorage for S3Bucket {
                                     .client
                                     .copy_object()
                                     .bucket(self.bucket_name.clone())
-                                    .key(&key)
+                                    .key(key)
                                     .set_storage_class(self.upload_storage_class.clone())
                                     .copy_source(&source_id)
                                     .send();
@@ -1093,8 +1027,8 @@ impl RemoteStorage for S3Bucket {
                         .and_then(|x| x)?;
                         tracing::info!(%version_id, %key, "Copied old version in S3");
                     }
-                    Version {
-                        kind: VersionKind::DeletionMarker,
+                    VerOrDelete {
+                        kind: VerOrDeleteKind::DeleteMarker,
                         ..
                     } => {
                         do_delete = true;
@@ -1102,7 +1036,7 @@ impl RemoteStorage for S3Bucket {
                 }
             };
             if do_delete {
-                if matches!(last_vd.kind, VersionKind::DeletionMarker) {
+                if matches!(last_vd.kind, VerOrDeleteKind::DeleteMarker) {
                     // Key has since been deleted (but there was some history), no need to do anything
                     tracing::trace!("Key {key} already deleted, skipping.");
                 } else {
@@ -1130,6 +1064,62 @@ impl RemoteStorage for S3Bucket {
     }
 }
 
+// Save RAM and only store the needed data instead of the entire ObjectVersion/DeleteMarkerEntry
+struct VerOrDelete {
+    kind: VerOrDeleteKind,
+    last_modified: DateTime,
+    version_id: String,
+    key: String,
+}
+
+#[derive(Debug)]
+enum VerOrDeleteKind {
+    Version,
+    DeleteMarker,
+}
+
+impl VerOrDelete {
+    fn with_kind(
+        kind: VerOrDeleteKind,
+        last_modified: Option<DateTime>,
+        version_id: Option<String>,
+        key: Option<String>,
+    ) -> anyhow::Result<Self> {
+        let lvk = (last_modified, version_id, key);
+        let (Some(last_modified), Some(version_id), Some(key)) = lvk else {
+            anyhow::bail!(
+                "One (or more) of last_modified, key, and id is None. \
+            Is versioning enabled in the bucket? last_modified={:?}, version_id={:?}, key={:?}",
+                lvk.0,
+                lvk.1,
+                lvk.2,
+            );
+        };
+        Ok(Self {
+            kind,
+            last_modified,
+            version_id,
+            key,
+        })
+    }
+    fn from_version(v: ObjectVersion) -> anyhow::Result<Self> {
+        Self::with_kind(
+            VerOrDeleteKind::Version,
+            v.last_modified,
+            v.version_id,
+            v.key,
+        )
+    }
+    fn from_delete_marker(v: DeleteMarkerEntry) -> anyhow::Result<Self> {
+        Self::with_kind(
+            VerOrDeleteKind::DeleteMarker,
+            v.last_modified,
+            v.version_id,
+            v.key,
+        )
+    }
+}
+
 #[cfg(test)]
 mod tests {
     use std::num::NonZeroUsize;

libs/remote_storage/src/simulate_failures.rs

@@ -139,20 +139,6 @@ impl RemoteStorage for UnreliableWrapper {
         self.inner.list(prefix, mode, max_keys, cancel).await
     }
 
-    async fn list_versions(
-        &self,
-        prefix: Option<&RemotePath>,
-        mode: ListingMode,
-        max_keys: Option<NonZeroU32>,
-        cancel: &CancellationToken,
-    ) -> Result<crate::VersionListing, DownloadError> {
-        self.attempt(RemoteOp::ListPrefixes(prefix.cloned()))
-            .map_err(DownloadError::Other)?;
-        self.inner
-            .list_versions(prefix, mode, max_keys, cancel)
-            .await
-    }
-
     async fn head_object(
         &self,
         key: &RemotePath,

pageserver/Cargo.toml

@@ -106,7 +106,6 @@ hex-literal.workspace = true
 tokio = { workspace = true, features = ["process", "sync", "fs", "rt", "io-util", "time", "test-util"] }
 indoc.workspace = true
 uuid.workspace = true
-rstest.workspace = true
 
 [[bench]]
 name = "bench_layer_map"

pageserver/benches/bench_ingest.rs

@@ -11,7 +11,6 @@ use pageserver::task_mgr::TaskKind;
 use pageserver::tenant::storage_layer::InMemoryLayer;
 use pageserver::{page_cache, virtual_file};
 use pageserver_api::key::Key;
-use pageserver_api::models::virtual_file::IoMode;
 use pageserver_api::shard::TenantShardId;
 use pageserver_api::value::Value;
 use tokio_util::sync::CancellationToken;
@@ -29,7 +28,6 @@ fn murmurhash32(mut h: u32) -> u32 {
     h
 }
 
-#[derive(serde::Serialize, Clone, Copy, Debug)]
 enum KeyLayout {
     /// Sequential unique keys
     Sequential,
@@ -39,7 +37,6 @@ enum KeyLayout {
     RandomReuse(u32),
 }
 
-#[derive(serde::Serialize, Clone, Copy, Debug)]
 enum WriteDelta {
     Yes,
     No,
@@ -141,15 +138,12 @@ async fn ingest(
 /// Wrapper to instantiate a tokio runtime
 fn ingest_main(
     conf: &'static PageServerConf,
-    io_mode: IoMode,
     put_size: usize,
     put_count: usize,
     key_layout: KeyLayout,
     write_delta: WriteDelta,
 ) {
-    pageserver::virtual_file::set_io_mode(io_mode);
-
-    let runtime = tokio::runtime::Builder::new_multi_thread()
+    let runtime = tokio::runtime::Builder::new_current_thread()
         .enable_all()
         .build()
         .unwrap();
@@ -180,245 +174,93 @@ fn criterion_benchmark(c: &mut Criterion) {
     virtual_file::init(
         16384,
         virtual_file::io_engine_for_bench(),
-        // immaterial, each `ingest_main` invocation below overrides this
         conf.virtual_file_io_mode,
-        // without actually doing syncs, buffered writes have an unfair advantage over direct IO writes
         virtual_file::SyncMode::Sync,
     );
     page_cache::init(conf.page_cache_size);
 
-    #[derive(serde::Serialize)]
-    struct ExplodedParameters {
-        io_mode: IoMode,
-        volume_mib: usize,
-        key_size: usize,
-        key_layout: KeyLayout,
-        write_delta: WriteDelta,
-    }
-    #[derive(Clone)]
-    struct HandPickedParameters {
-        volume_mib: usize,
-        key_size: usize,
-        key_layout: KeyLayout,
-        write_delta: WriteDelta,
-    }
-    let expect = vec![
-        // Small values (100b) tests
-        HandPickedParameters {
-            volume_mib: 128,
-            key_size: 100,
-            key_layout: KeyLayout::Sequential,
-            write_delta: WriteDelta::Yes,
-        },
-        HandPickedParameters {
-            volume_mib: 128,
-            key_size: 100,
-            key_layout: KeyLayout::Random,
-            write_delta: WriteDelta::Yes,
-        },
-        HandPickedParameters {
-            volume_mib: 128,
-            key_size: 100,
-            key_layout: KeyLayout::RandomReuse(0x3ff),
-            write_delta: WriteDelta::Yes,
-        },
-        HandPickedParameters {
-            volume_mib: 128,
-            key_size: 100,
-            key_layout: KeyLayout::Sequential,
-            write_delta: WriteDelta::No,
-        },
-        // Large values (8k) tests
-        HandPickedParameters {
-            volume_mib: 128,
-            key_size: 8192,
-            key_layout: KeyLayout::Sequential,
-            write_delta: WriteDelta::Yes,
-        },
-        HandPickedParameters {
-            volume_mib: 128,
-            key_size: 8192,
-            key_layout: KeyLayout::Sequential,
-            write_delta: WriteDelta::No,
-        },
-    ];
-    let exploded_parameters = {
-        let mut out = Vec::new();
-        for io_mode in [
-            IoMode::Buffered,
-            #[cfg(target_os = "linux")]
-            IoMode::Direct,
-            #[cfg(target_os = "linux")]
-            IoMode::DirectRw,
-        ] {
-            for param in expect.clone() {
-                let HandPickedParameters {
-                    volume_mib,
-                    key_size,
-                    key_layout,
-                    write_delta,
-                } = param;
-                out.push(ExplodedParameters {
-                    io_mode,
-                    volume_mib,
-                    key_size,
-                    key_layout,
-                    write_delta,
-                });
-            }
-        }
-        out
-    };
-    impl ExplodedParameters {
-        fn benchmark_id(&self) -> String {
-            let ExplodedParameters {
-                io_mode,
-                volume_mib,
-                key_size,
-                key_layout,
-                write_delta,
-            } = self;
-            format!(
-                "io_mode={io_mode:?} volume_mib={volume_mib:?} key_size_bytes={key_size:?} key_layout={key_layout:?} write_delta={write_delta:?}"
-            )
-        }
-    }
-    let mut group = c.benchmark_group("ingest");
-    for params in exploded_parameters {
-        let id = params.benchmark_id();
-        let ExplodedParameters {
-            io_mode,
-            volume_mib,
-            key_size,
-            key_layout,
-            write_delta,
-        } = params;
-        let put_count = volume_mib * 1024 * 1024 / key_size;
-        group.throughput(criterion::Throughput::Bytes((key_size * put_count) as u64));
+    {
+        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(id, |b| {
-            b.iter(|| ingest_main(conf, io_mode, key_size, put_count, key_layout, write_delta))
+        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);
-
-/*
-cargo bench --bench bench_ingest
-
-im4gn.2xlarge:
-
-ingest/io_mode=Buffered volume_mib=128 key_size_bytes=100 key_layout=Sequential write_delta=Yes
-                        time:   [1.2901 s 1.2943 s 1.2991 s]
-                        thrpt:  [98.533 MiB/s 98.892 MiB/s 99.220 MiB/s]
-ingest/io_mode=Buffered volume_mib=128 key_size_bytes=100 key_layout=Random write_delta=Yes
-                        time:   [2.1387 s 2.1623 s 2.1845 s]
-                        thrpt:  [58.595 MiB/s 59.197 MiB/s 59.851 MiB/s]
-ingest/io_mode=Buffered volume_mib=128 key_size_bytes=100 key_layout=RandomReuse(1023) write_delta=Y...
-                        time:   [1.2036 s 1.2074 s 1.2122 s]
-                        thrpt:  [105.60 MiB/s 106.01 MiB/s 106.35 MiB/s]
-ingest/io_mode=Buffered volume_mib=128 key_size_bytes=100 key_layout=Sequential write_delta=No
-                        time:   [520.55 ms 521.46 ms 522.57 ms]
-                        thrpt:  [244.94 MiB/s 245.47 MiB/s 245.89 MiB/s]
-ingest/io_mode=Buffered volume_mib=128 key_size_bytes=8192 key_layout=Sequential write_delta=Yes
-                        time:   [440.33 ms 442.24 ms 444.10 ms]
-                        thrpt:  [288.22 MiB/s 289.43 MiB/s 290.69 MiB/s]
-ingest/io_mode=Buffered volume_mib=128 key_size_bytes=8192 key_layout=Sequential write_delta=No
-                        time:   [168.78 ms 169.42 ms 170.18 ms]
-                        thrpt:  [752.16 MiB/s 755.52 MiB/s 758.40 MiB/s]
-ingest/io_mode=Direct volume_mib=128 key_size_bytes=100 key_layout=Sequential write_delta=Yes
-                        time:   [1.2978 s 1.3094 s 1.3227 s]
-                        thrpt:  [96.775 MiB/s 97.758 MiB/s 98.632 MiB/s]
-ingest/io_mode=Direct volume_mib=128 key_size_bytes=100 key_layout=Random write_delta=Yes
-                        time:   [2.1976 s 2.2067 s 2.2154 s]
-                        thrpt:  [57.777 MiB/s 58.006 MiB/s 58.245 MiB/s]
-ingest/io_mode=Direct volume_mib=128 key_size_bytes=100 key_layout=RandomReuse(1023) write_delta=Yes
-                        time:   [1.2103 s 1.2160 s 1.2233 s]
-                        thrpt:  [104.64 MiB/s 105.26 MiB/s 105.76 MiB/s]
-ingest/io_mode=Direct volume_mib=128 key_size_bytes=100 key_layout=Sequential write_delta=No
-                        time:   [525.05 ms 526.37 ms 527.79 ms]
-                        thrpt:  [242.52 MiB/s 243.17 MiB/s 243.79 MiB/s]
-ingest/io_mode=Direct volume_mib=128 key_size_bytes=8192 key_layout=Sequential write_delta=Yes
-                        time:   [443.06 ms 444.88 ms 447.15 ms]
-                        thrpt:  [286.26 MiB/s 287.72 MiB/s 288.90 MiB/s]
-ingest/io_mode=Direct volume_mib=128 key_size_bytes=8192 key_layout=Sequential write_delta=No
-                        time:   [169.40 ms 169.80 ms 170.17 ms]
-                        thrpt:  [752.21 MiB/s 753.81 MiB/s 755.60 MiB/s]
-ingest/io_mode=DirectRw volume_mib=128 key_size_bytes=100 key_layout=Sequential write_delta=Yes
-                        time:   [1.2844 s 1.2915 s 1.2990 s]
-                        thrpt:  [98.536 MiB/s 99.112 MiB/s 99.657 MiB/s]
-ingest/io_mode=DirectRw volume_mib=128 key_size_bytes=100 key_layout=Random write_delta=Yes
-                        time:   [2.1431 s 2.1663 s 2.1900 s]
-                        thrpt:  [58.446 MiB/s 59.087 MiB/s 59.726 MiB/s]
-ingest/io_mode=DirectRw volume_mib=128 key_size_bytes=100 key_layout=RandomReuse(1023) write_delta=Y...
-                        time:   [1.1906 s 1.1926 s 1.1947 s]
-                        thrpt:  [107.14 MiB/s 107.33 MiB/s 107.51 MiB/s]
-ingest/io_mode=DirectRw volume_mib=128 key_size_bytes=100 key_layout=Sequential write_delta=No
-                        time:   [516.86 ms 518.25 ms 519.47 ms]
-                        thrpt:  [246.40 MiB/s 246.98 MiB/s 247.65 MiB/s]
-ingest/io_mode=DirectRw volume_mib=128 key_size_bytes=8192 key_layout=Sequential write_delta=Yes
-                        time:   [536.50 ms 536.53 ms 536.60 ms]
-                        thrpt:  [238.54 MiB/s 238.57 MiB/s 238.59 MiB/s]
-ingest/io_mode=DirectRw volume_mib=128 key_size_bytes=8192 key_layout=Sequential write_delta=No
-                        time:   [267.77 ms 267.90 ms 268.04 ms]
-                        thrpt:  [477.53 MiB/s 477.79 MiB/s 478.02 MiB/s]
-
-Hetzner AX102:
-
-ingest/io_mode=Buffered volume_mib=128 key_size_bytes=100 key_layout=Sequential write_delta=Yes
-                        time:   [836.58 ms 861.93 ms 886.57 ms]
-                        thrpt:  [144.38 MiB/s 148.50 MiB/s 153.00 MiB/s]
-ingest/io_mode=Buffered volume_mib=128 key_size_bytes=100 key_layout=Random write_delta=Yes
-                        time:   [1.2782 s 1.3191 s 1.3665 s]
-                        thrpt:  [93.668 MiB/s 97.037 MiB/s 100.14 MiB/s]
-ingest/io_mode=Buffered volume_mib=128 key_size_bytes=100 key_layout=RandomReuse(1023) write_delta=Y...
-                        time:   [791.27 ms 807.08 ms 822.95 ms]
-                        thrpt:  [155.54 MiB/s 158.60 MiB/s 161.77 MiB/s]
-ingest/io_mode=Buffered volume_mib=128 key_size_bytes=100 key_layout=Sequential write_delta=No
-                        time:   [310.78 ms 314.66 ms 318.47 ms]
-                        thrpt:  [401.92 MiB/s 406.79 MiB/s 411.87 MiB/s]
-ingest/io_mode=Buffered volume_mib=128 key_size_bytes=8192 key_layout=Sequential write_delta=Yes
-                        time:   [377.11 ms 387.77 ms 399.21 ms]
-                        thrpt:  [320.63 MiB/s 330.10 MiB/s 339.42 MiB/s]
-ingest/io_mode=Buffered volume_mib=128 key_size_bytes=8192 key_layout=Sequential write_delta=No
-                        time:   [128.37 ms 132.96 ms 138.55 ms]
-                        thrpt:  [923.83 MiB/s 962.69 MiB/s 997.11 MiB/s]
-ingest/io_mode=Direct volume_mib=128 key_size_bytes=100 key_layout=Sequential write_delta=Yes
-                        time:   [900.38 ms 914.88 ms 928.86 ms]
-                        thrpt:  [137.80 MiB/s 139.91 MiB/s 142.16 MiB/s]
-ingest/io_mode=Direct volume_mib=128 key_size_bytes=100 key_layout=Random write_delta=Yes
-                        time:   [1.2538 s 1.2936 s 1.3313 s]
-                        thrpt:  [96.149 MiB/s 98.946 MiB/s 102.09 MiB/s]
-ingest/io_mode=Direct volume_mib=128 key_size_bytes=100 key_layout=RandomReuse(1023) write_delta=Yes
-                        time:   [787.17 ms 803.89 ms 820.63 ms]
-                        thrpt:  [155.98 MiB/s 159.23 MiB/s 162.61 MiB/s]
-ingest/io_mode=Direct volume_mib=128 key_size_bytes=100 key_layout=Sequential write_delta=No
-                        time:   [318.78 ms 321.89 ms 324.74 ms]
-                        thrpt:  [394.16 MiB/s 397.65 MiB/s 401.53 MiB/s]
-ingest/io_mode=Direct volume_mib=128 key_size_bytes=8192 key_layout=Sequential write_delta=Yes
-                        time:   [374.01 ms 383.45 ms 393.20 ms]
-                        thrpt:  [325.53 MiB/s 333.81 MiB/s 342.24 MiB/s]
-ingest/io_mode=Direct volume_mib=128 key_size_bytes=8192 key_layout=Sequential write_delta=No
-                        time:   [137.98 ms 141.31 ms 143.57 ms]
-                        thrpt:  [891.58 MiB/s 905.79 MiB/s 927.66 MiB/s]
-ingest/io_mode=DirectRw volume_mib=128 key_size_bytes=100 key_layout=Sequential write_delta=Yes
-                        time:   [613.69 ms 622.48 ms 630.97 ms]
-                        thrpt:  [202.86 MiB/s 205.63 MiB/s 208.57 MiB/s]
-ingest/io_mode=DirectRw volume_mib=128 key_size_bytes=100 key_layout=Random write_delta=Yes
-                        time:   [1.0299 s 1.0766 s 1.1273 s]
-                        thrpt:  [113.55 MiB/s 118.90 MiB/s 124.29 MiB/s]
-ingest/io_mode=DirectRw volume_mib=128 key_size_bytes=100 key_layout=RandomReuse(1023) write_delta=Y...
-                        time:   [637.80 ms 647.78 ms 658.01 ms]
-                        thrpt:  [194.53 MiB/s 197.60 MiB/s 200.69 MiB/s]
-ingest/io_mode=DirectRw volume_mib=128 key_size_bytes=100 key_layout=Sequential write_delta=No
-                        time:   [266.09 ms 267.20 ms 268.31 ms]
-                        thrpt:  [477.06 MiB/s 479.04 MiB/s 481.04 MiB/s]
-ingest/io_mode=DirectRw volume_mib=128 key_size_bytes=8192 key_layout=Sequential write_delta=Yes
-                        time:   [269.34 ms 273.27 ms 277.69 ms]
-                        thrpt:  [460.95 MiB/s 468.40 MiB/s 475.24 MiB/s]
-ingest/io_mode=DirectRw volume_mib=128 key_size_bytes=8192 key_layout=Sequential write_delta=No
-                        time:   [123.18 ms 124.24 ms 125.15 ms]
-                        thrpt:  [1022.8 MiB/s 1.0061 GiB/s 1.0148 GiB/s]
-*/

pageserver/client/src/mgmt_api.rs

@@ -419,23 +419,6 @@ impl Client {
         }
     }
 
-    pub async fn timeline_detail(
-        &self,
-        tenant_shard_id: TenantShardId,
-        timeline_id: TimelineId,
-    ) -> Result<TimelineInfo> {
-        let uri = format!(
-            "{}/v1/tenant/{tenant_shard_id}/timeline/{timeline_id}",
-            self.mgmt_api_endpoint
-        );
-
-        self.request(Method::GET, &uri, ())
-            .await?
-            .json()
-            .await
-            .map_err(Error::ReceiveBody)
-    }
-
     pub async fn timeline_archival_config(
         &self,
         tenant_shard_id: TenantShardId,

pageserver/src/config.rs

@@ -225,11 +225,6 @@ pub struct PageServerConf {
     /// Does not force TLS: the client negotiates TLS usage during the handshake.
     /// Uses key and certificate from ssl_key_file/ssl_cert_file.
     pub enable_tls_page_service_api: bool,
-
-    /// Run in development mode, which disables certain safety checks
-    /// such as authentication requirements for HTTP and PostgreSQL APIs.
-    /// This is insecure and should only be used in development environments.
-    pub dev_mode: bool,
 }
 
 /// Token for authentication to safekeepers
@@ -403,7 +398,6 @@ impl PageServerConf {
             generate_unarchival_heatmap,
             tracing,
             enable_tls_page_service_api,
-            dev_mode,
         } = config_toml;
 
         let mut conf = PageServerConf {
@@ -455,7 +449,6 @@ impl PageServerConf {
             get_vectored_concurrent_io,
             tracing,
             enable_tls_page_service_api,
-            dev_mode,
 
             // ------------------------------------------------------------
             // fields that require additional validation or custom handling

pageserver/src/consumption_metrics/metrics.rs

@@ -27,9 +27,6 @@ pub(super) enum Name {
     /// Timeline logical size
     #[serde(rename = "timeline_logical_size")]
     LogicalSize,
-    /// Timeline delta from parent (WAL bytes clamped to logical size)
-    #[serde(rename = "timeline_changed_bytes_from_parent")]
-    ChangedBytesFromParent,
     /// Tenant remote size
     #[serde(rename = "remote_storage_size")]
     RemoteSize,
@@ -178,24 +175,6 @@ impl MetricsKey {
         .absolute_values()
     }
 
-    /// [`Timeline::get_last_record_lsn`] - [`Timeline::get_ancestor_lsn`], clamped to
-    /// [`Timeline::get_current_logical_size`].
-    ///
-    /// [`Timeline::get_last_record_lsn`]: crate::tenant::Timeline::get_last_record_lsn
-    /// [`Timeline::get_ancestor_lsn`]: crate::tenant::Timeline::get_ancestor_lsn
-    /// [`Timeline::get_current_logical_size`]: crate::tenant::Timeline::get_current_logical_size
-    const fn timeline_changed_bytes_from_parent(
-        tenant_id: TenantId,
-        timeline_id: TimelineId,
-    ) -> AbsoluteValueFactory {
-        MetricsKey {
-            tenant_id,
-            timeline_id: Some(timeline_id),
-            metric: Name::ChangedBytesFromParent,
-        }
-        .absolute_values()
-    }
-
     /// [`TenantShard::remote_size`]
     ///
     /// [`TenantShard::remote_size`]: crate::tenant::TenantShard::remote_size
@@ -392,7 +371,6 @@ struct TimelineSnapshot {
     loaded_at: (Lsn, SystemTime),
     last_record_lsn: Lsn,
     current_exact_logical_size: Option<u64>,
-    changed_bytes_from_parent: Option<u64>,
 }
 
 impl TimelineSnapshot {
@@ -428,22 +406,10 @@ impl TimelineSnapshot {
                 }
             };
 
-            // This is an approximation of how much data has changed on this branch vs. its
-            // ancestor: the number of bytes written to the WAL, clamped to the size of the branch.
-            let changed_bytes_from_parent = current_exact_logical_size.and_then(|size| {
-                if t.get_ancestor_lsn() == Lsn::MAX {
-                    return None;
-                }
-                t.get_last_record_lsn()
-                    .checked_sub(t.get_ancestor_lsn())
-                    .map(|wal_bytes| wal_bytes.0.min(size))
-            });
-
             Ok(Some(TimelineSnapshot {
                 loaded_at,
                 last_record_lsn,
                 current_exact_logical_size,
-                changed_bytes_from_parent,
             }))
         }
     }
@@ -521,17 +487,6 @@ impl TimelineSnapshot {
                 metrics.push(factory.at(now, size));
             }
         }
-
-        {
-            let factory = MetricsKey::timeline_changed_bytes_from_parent(tenant_id, timeline_id);
-            let current_or_previous = self
-                .changed_bytes_from_parent
-                .or_else(|| cache.get(factory.key()).map(|item| item.value));
-
-            if let Some(size) = current_or_previous {
-                metrics.push(factory.at(now, size));
-            }
-        }
     }
 }
 

pageserver/src/consumption_metrics/metrics/tests.rs

@@ -18,7 +18,6 @@ fn startup_collected_timeline_metrics_before_advancing() {
         loaded_at: (disk_consistent_lsn, SystemTime::now()),
         last_record_lsn: disk_consistent_lsn,
         current_exact_logical_size: Some(0x42000),
-        changed_bytes_from_parent: Some(0x1000),
     };
 
     let now = DateTime::<Utc>::from(SystemTime::now());
@@ -34,8 +33,7 @@ fn startup_collected_timeline_metrics_before_advancing() {
                 0
             ),
             MetricsKey::written_size(tenant_id, timeline_id).at(now, disk_consistent_lsn.0),
-            MetricsKey::timeline_logical_size(tenant_id, timeline_id).at(now, 0x42000),
-            MetricsKey::timeline_changed_bytes_from_parent(tenant_id, timeline_id).at(now, 0x1000)
+            MetricsKey::timeline_logical_size(tenant_id, timeline_id).at(now, 0x42000)
         ]
     );
 }
@@ -62,7 +60,6 @@ fn startup_collected_timeline_metrics_second_round() {
         loaded_at: (disk_consistent_lsn, init),
         last_record_lsn: disk_consistent_lsn,
         current_exact_logical_size: Some(0x42000),
-        changed_bytes_from_parent: Some(0x1000),
     };
 
     snap.to_metrics(tenant_id, timeline_id, now, &mut metrics, &cache);
@@ -72,8 +69,7 @@ fn startup_collected_timeline_metrics_second_round() {
         &[
             MetricsKey::written_size_delta(tenant_id, timeline_id).from_until(before, now, 0),
             MetricsKey::written_size(tenant_id, timeline_id).at(now, disk_consistent_lsn.0),
-            MetricsKey::timeline_logical_size(tenant_id, timeline_id).at(now, 0x42000),
-            MetricsKey::timeline_changed_bytes_from_parent(tenant_id, timeline_id).at(now, 0x1000)
+            MetricsKey::timeline_logical_size(tenant_id, timeline_id).at(now, 0x42000)
         ]
     );
 }
@@ -108,7 +104,6 @@ fn startup_collected_timeline_metrics_nth_round_at_same_lsn() {
         loaded_at: (disk_consistent_lsn, init),
         last_record_lsn: disk_consistent_lsn,
         current_exact_logical_size: Some(0x42000),
-        changed_bytes_from_parent: Some(0x1000),
     };
 
     snap.to_metrics(tenant_id, timeline_id, now, &mut metrics, &cache);
@@ -118,8 +113,7 @@ fn startup_collected_timeline_metrics_nth_round_at_same_lsn() {
         &[
             MetricsKey::written_size_delta(tenant_id, timeline_id).from_until(just_before, now, 0),
             MetricsKey::written_size(tenant_id, timeline_id).at(now, disk_consistent_lsn.0),
-            MetricsKey::timeline_logical_size(tenant_id, timeline_id).at(now, 0x42000),
-            MetricsKey::timeline_changed_bytes_from_parent(tenant_id, timeline_id).at(now, 0x1000)
+            MetricsKey::timeline_logical_size(tenant_id, timeline_id).at(now, 0x42000)
         ]
     );
 }
@@ -147,7 +141,6 @@ fn post_restart_written_sizes_with_rolled_back_last_record_lsn() {
         loaded_at: (Lsn(50), at_restart),
         last_record_lsn: Lsn(50),
         current_exact_logical_size: None,
-        changed_bytes_from_parent: None,
     };
 
     let mut cache = HashMap::from([
@@ -209,7 +202,6 @@ fn post_restart_current_exact_logical_size_uses_cached() {
         loaded_at: (Lsn(50), at_restart),
         last_record_lsn: Lsn(50),
         current_exact_logical_size: None,
-        changed_bytes_from_parent: Some(0x1000),
     };
 
     let cache = HashMap::from([MetricsKey::timeline_logical_size(tenant_id, timeline_id)

pageserver/src/controller_upcall_client.rs

@@ -3,11 +3,10 @@ use std::collections::HashMap;
 use futures::Future;
 use pageserver_api::config::NodeMetadata;
 use pageserver_api::controller_api::{AvailabilityZone, NodeRegisterRequest};
-use pageserver_api::models::ShardImportStatus;
 use pageserver_api::shard::TenantShardId;
 use pageserver_api::upcall_api::{
-    PutTimelineImportStatusRequest, ReAttachRequest, ReAttachResponse, ReAttachResponseTenant,
-    ValidateRequest, ValidateRequestTenant, ValidateResponse,
+    ReAttachRequest, ReAttachResponse, ReAttachResponseTenant, ValidateRequest,
+    ValidateRequestTenant, ValidateResponse,
 };
 use reqwest::Certificate;
 use serde::Serialize;
@@ -15,7 +14,7 @@ use serde::de::DeserializeOwned;
 use tokio_util::sync::CancellationToken;
 use url::Url;
 use utils::generation::Generation;
-use utils::id::{NodeId, TimelineId};
+use utils::id::NodeId;
 use utils::{backoff, failpoint_support};
 
 use crate::config::PageServerConf;
@@ -47,12 +46,6 @@ pub trait StorageControllerUpcallApi {
         &self,
         tenants: Vec<(TenantShardId, Generation)>,
     ) -> impl Future<Output = Result<HashMap<TenantShardId, bool>, RetryForeverError>> + Send;
-    fn put_timeline_import_status(
-        &self,
-        tenant_shard_id: TenantShardId,
-        timeline_id: TimelineId,
-        status: ShardImportStatus,
-    ) -> impl Future<Output = Result<(), RetryForeverError>> + Send;
 }
 
 impl StorageControllerUpcallClient {
@@ -280,30 +273,4 @@ impl StorageControllerUpcallApi for StorageControllerUpcallClient {
 
         Ok(result.into_iter().collect())
     }
-
-    /// Send a shard import status to the storage controller
-    ///
-    /// The implementation must have at-least-once delivery semantics.
-    /// To this end, we retry the request until it succeeds. If the pageserver
-    /// restarts or crashes, the shard import will start again from the beggining.
-    #[tracing::instrument(skip_all)] // so that warning logs from retry_http_forever have context
-    async fn put_timeline_import_status(
-        &self,
-        tenant_shard_id: TenantShardId,
-        timeline_id: TimelineId,
-        status: ShardImportStatus,
-    ) -> Result<(), RetryForeverError> {
-        let url = self
-            .base_url
-            .join("timeline_import_status")
-            .expect("Failed to build path");
-
-        let request = PutTimelineImportStatusRequest {
-            tenant_shard_id,
-            timeline_id,
-            status,
-        };
-
-        self.retry_http_forever(&url, request).await
-    }
 }

pageserver/src/deletion_queue.rs

@@ -787,15 +787,6 @@ mod test {
 
             Ok(result)
         }
-
-        async fn put_timeline_import_status(
-            &self,
-            _tenant_shard_id: TenantShardId,
-            _timeline_id: TimelineId,
-            _status: pageserver_api::models::ShardImportStatus,
-        ) -> Result<(), RetryForeverError> {
-            unimplemented!()
-        }
     }
 
     async fn setup(test_name: &str) -> anyhow::Result<TestSetup> {

pageserver/src/metrics.rs

@@ -1289,7 +1289,6 @@ pub(crate) enum StorageIoOperation {
     Seek,
     Fsync,
     Metadata,
-    SetLen,
 }
 
 impl StorageIoOperation {
@@ -1304,7 +1303,6 @@ impl StorageIoOperation {
             StorageIoOperation::Seek => "seek",
             StorageIoOperation::Fsync => "fsync",
             StorageIoOperation::Metadata => "metadata",
-            StorageIoOperation::SetLen => "set_len",
         }
     }
 }

pageserver/src/tenant.rs

@@ -3816,24 +3816,6 @@ impl TenantShard {
                 MaybeDeletedIndexPart::IndexPart(p) => p,
             };
 
-            // A shard split may not take place while a timeline import is on-going
-            // for the tenant. Timeline imports run as part of each tenant shard
-            // and rely on the sharding scheme to split the work among pageservers.
-            // If we were to split in the middle of this process, we would have to
-            // either ensure that it's driven to completion on the old shard set
-            // or transfer it to the new shard set. It's technically possible, but complex.
-            match index_part.import_pgdata {
-                Some(ref import) if !import.is_done() => {
-                    anyhow::bail!(
-                        "Cannot split due to import with idempotency key: {:?}",
-                        import.idempotency_key()
-                    );
-                }
-                Some(_) | None => {
-                    // fallthrough
-                }
-            }
-
             for child_shard in child_shards {
                 tracing::info!(%timeline_id, "Uploading index_part for child {}", child_shard.to_index());
                 upload_index_part(

pageserver/src/tenant/blob_io.rs

@@ -15,23 +15,21 @@
 //! len >= 128: 1CCCXXXX XXXXXXXX XXXXXXXX XXXXXXXX
 //!
 use std::cmp::min;
+use std::io::Error;
 
-use anyhow::Context;
 use async_compression::Level;
 use bytes::{BufMut, BytesMut};
 use pageserver_api::models::ImageCompressionAlgorithm;
 use tokio::io::AsyncWriteExt;
-use tokio_epoll_uring::IoBuf;
+use tokio_epoll_uring::{BoundedBuf, IoBuf, Slice};
 use tokio_util::sync::CancellationToken;
 use tracing::warn;
 
 use crate::context::RequestContext;
 use crate::page_cache::PAGE_SZ;
 use crate::tenant::block_io::BlockCursor;
-use crate::virtual_file::IoBufferMut;
+use crate::virtual_file::VirtualFile;
 use crate::virtual_file::owned_buffers_io::io_buf_ext::{FullSlice, IoBufExt};
-use crate::virtual_file::owned_buffers_io::write::{BufferedWriter, FlushTaskError};
-use crate::virtual_file::owned_buffers_io::write::{BufferedWriterShutdownMode, OwnedAsyncWriter};
 
 #[derive(Copy, Clone, Debug)]
 pub struct CompressionInfo {
@@ -52,9 +50,12 @@ pub struct Header {
 
 impl Header {
     /// Decodes a header from a byte slice.
-    pub fn decode(bytes: &[u8]) -> anyhow::Result<Self> {
+    pub fn decode(bytes: &[u8]) -> Result<Self, std::io::Error> {
         let Some(&first_header_byte) = bytes.first() else {
-            anyhow::bail!("zero-length blob header");
+            return Err(std::io::Error::new(
+                std::io::ErrorKind::InvalidData,
+                "zero-length blob header",
+            ));
         };
 
         // If the first bit is 0, this is just a 1-byte length prefix up to 128 bytes.
@@ -68,9 +69,12 @@ impl Header {
 
         // Otherwise, this is a 4-byte header containing compression information and length.
         const HEADER_LEN: usize = 4;
-        let mut header_buf: [u8; HEADER_LEN] = bytes[0..HEADER_LEN]
-            .try_into()
-            .map_err(|_| anyhow::anyhow!("blob header too short: {bytes:?}"))?;
+        let mut header_buf: [u8; HEADER_LEN] = bytes[0..HEADER_LEN].try_into().map_err(|_| {
+            std::io::Error::new(
+                std::io::ErrorKind::InvalidData,
+                format!("blob header too short: {bytes:?}"),
+            )
+        })?;
 
         // TODO: verify the compression bits and convert to an enum.
         let compression_bits = header_buf[0] & LEN_COMPRESSION_BIT_MASK;
@@ -90,16 +94,6 @@ impl Header {
     }
 }
 
-#[derive(Debug, thiserror::Error)]
-pub enum WriteBlobError {
-    #[error(transparent)]
-    Flush(FlushTaskError),
-    #[error("blob too large ({len} bytes)")]
-    BlobTooLarge { len: usize },
-    #[error(transparent)]
-    WriteBlobRaw(anyhow::Error),
-}
-
 impl BlockCursor<'_> {
     /// Read a blob into a new buffer.
     pub async fn read_blob(
@@ -219,64 +213,143 @@ pub(super) const BYTE_UNCOMPRESSED: u8 = 0x80;
 pub(super) const BYTE_ZSTD: u8 = BYTE_UNCOMPRESSED | 0x10;
 
 /// A wrapper of `VirtualFile` that allows users to write blobs.
-pub struct BlobWriter<W> {
+///
+/// If a `BlobWriter` is dropped, the internal buffer will be
+/// discarded. You need to call [`flush_buffer`](Self::flush_buffer)
+/// manually before dropping.
+pub struct BlobWriter<const BUFFERED: bool> {
+    inner: VirtualFile,
+    offset: u64,
+    /// A buffer to save on write calls, only used if BUFFERED=true
+    buf: Vec<u8>,
     /// We do tiny writes for the length headers; they need to be in an owned buffer;
     io_buf: Option<BytesMut>,
-    writer: BufferedWriter<IoBufferMut, W>,
-    offset: u64,
 }
 
-impl<W> BlobWriter<W>
-where
-    W: OwnedAsyncWriter + std::fmt::Debug + Send + Sync + 'static,
-{
-    /// See [`BufferedWriter`] struct-level doc comment for semantics of `start_offset`.
+impl<const BUFFERED: bool> BlobWriter<BUFFERED> {
     pub fn new(
-        file: W,
+        inner: VirtualFile,
         start_offset: u64,
-        gate: &utils::sync::gate::Gate,
-        cancel: CancellationToken,
-        ctx: &RequestContext,
-        flush_task_span: tracing::Span,
-    ) -> anyhow::Result<Self> {
-        Ok(Self {
-            io_buf: Some(BytesMut::new()),
-            writer: BufferedWriter::new(
-                file,
-                start_offset,
-                || IoBufferMut::with_capacity(Self::CAPACITY),
-                gate.enter()?,
-                cancel,
-                ctx,
-                flush_task_span,
-            ),
+        _gate: &utils::sync::gate::Gate,
+        _cancel: CancellationToken,
+        _ctx: &RequestContext,
+    ) -> Self {
+        Self {
+            inner,
             offset: start_offset,
-        })
+            buf: Vec::with_capacity(Self::CAPACITY),
+            io_buf: Some(BytesMut::new()),
+        }
     }
 
     pub fn size(&self) -> u64 {
         self.offset
     }
 
-    const CAPACITY: usize = 64 * 1024;
+    const CAPACITY: usize = if BUFFERED { 64 * 1024 } else { 0 };
 
-    /// Writes `src_buf` to the file at the current offset.
+    /// Writes the given buffer directly to the underlying `VirtualFile`.
+    /// You need to make sure that the internal buffer is empty, otherwise
+    /// data will be written in wrong order.
+    #[inline(always)]
+    async fn write_all_unbuffered<Buf: IoBuf + Send>(
+        &mut self,
+        src_buf: FullSlice<Buf>,
+        ctx: &RequestContext,
+    ) -> (FullSlice<Buf>, Result<(), Error>) {
+        let (src_buf, res) = self.inner.write_all(src_buf, ctx).await;
+        let nbytes = match res {
+            Ok(nbytes) => nbytes,
+            Err(e) => return (src_buf, Err(e)),
+        };
+        self.offset += nbytes as u64;
+        (src_buf, Ok(()))
+    }
+
+    #[inline(always)]
+    /// Flushes the internal buffer to the underlying `VirtualFile`.
+    pub async fn flush_buffer(&mut self, ctx: &RequestContext) -> Result<(), Error> {
+        let buf = std::mem::take(&mut self.buf);
+        let (slice, res) = self.inner.write_all(buf.slice_len(), ctx).await;
+        res?;
+        let mut buf = slice.into_raw_slice().into_inner();
+        buf.clear();
+        self.buf = buf;
+        Ok(())
+    }
+
+    #[inline(always)]
+    /// Writes as much of `src_buf` into the internal buffer as it fits
+    fn write_into_buffer(&mut self, src_buf: &[u8]) -> usize {
+        let remaining = Self::CAPACITY - self.buf.len();
+        let to_copy = src_buf.len().min(remaining);
+        self.buf.extend_from_slice(&src_buf[..to_copy]);
+        self.offset += to_copy as u64;
+        to_copy
+    }
+
+    /// Internal, possibly buffered, write function
     async fn write_all<Buf: IoBuf + Send>(
         &mut self,
         src_buf: FullSlice<Buf>,
         ctx: &RequestContext,
-    ) -> (FullSlice<Buf>, Result<(), FlushTaskError>) {
-        let res = self
-            .writer
-            // TODO: why are we taking a FullSlice if we're going to pass a borrow downstack?
-            // Can remove all the complexity around owned buffers upstack
-            .write_buffered_borrowed(&src_buf, ctx)
-            .await
-            .map(|len| {
-                self.offset += len as u64;
-            });
+    ) -> (FullSlice<Buf>, Result<(), Error>) {
+        let src_buf = src_buf.into_raw_slice();
+        let src_buf_bounds = src_buf.bounds();
+        let restore = move |src_buf_slice: Slice<_>| {
+            FullSlice::must_new(Slice::from_buf_bounds(
+                src_buf_slice.into_inner(),
+                src_buf_bounds,
+            ))
+        };
 
-        (src_buf, res)
+        if !BUFFERED {
+            assert!(self.buf.is_empty());
+            return self
+                .write_all_unbuffered(FullSlice::must_new(src_buf), ctx)
+                .await;
+        }
+        let remaining = Self::CAPACITY - self.buf.len();
+        let src_buf_len = src_buf.bytes_init();
+        if src_buf_len == 0 {
+            return (restore(src_buf), Ok(()));
+        }
+        let mut src_buf = src_buf.slice(0..src_buf_len);
+        // First try to copy as much as we can into the buffer
+        if remaining > 0 {
+            let copied = self.write_into_buffer(&src_buf);
+            src_buf = src_buf.slice(copied..);
+        }
+        // Then, if the buffer is full, flush it out
+        if self.buf.len() == Self::CAPACITY {
+            if let Err(e) = self.flush_buffer(ctx).await {
+                return (restore(src_buf), Err(e));
+            }
+        }
+        // Finally, write the tail of src_buf:
+        // If it wholly fits into the buffer without
+        // completely filling it, then put it there.
+        // If not, write it out directly.
+        let src_buf = if !src_buf.is_empty() {
+            assert_eq!(self.buf.len(), 0);
+            if src_buf.len() < Self::CAPACITY {
+                let copied = self.write_into_buffer(&src_buf);
+                // We just verified above that src_buf fits into our internal buffer.
+                assert_eq!(copied, src_buf.len());
+                restore(src_buf)
+            } else {
+                let (src_buf, res) = self
+                    .write_all_unbuffered(FullSlice::must_new(src_buf), ctx)
+                    .await;
+                if let Err(e) = res {
+                    return (src_buf, Err(e));
+                }
+                src_buf
+            }
+        } else {
+            restore(src_buf)
+        };
+        (src_buf, Ok(()))
     }
 
     /// Write a blob of data. Returns the offset that it was written to,
@@ -285,7 +358,7 @@ where
         &mut self,
         srcbuf: FullSlice<Buf>,
         ctx: &RequestContext,
-    ) -> (FullSlice<Buf>, Result<u64, WriteBlobError>) {
+    ) -> (FullSlice<Buf>, Result<u64, Error>) {
         let (buf, res) = self
             .write_blob_maybe_compressed(srcbuf, ctx, ImageCompressionAlgorithm::Disabled)
             .await;
@@ -299,10 +372,7 @@ where
         srcbuf: FullSlice<Buf>,
         ctx: &RequestContext,
         algorithm: ImageCompressionAlgorithm,
-    ) -> (
-        FullSlice<Buf>,
-        Result<(u64, CompressionInfo), WriteBlobError>,
-    ) {
+    ) -> (FullSlice<Buf>, Result<(u64, CompressionInfo), Error>) {
         let offset = self.offset;
         let mut compression_info = CompressionInfo {
             written_compressed: false,
@@ -318,16 +388,14 @@ where
             if len < 128 {
                 // Short blob. Write a 1-byte length header
                 io_buf.put_u8(len as u8);
-                let (slice, res) = self.write_all(io_buf.slice_len(), ctx).await;
-                let res = res.map_err(WriteBlobError::Flush);
-                ((slice, res), srcbuf)
+                (self.write_all(io_buf.slice_len(), ctx).await, srcbuf)
             } else {
                 // Write a 4-byte length header
                 if len > MAX_SUPPORTED_BLOB_LEN {
                     return (
                         (
                             io_buf.slice_len(),
-                            Err(WriteBlobError::BlobTooLarge { len }),
+                            Err(Error::other(format!("blob too large ({len} bytes)"))),
                         ),
                         srcbuf,
                     );
@@ -361,9 +429,7 @@ where
                 assert_eq!(len_buf[0] & 0xf0, 0);
                 len_buf[0] |= high_bit_mask;
                 io_buf.extend_from_slice(&len_buf[..]);
-                let (slice, res) = self.write_all(io_buf.slice_len(), ctx).await;
-                let res = res.map_err(WriteBlobError::Flush);
-                ((slice, res), srcbuf)
+                (self.write_all(io_buf.slice_len(), ctx).await, srcbuf)
             }
         }
         .await;
@@ -378,7 +444,6 @@ where
         } else {
             self.write_all(srcbuf, ctx).await
         };
-        let res = res.map_err(WriteBlobError::Flush);
         (srcbuf, res.map(|_| (offset, compression_info)))
     }
 
@@ -387,12 +452,9 @@ where
         &mut self,
         raw_with_header: FullSlice<Buf>,
         ctx: &RequestContext,
-    ) -> (FullSlice<Buf>, Result<u64, WriteBlobError>) {
+    ) -> (FullSlice<Buf>, Result<u64, Error>) {
         // Verify the header, to ensure we don't write invalid/corrupt data.
-        let header = match Header::decode(&raw_with_header)
-            .context("decoding blob header")
-            .map_err(WriteBlobError::WriteBlobRaw)
-        {
+        let header = match Header::decode(&raw_with_header) {
             Ok(header) => header,
             Err(err) => return (raw_with_header, Err(err)),
         };
@@ -401,26 +463,42 @@ where
             let raw_len = raw_with_header.len();
             return (
                 raw_with_header,
-                Err(WriteBlobError::WriteBlobRaw(anyhow::anyhow!(
-                    "header length mismatch: {header_total_len} != {raw_len}"
-                ))),
+                Err(std::io::Error::new(
+                    std::io::ErrorKind::InvalidData,
+                    format!("header length mismatch: {header_total_len} != {raw_len}"),
+                )),
             );
         }
 
         let offset = self.offset;
         let (raw_with_header, result) = self.write_all(raw_with_header, ctx).await;
-        let result = result.map_err(WriteBlobError::Flush);
         (raw_with_header, result.map(|_| offset))
     }
+}
 
-    /// Finish this blob writer and return the underlying `W`.
-    pub async fn shutdown(
-        self,
-        mode: BufferedWriterShutdownMode,
-        ctx: &RequestContext,
-    ) -> Result<W, FlushTaskError> {
-        let (_, file) = self.writer.shutdown(mode, ctx).await?;
-        Ok(file)
+impl BlobWriter<true> {
+    /// Access the underlying `VirtualFile`.
+    ///
+    /// This function flushes the internal buffer before giving access
+    /// to the underlying `VirtualFile`.
+    pub async fn into_inner(mut self, ctx: &RequestContext) -> Result<VirtualFile, Error> {
+        self.flush_buffer(ctx).await?;
+        Ok(self.inner)
+    }
+
+    /// Access the underlying `VirtualFile`.
+    ///
+    /// Unlike [`into_inner`](Self::into_inner), this doesn't flush
+    /// the internal buffer before giving access.
+    pub fn into_inner_no_flush(self) -> VirtualFile {
+        self.inner
+    }
+}
+
+impl BlobWriter<false> {
+    /// Access the underlying `VirtualFile`.
+    pub fn into_inner(self) -> VirtualFile {
+        self.inner
     }
 }
 
@@ -429,25 +507,21 @@ pub(crate) mod tests {
     use camino::Utf8PathBuf;
     use camino_tempfile::Utf8TempDir;
     use rand::{Rng, SeedableRng};
-    use tracing::info_span;
 
     use super::*;
     use crate::context::DownloadBehavior;
     use crate::task_mgr::TaskKind;
     use crate::tenant::block_io::BlockReaderRef;
-    use crate::virtual_file;
-    use crate::virtual_file::TempVirtualFile;
-    use crate::virtual_file::VirtualFile;
 
-    async fn round_trip_test(blobs: &[Vec<u8>]) -> anyhow::Result<()> {
-        round_trip_test_compressed(blobs, false).await
+    async fn round_trip_test<const BUFFERED: bool>(blobs: &[Vec<u8>]) -> Result<(), Error> {
+        round_trip_test_compressed::<BUFFERED>(blobs, false).await
     }
 
-    pub(crate) async fn write_maybe_compressed(
+    pub(crate) async fn write_maybe_compressed<const BUFFERED: bool>(
         blobs: &[Vec<u8>],
         compression: bool,
         ctx: &RequestContext,
-    ) -> anyhow::Result<(Utf8TempDir, Utf8PathBuf, Vec<u64>)> {
+    ) -> Result<(Utf8TempDir, Utf8PathBuf, Vec<u64>), Error> {
         let temp_dir = camino_tempfile::tempdir()?;
         let pathbuf = temp_dir.path().join("file");
         let gate = utils::sync::gate::Gate::default();
@@ -456,19 +530,8 @@ pub(crate) mod tests {
         // Write part (in block to drop the file)
         let mut offsets = Vec::new();
         {
-            let file = TempVirtualFile::new(
-                VirtualFile::open_with_options_v2(
-                    pathbuf.as_path(),
-                    virtual_file::OpenOptions::new()
-                        .create_new(true)
-                        .write(true),
-                    ctx,
-                )
-                .await?,
-                gate.enter()?,
-            );
-            let mut wtr =
-                BlobWriter::new(file, 0, &gate, cancel.clone(), ctx, info_span!("test")).unwrap();
+            let file = VirtualFile::create(pathbuf.as_path(), ctx).await?;
+            let mut wtr = BlobWriter::<BUFFERED>::new(file, 0, &gate, cancel.clone(), ctx);
             for blob in blobs.iter() {
                 let (_, res) = if compression {
                     let res = wtr
@@ -485,28 +548,26 @@ pub(crate) mod tests {
                 let offs = res?;
                 offsets.push(offs);
             }
-            let file = wtr
-                .shutdown(
-                    BufferedWriterShutdownMode::ZeroPadToNextMultiple(PAGE_SZ),
-                    ctx,
-                )
-                .await?;
-            file.disarm_into_inner()
-        };
+            // Write out one page worth of zeros so that we can
+            // read again with read_blk
+            let (_, res) = wtr.write_blob(vec![0; PAGE_SZ].slice_len(), ctx).await;
+            let offs = res?;
+            println!("Writing final blob at offs={offs}");
+            wtr.flush_buffer(ctx).await?;
+        }
         Ok((temp_dir, pathbuf, offsets))
     }
 
-    async fn round_trip_test_compressed(
+    async fn round_trip_test_compressed<const BUFFERED: bool>(
         blobs: &[Vec<u8>],
         compression: bool,
-    ) -> anyhow::Result<()> {
+    ) -> Result<(), Error> {
         let ctx =
             RequestContext::new(TaskKind::UnitTest, DownloadBehavior::Error).with_scope_unit_test();
         let (_temp_dir, pathbuf, offsets) =
-            write_maybe_compressed(blobs, compression, &ctx).await?;
+            write_maybe_compressed::<BUFFERED>(blobs, compression, &ctx).await?;
 
-        println!("Done writing!");
-        let file = VirtualFile::open_v2(pathbuf, &ctx).await?;
+        let file = VirtualFile::open(pathbuf, &ctx).await?;
         let rdr = BlockReaderRef::VirtualFile(&file);
         let rdr = BlockCursor::new_with_compression(rdr, compression);
         for (idx, (blob, offset)) in blobs.iter().zip(offsets.iter()).enumerate() {
@@ -525,27 +586,30 @@ pub(crate) mod tests {
     }
 
     #[tokio::test]
-    async fn test_one() -> anyhow::Result<()> {
+    async fn test_one() -> Result<(), Error> {
         let blobs = &[vec![12, 21, 22]];
-        round_trip_test(blobs).await?;
+        round_trip_test::<false>(blobs).await?;
+        round_trip_test::<true>(blobs).await?;
         Ok(())
     }
 
     #[tokio::test]
-    async fn test_hello_simple() -> anyhow::Result<()> {
+    async fn test_hello_simple() -> Result<(), Error> {
         let blobs = &[
             vec![0, 1, 2, 3],
             b"Hello, World!".to_vec(),
             Vec::new(),
             b"foobar".to_vec(),
         ];
-        round_trip_test(blobs).await?;
-        round_trip_test_compressed(blobs, true).await?;
+        round_trip_test::<false>(blobs).await?;
+        round_trip_test::<true>(blobs).await?;
+        round_trip_test_compressed::<false>(blobs, true).await?;
+        round_trip_test_compressed::<true>(blobs, true).await?;
         Ok(())
     }
 
     #[tokio::test]
-    async fn test_really_big_array() -> anyhow::Result<()> {
+    async fn test_really_big_array() -> Result<(), Error> {
         let blobs = &[
             b"test".to_vec(),
             random_array(10 * PAGE_SZ),
@@ -554,22 +618,25 @@ pub(crate) mod tests {
             vec![0xf3; 24 * PAGE_SZ],
             b"foobar".to_vec(),
         ];
-        round_trip_test(blobs).await?;
-        round_trip_test_compressed(blobs, true).await?;
+        round_trip_test::<false>(blobs).await?;
+        round_trip_test::<true>(blobs).await?;
+        round_trip_test_compressed::<false>(blobs, true).await?;
+        round_trip_test_compressed::<true>(blobs, true).await?;
         Ok(())
     }
 
     #[tokio::test]
-    async fn test_arrays_inc() -> anyhow::Result<()> {
+    async fn test_arrays_inc() -> Result<(), Error> {
         let blobs = (0..PAGE_SZ / 8)
             .map(|v| random_array(v * 16))
             .collect::<Vec<_>>();
-        round_trip_test(&blobs).await?;
+        round_trip_test::<false>(&blobs).await?;
+        round_trip_test::<true>(&blobs).await?;
         Ok(())
     }
 
     #[tokio::test]
-    async fn test_arrays_random_size() -> anyhow::Result<()> {
+    async fn test_arrays_random_size() -> Result<(), Error> {
         let mut rng = rand::rngs::StdRng::seed_from_u64(42);
         let blobs = (0..1024)
             .map(|_| {
@@ -581,18 +648,20 @@ pub(crate) mod tests {
                 random_array(sz.into())
             })
             .collect::<Vec<_>>();
-        round_trip_test(&blobs).await?;
+        round_trip_test::<false>(&blobs).await?;
+        round_trip_test::<true>(&blobs).await?;
         Ok(())
     }
 
     #[tokio::test]
-    async fn test_arrays_page_boundary() -> anyhow::Result<()> {
+    async fn test_arrays_page_boundary() -> Result<(), Error> {
         let blobs = &[
             random_array(PAGE_SZ - 4),
             random_array(PAGE_SZ - 4),
             random_array(PAGE_SZ - 4),
         ];
-        round_trip_test(blobs).await?;
+        round_trip_test::<false>(blobs).await?;
+        round_trip_test::<true>(blobs).await?;
         Ok(())
     }
 }

pageserver/src/tenant/block_io.rs

@@ -4,12 +4,14 @@
 
 use std::ops::Deref;
 
+use bytes::Bytes;
+
 use super::storage_layer::delta_layer::{Adapter, DeltaLayerInner};
 use crate::context::RequestContext;
 use crate::page_cache::{self, FileId, PAGE_SZ, PageReadGuard, PageWriteGuard, ReadBufResult};
 #[cfg(test)]
 use crate::virtual_file::IoBufferMut;
-use crate::virtual_file::{IoBuffer, VirtualFile};
+use crate::virtual_file::VirtualFile;
 
 /// This is implemented by anything that can read 8 kB (PAGE_SZ)
 /// blocks, using the page cache
@@ -245,17 +247,17 @@ pub trait BlockWriter {
     /// 'buf' must be of size PAGE_SZ. Returns the block number the page was
     /// written to.
     ///
-    fn write_blk(&mut self, buf: IoBuffer) -> Result<u32, std::io::Error>;
+    fn write_blk(&mut self, buf: Bytes) -> Result<u32, std::io::Error>;
 }
 
 ///
 /// A simple in-memory buffer of blocks.
 ///
 pub struct BlockBuf {
-    pub blocks: Vec<IoBuffer>,
+    pub blocks: Vec<Bytes>,
 }
 impl BlockWriter for BlockBuf {
-    fn write_blk(&mut self, buf: IoBuffer) -> Result<u32, std::io::Error> {
+    fn write_blk(&mut self, buf: Bytes) -> Result<u32, std::io::Error> {
         assert!(buf.len() == PAGE_SZ);
         let blknum = self.blocks.len();
         self.blocks.push(buf);

pageserver/src/tenant/disk_btree.rs

@@ -25,7 +25,7 @@ use std::{io, result};
 
 use async_stream::try_stream;
 use byteorder::{BE, ReadBytesExt};
-use bytes::BufMut;
+use bytes::{BufMut, Bytes, BytesMut};
 use either::Either;
 use futures::{Stream, StreamExt};
 use hex;
@@ -34,7 +34,6 @@ use tracing::error;
 
 use crate::context::RequestContext;
 use crate::tenant::block_io::{BlockReader, BlockWriter};
-use crate::virtual_file::{IoBuffer, IoBufferMut, owned_buffers_io::write::Buffer};
 
 // The maximum size of a value stored in the B-tree. 5 bytes is enough currently.
 pub const VALUE_SZ: usize = 5;
@@ -788,12 +787,12 @@ impl<const L: usize> BuildNode<L> {
     ///
     /// Serialize the node to on-disk format.
     ///
-    fn pack(&self) -> IoBuffer {
+    fn pack(&self) -> Bytes {
         assert!(self.keys.len() == self.num_children as usize * self.suffix_len);
         assert!(self.values.len() == self.num_children as usize * VALUE_SZ);
         assert!(self.num_children > 0);
 
-        let mut buf = IoBufferMut::with_capacity(PAGE_SZ);
+        let mut buf = BytesMut::new();
 
         buf.put_u16(self.num_children);
         buf.put_u8(self.level);
@@ -806,7 +805,7 @@ impl<const L: usize> BuildNode<L> {
         assert!(buf.len() == self.size);
 
         assert!(buf.len() <= PAGE_SZ);
-        buf.extend_with(0, PAGE_SZ - buf.len());
+        buf.resize(PAGE_SZ, 0);
         buf.freeze()
     }
 
@@ -840,7 +839,7 @@ pub(crate) mod tests {
 
     #[derive(Clone, Default)]
     pub(crate) struct TestDisk {
-        blocks: Vec<IoBuffer>,
+        blocks: Vec<Bytes>,
     }
     impl TestDisk {
         fn new() -> Self {
@@ -858,7 +857,7 @@ pub(crate) mod tests {
         }
     }
     impl BlockWriter for &mut TestDisk {
-        fn write_blk(&mut self, buf: IoBuffer) -> io::Result<u32> {
+        fn write_blk(&mut self, buf: Bytes) -> io::Result<u32> {
             let blknum = self.blocks.len();
             self.blocks.push(buf);
             Ok(blknum as u32)

pageserver/src/tenant/ephemeral_file.rs

@@ -12,7 +12,6 @@ use tokio_epoll_uring::{BoundedBuf, Slice};
 use tokio_util::sync::CancellationToken;
 use tracing::{error, info_span};
 use utils::id::TimelineId;
-use utils::sync::gate::GateGuard;
 
 use crate::assert_u64_eq_usize::{U64IsUsize, UsizeIsU64};
 use crate::config::PageServerConf;
@@ -22,33 +21,16 @@ use crate::tenant::storage_layer::inmemory_layer::vectored_dio_read::File;
 use crate::virtual_file::owned_buffers_io::io_buf_aligned::IoBufAlignedMut;
 use crate::virtual_file::owned_buffers_io::slice::SliceMutExt;
 use crate::virtual_file::owned_buffers_io::write::{Buffer, FlushTaskError};
-use crate::virtual_file::{self, IoBufferMut, TempVirtualFile, VirtualFile, owned_buffers_io};
-
-use self::owned_buffers_io::write::OwnedAsyncWriter;
+use crate::virtual_file::{self, IoBufferMut, VirtualFile, owned_buffers_io};
 
 pub struct EphemeralFile {
     _tenant_shard_id: TenantShardId,
     _timeline_id: TimelineId,
     page_cache_file_id: page_cache::FileId,
     bytes_written: u64,
-    file: TempVirtualFileCoOwnedByEphemeralFileAndBufferedWriter,
-    buffered_writer: BufferedWriter,
-}
-
-type BufferedWriter = owned_buffers_io::write::BufferedWriter<
-    IoBufferMut,
-    TempVirtualFileCoOwnedByEphemeralFileAndBufferedWriter,
->;
-
-/// A TempVirtualFile that is co-owned by the [`EphemeralFile`]` and [`BufferedWriter`].
-///
-/// (Actually [`BufferedWriter`] internally is just a client to a background flush task.
-/// The co-ownership is between [`EphemeralFile`] and that flush task.)
-///
-/// Co-ownership allows us to serve reads for data that has already been flushed by the [`BufferedWriter`].
-#[derive(Debug, Clone)]
-struct TempVirtualFileCoOwnedByEphemeralFileAndBufferedWriter {
-    inner: Arc<TempVirtualFile>,
+    buffered_writer: owned_buffers_io::write::BufferedWriter<IoBufferMut, VirtualFile>,
+    /// Gate guard is held on as long as we need to do operations in the path (delete on drop)
+    _gate_guard: utils::sync::gate::GateGuard,
 }
 
 const TAIL_SZ: usize = 64 * 1024;
@@ -62,12 +44,9 @@ impl EphemeralFile {
         cancel: &CancellationToken,
         ctx: &RequestContext,
     ) -> anyhow::Result<EphemeralFile> {
-        // TempVirtualFile requires us to never reuse a filename while an old
-        // instance of TempVirtualFile created with that filename is not done dropping yet.
-        // So, we use a monotonic counter to disambiguate the filenames.
-        static NEXT_TEMP_DISAMBIGUATOR: AtomicU64 = AtomicU64::new(1);
+        static NEXT_FILENAME: AtomicU64 = AtomicU64::new(1);
         let filename_disambiguator =
-            NEXT_TEMP_DISAMBIGUATOR.fetch_add(1, std::sync::atomic::Ordering::Relaxed);
+            NEXT_FILENAME.fetch_add(1, std::sync::atomic::Ordering::Relaxed);
 
         let filename = conf
             .timeline_path(&tenant_shard_id, &timeline_id)
@@ -75,17 +54,16 @@ impl EphemeralFile {
                 "ephemeral-{filename_disambiguator}"
             )));
 
-        let file = TempVirtualFileCoOwnedByEphemeralFileAndBufferedWriter::new(
+        let file = Arc::new(
             VirtualFile::open_with_options_v2(
                 &filename,
                 virtual_file::OpenOptions::new()
-                    .create_new(true)
                     .read(true)
-                    .write(true),
+                    .write(true)
+                    .create(true),
                 ctx,
             )
             .await?,
-            gate.enter()?,
         );
 
         let page_cache_file_id = page_cache::next_file_id(); // XXX get rid, we're not page-caching anymore
@@ -95,60 +73,37 @@ impl EphemeralFile {
             _timeline_id: timeline_id,
             page_cache_file_id,
             bytes_written: 0,
-            file: file.clone(),
-            buffered_writer: BufferedWriter::new(
+            buffered_writer: owned_buffers_io::write::BufferedWriter::new(
                 file,
-                0,
                 || IoBufferMut::with_capacity(TAIL_SZ),
                 gate.enter()?,
                 cancel.child_token(),
                 ctx,
                 info_span!(parent: None, "ephemeral_file_buffered_writer", tenant_id=%tenant_shard_id.tenant_id, shard_id=%tenant_shard_id.shard_slug(), timeline_id=%timeline_id, path = %filename),
             ),
+            _gate_guard: gate.enter()?,
         })
     }
 }
 
-impl TempVirtualFileCoOwnedByEphemeralFileAndBufferedWriter {
-    fn new(file: VirtualFile, gate_guard: GateGuard) -> Self {
-        Self {
-            inner: Arc::new(TempVirtualFile::new(file, gate_guard)),
+impl Drop for EphemeralFile {
+    fn drop(&mut self) {
+        // unlink the file
+        // we are clear to do this, because we have entered a gate
+        let path = self.buffered_writer.as_inner().path();
+        let res = std::fs::remove_file(path);
+        if let Err(e) = res {
+            if e.kind() != std::io::ErrorKind::NotFound {
+                // just never log the not found errors, we cannot do anything for them; on detach
+                // the tenant directory is already gone.
+                //
+                // not found files might also be related to https://github.com/neondatabase/neon/issues/2442
+                error!("could not remove ephemeral file '{path}': {e}");
+            }
         }
     }
 }
 
-impl OwnedAsyncWriter for TempVirtualFileCoOwnedByEphemeralFileAndBufferedWriter {
-    fn write_all_at<Buf: owned_buffers_io::io_buf_aligned::IoBufAligned + Send>(
-        &self,
-        buf: owned_buffers_io::io_buf_ext::FullSlice<Buf>,
-        offset: u64,
-        ctx: &RequestContext,
-    ) -> impl std::future::Future<
-        Output = (
-            owned_buffers_io::io_buf_ext::FullSlice<Buf>,
-            std::io::Result<()>,
-        ),
-    > + Send {
-        self.inner.write_all_at(buf, offset, ctx)
-    }
-
-    fn set_len(
-        &self,
-        len: u64,
-        ctx: &RequestContext,
-    ) -> impl Future<Output = std::io::Result<()>> + Send {
-        self.inner.set_len(len, ctx)
-    }
-}
-
-impl std::ops::Deref for TempVirtualFileCoOwnedByEphemeralFileAndBufferedWriter {
-    type Target = VirtualFile;
-
-    fn deref(&self) -> &Self::Target {
-        &self.inner
-    }
-}
-
 #[derive(Debug, thiserror::Error)]
 pub(crate) enum EphemeralFileWriteError {
     #[error("{0}")]
@@ -307,9 +262,9 @@ impl super::storage_layer::inmemory_layer::vectored_dio_read::File for Ephemeral
         let mutable_range = Range(std::cmp::max(start, submitted_offset), end);
 
         let dst = if written_range.len() > 0 {
+            let file: &VirtualFile = self.buffered_writer.as_inner();
             let bounds = dst.bounds();
-            let slice = self
-                .file
+            let slice = file
                 .read_exact_at(dst.slice(0..written_range.len().into_usize()), start, ctx)
                 .await?;
             Slice::from_buf_bounds(Slice::into_inner(slice), bounds)
@@ -501,7 +456,7 @@ mod tests {
             assert_eq!(&buf, &content[range]);
         }
 
-        let file_contents = std::fs::read(file.file.path()).unwrap();
+        let file_contents = std::fs::read(file.buffered_writer.as_inner().path()).unwrap();
         assert!(file_contents == content[0..cap * 2]);
 
         let maybe_flushed_buffer_contents = file.buffered_writer.inspect_maybe_flushed().unwrap();
@@ -534,7 +489,7 @@ mod tests {
         // assert the state is as this test expects it to be
         let load_io_buf_res = file.load_to_io_buf(&ctx).await.unwrap();
         assert_eq!(&load_io_buf_res[..], &content[0..cap * 2 + cap / 2]);
-        let md = file.file.path().metadata().unwrap();
+        let md = file.buffered_writer.as_inner().path().metadata().unwrap();
         assert_eq!(
             md.len(),
             2 * cap.into_u64(),

pageserver/src/tenant/remote_timeline_client/download.rs

@@ -6,7 +6,6 @@
 use std::collections::HashSet;
 use std::future::Future;
 use std::str::FromStr;
-use std::sync::atomic::AtomicU64;
 use std::time::SystemTime;
 
 use anyhow::{Context, anyhow};
@@ -16,7 +15,7 @@ use remote_storage::{
     DownloadError, DownloadKind, DownloadOpts, GenericRemoteStorage, ListingMode, RemotePath,
 };
 use tokio::fs::{self, File, OpenOptions};
-use tokio::io::AsyncSeekExt;
+use tokio::io::{AsyncSeekExt, AsyncWriteExt};
 use tokio_util::io::StreamReader;
 use tokio_util::sync::CancellationToken;
 use tracing::warn;
@@ -41,10 +40,7 @@ use crate::span::{
 use crate::tenant::Generation;
 use crate::tenant::remote_timeline_client::{remote_layer_path, remote_timelines_path};
 use crate::tenant::storage_layer::LayerName;
-use crate::virtual_file;
-use crate::virtual_file::owned_buffers_io::write::FlushTaskError;
-use crate::virtual_file::{IoBufferMut, MaybeFatalIo, VirtualFile};
-use crate::virtual_file::{TempVirtualFile, owned_buffers_io};
+use crate::virtual_file::{MaybeFatalIo, VirtualFile, on_fatal_io_error};
 
 ///
 /// If 'metadata' is given, we will validate that the downloaded file's size matches that
@@ -76,34 +72,21 @@ pub async fn download_layer_file<'a>(
         layer_metadata.generation,
     );
 
-    let (bytes_amount, temp_file) = download_retry(
+    // Perform a rename inspired by durable_rename from file_utils.c.
+    // The sequence:
+    //     write(tmp)
+    //     fsync(tmp)
+    //     rename(tmp, new)
+    //     fsync(new)
+    //     fsync(parent)
+    // For more context about durable_rename check this email from postgres mailing list:
+    // https://www.postgresql.org/message-id/56583BDD.9060302@2ndquadrant.com
+    // If pageserver crashes the temp file will be deleted on startup and re-downloaded.
+    let temp_file_path = path_with_suffix_extension(local_path, TEMP_DOWNLOAD_EXTENSION);
+
+    let bytes_amount = download_retry(
         || async {
-            // TempVirtualFile requires us to never reuse a filename while an old
-            // instance of TempVirtualFile created with that filename is not done dropping yet.
-            // So, we use a monotonic counter to disambiguate the filenames.
-            static NEXT_TEMP_DISAMBIGUATOR: AtomicU64 = AtomicU64::new(1);
-            let filename_disambiguator =
-                NEXT_TEMP_DISAMBIGUATOR.fetch_add(1, std::sync::atomic::Ordering::Relaxed);
-
-            let temp_file_path = path_with_suffix_extension(
-                local_path,
-                &format!("{filename_disambiguator:x}.{TEMP_DOWNLOAD_EXTENSION}"),
-            );
-
-            let temp_file = TempVirtualFile::new(
-                VirtualFile::open_with_options_v2(
-                    &temp_file_path,
-                    virtual_file::OpenOptions::new()
-                        .create_new(true)
-                        .write(true),
-                    ctx,
-                )
-                .await
-                .with_context(|| format!("create a temp file for layer download: {temp_file_path}"))
-                .map_err(DownloadError::Other)?,
-                gate.enter().map_err(|_| DownloadError::Cancelled)?,
-            );
-            download_object(storage, &remote_path, temp_file, gate, cancel, ctx).await
+            download_object(storage, &remote_path, &temp_file_path, gate, cancel, ctx).await
         },
         &format!("download {remote_path:?}"),
         cancel,
@@ -113,8 +96,7 @@ pub async fn download_layer_file<'a>(
     let expected = layer_metadata.file_size;
     if expected != bytes_amount {
         return Err(DownloadError::Other(anyhow!(
-            "According to layer file metadata should have downloaded {expected} bytes but downloaded {bytes_amount} bytes into file {:?}",
-            temp_file.path()
+            "According to layer file metadata should have downloaded {expected} bytes but downloaded {bytes_amount} bytes into file {temp_file_path:?}",
         )));
     }
 
@@ -124,28 +106,11 @@ pub async fn download_layer_file<'a>(
         )))
     });
 
-    // Try rename before disarming the temp file.
-    // That way, if rename fails for whatever reason, we clean up the temp file on the return path.
-
-    fs::rename(temp_file.path(), &local_path)
+    fs::rename(&temp_file_path, &local_path)
         .await
         .with_context(|| format!("rename download layer file to {local_path}"))
         .map_err(DownloadError::Other)?;
 
-    // The temp file's VirtualFile points to the temp_file_path which we moved above.
-    // Drop it immediately, it's invalid.
-    // This will get better in https://github.com/neondatabase/neon/issues/11692
-    let _: VirtualFile = temp_file.disarm_into_inner();
-    // NB: The gate guard that was stored in `temp_file` is dropped but we continue
-    // to operate on it and on the parent timeline directory.
-    // Those operations are safe to do because higher-level code is holding another gate guard:
-    // - attached mode: the download task spawned by struct Layer is holding the gate guard
-    // - secondary mode: The TenantDownloader::download holds the gate open
-
-    // The rename above is not durable yet.
-    // It doesn't matter for crash consistency because pageserver startup deletes temp
-    // files and we'll re-download on demand if necessary.
-
     // We use fatal_err() below because the after the rename above,
     // the in-memory state of the filesystem already has the layer file in its final place,
     // and subsequent pageserver code could think it's durable while it really isn't.
@@ -181,64 +146,147 @@ pub async fn download_layer_file<'a>(
 async fn download_object(
     storage: &GenericRemoteStorage,
     src_path: &RemotePath,
-    destination_file: TempVirtualFile,
-    gate: &utils::sync::gate::Gate,
+    dst_path: &Utf8PathBuf,
+    #[cfg_attr(target_os = "macos", allow(unused_variables))] gate: &utils::sync::gate::Gate,
     cancel: &CancellationToken,
-    ctx: &RequestContext,
-) -> Result<(u64, TempVirtualFile), DownloadError> {
-    let mut download = storage
-        .download(src_path, &DownloadOpts::default(), cancel)
-        .await?;
+    #[cfg_attr(target_os = "macos", allow(unused_variables))] ctx: &RequestContext,
+) -> Result<u64, DownloadError> {
+    let res = match crate::virtual_file::io_engine::get() {
+        crate::virtual_file::io_engine::IoEngine::NotSet => panic!("unset"),
+        crate::virtual_file::io_engine::IoEngine::StdFs => {
+            async {
+                let destination_file = tokio::fs::File::create(dst_path)
+                    .await
+                    .with_context(|| format!("create a destination file for layer '{dst_path}'"))
+                    .map_err(DownloadError::Other)?;
 
-    pausable_failpoint!("before-downloading-layer-stream-pausable");
+                let download = storage
+                    .download(src_path, &DownloadOpts::default(), cancel)
+                    .await?;
 
-    let dst_path = destination_file.path().to_owned();
-    let mut buffered = owned_buffers_io::write::BufferedWriter::<IoBufferMut, _>::new(
-        destination_file,
-        0,
-        || IoBufferMut::with_capacity(super::BUFFER_SIZE),
-        gate.enter().map_err(|_| DownloadError::Cancelled)?,
-        cancel.child_token(),
-        ctx,
-        tracing::info_span!(parent: None, "download_object_buffered_writer", %dst_path),
-    );
+                pausable_failpoint!("before-downloading-layer-stream-pausable");
 
-    // TODO: use vectored write (writev) once supported by tokio-epoll-uring.
-    // There's chunks_vectored() on the stream.
-    let (bytes_amount, destination_file) = async {
-        while let Some(res) = futures::StreamExt::next(&mut download.download_stream).await {
-            let chunk = match res {
-                Ok(chunk) => chunk,
-                Err(e) => return Err(DownloadError::from(e)),
-            };
-            buffered
-                .write_buffered_borrowed(&chunk, ctx)
-                .await
-                .map_err(|e| match e {
-                    FlushTaskError::Cancelled => DownloadError::Cancelled,
-                })?;
-        }
-        buffered
-            .shutdown(
-                owned_buffers_io::write::BufferedWriterShutdownMode::PadThenTruncate,
-                ctx,
-            )
+                let mut buf_writer =
+                    tokio::io::BufWriter::with_capacity(super::BUFFER_SIZE, destination_file);
+
+                let mut reader = tokio_util::io::StreamReader::new(download.download_stream);
+
+                let bytes_amount = tokio::io::copy_buf(&mut reader, &mut buf_writer).await?;
+                buf_writer.flush().await?;
+
+                let mut destination_file = buf_writer.into_inner();
+
+                // Tokio doc here: https://docs.rs/tokio/1.17.0/tokio/fs/struct.File.html states that:
+                // A file will not be closed immediately when it goes out of scope if there are any IO operations
+                // that have not yet completed. To ensure that a file is closed immediately when it is dropped,
+                // you should call flush before dropping it.
+                //
+                // From the tokio code I see that it waits for pending operations to complete. There shouldt be any because
+                // we assume that `destination_file` file is fully written. I e there is no pending .write(...).await operations.
+                // But for additional safety lets check/wait for any pending operations.
+                destination_file
+                    .flush()
+                    .await
+                    .maybe_fatal_err("download_object sync_all")
+                    .with_context(|| format!("flush source file at {dst_path}"))
+                    .map_err(DownloadError::Other)?;
+
+                // not using sync_data because it can lose file size update
+                destination_file
+                    .sync_all()
+                    .await
+                    .maybe_fatal_err("download_object sync_all")
+                    .with_context(|| format!("failed to fsync source file at {dst_path}"))
+                    .map_err(DownloadError::Other)?;
+
+                Ok(bytes_amount)
+            }
             .await
-            .map_err(|e| match e {
-                FlushTaskError::Cancelled => DownloadError::Cancelled,
-            })
+        }
+        #[cfg(target_os = "linux")]
+        crate::virtual_file::io_engine::IoEngine::TokioEpollUring => {
+            use crate::virtual_file::owned_buffers_io::write::FlushTaskError;
+            use std::sync::Arc;
+
+            use crate::virtual_file::{IoBufferMut, owned_buffers_io};
+            async {
+                let destination_file = Arc::new(
+                    VirtualFile::create(dst_path, ctx)
+                        .await
+                        .with_context(|| {
+                            format!("create a destination file for layer '{dst_path}'")
+                        })
+                        .map_err(DownloadError::Other)?,
+                );
+
+                let mut download = storage
+                    .download(src_path, &DownloadOpts::default(), cancel)
+                    .await?;
+
+                pausable_failpoint!("before-downloading-layer-stream-pausable");
+
+                let mut buffered = owned_buffers_io::write::BufferedWriter::<IoBufferMut, _>::new(
+                    destination_file,
+                    || IoBufferMut::with_capacity(super::BUFFER_SIZE),
+                    gate.enter().map_err(|_| DownloadError::Cancelled)?,
+                    cancel.child_token(),
+                    ctx,
+                    tracing::info_span!(parent: None, "download_object_buffered_writer", %dst_path),
+                );
+
+                // TODO: use vectored write (writev) once supported by tokio-epoll-uring.
+                // There's chunks_vectored() on the stream.
+                let (bytes_amount, destination_file) = async {
+                    while let Some(res) =
+                        futures::StreamExt::next(&mut download.download_stream).await
+                    {
+                        let chunk = match res {
+                            Ok(chunk) => chunk,
+                            Err(e) => return Err(DownloadError::from(e)),
+                        };
+                        buffered
+                            .write_buffered_borrowed(&chunk, ctx)
+                            .await
+                            .map_err(|e| match e {
+                                FlushTaskError::Cancelled => DownloadError::Cancelled,
+                            })?;
+                    }
+                    let inner = buffered
+                        .flush_and_into_inner(ctx)
+                        .await
+                        .map_err(|e| match e {
+                            FlushTaskError::Cancelled => DownloadError::Cancelled,
+                        })?;
+                    Ok(inner)
+                }
+                .await?;
+
+                // not using sync_data because it can lose file size update
+                destination_file
+                    .sync_all()
+                    .await
+                    .maybe_fatal_err("download_object sync_all")
+                    .with_context(|| format!("failed to fsync source file at {dst_path}"))
+                    .map_err(DownloadError::Other)?;
+
+                Ok(bytes_amount)
+            }
+            .await
+        }
+    };
+
+    // in case the download failed, clean up
+    match res {
+        Ok(bytes_amount) => Ok(bytes_amount),
+        Err(e) => {
+            if let Err(e) = tokio::fs::remove_file(dst_path).await {
+                if e.kind() != std::io::ErrorKind::NotFound {
+                    on_fatal_io_error(&e, &format!("Removing temporary file {dst_path}"));
+                }
+            }
+            Err(e)
+        }
     }
-    .await?;
-
-    // not using sync_data because it can lose file size update
-    destination_file
-        .sync_all()
-        .await
-        .maybe_fatal_err("download_object sync_all")
-        .with_context(|| format!("failed to fsync source file at {dst_path}"))
-        .map_err(DownloadError::Other)?;
-
-    Ok((bytes_amount, destination_file))
 }
 
 const TEMP_DOWNLOAD_EXTENSION: &str = "temp_download";

pageserver/src/tenant/secondary/downloader.rs

@@ -646,7 +646,7 @@ enum UpdateError {
     NoData,
     #[error("Insufficient local storage space")]
     NoSpace,
-    #[error("Failed to download: {0}")]
+    #[error("Failed to download")]
     DownloadError(DownloadError),
     #[error(transparent)]
     Deserialize(#[from] serde_json::Error),
@@ -1521,11 +1521,12 @@ async fn load_heatmap(
     path: &Utf8PathBuf,
     ctx: &RequestContext,
 ) -> Result<Option<HeatMapTenant>, anyhow::Error> {
-    let st = match VirtualFile::read_to_string(path, ctx).await {
-        Ok(st) => st,
+    let mut file = match VirtualFile::open(path, ctx).await {
+        Ok(file) => file,
         Err(e) if e.kind() == std::io::ErrorKind::NotFound => return Ok(None),
         Err(e) => Err(e)?,
     };
+    let st = file.read_to_string(ctx).await?;
     let htm = serde_json::from_str(&st)?;
     Ok(Some(htm))
 }

pageserver/src/tenant/storage_layer/delta_layer.rs

@@ -29,11 +29,11 @@
 //!
 use std::collections::{HashMap, VecDeque};
 use std::fs::File;
+use std::io::SeekFrom;
 use std::ops::Range;
 use std::os::unix::fs::FileExt;
 use std::str::FromStr;
 use std::sync::Arc;
-use std::sync::atomic::AtomicU64;
 
 use anyhow::{Context, Result, bail, ensure};
 use camino::{Utf8Path, Utf8PathBuf};
@@ -45,13 +45,14 @@ use pageserver_api::keyspace::KeySpace;
 use pageserver_api::models::ImageCompressionAlgorithm;
 use pageserver_api::shard::TenantShardId;
 use pageserver_api::value::Value;
+use rand::Rng;
+use rand::distributions::Alphanumeric;
 use serde::{Deserialize, Serialize};
 use tokio::sync::OnceCell;
 use tokio_epoll_uring::IoBuf;
 use tokio_util::sync::CancellationToken;
 use tracing::*;
 use utils::bin_ser::BeSer;
-use utils::bin_ser::SerializeError;
 use utils::id::{TenantId, TimelineId};
 use utils::lsn::Lsn;
 
@@ -73,10 +74,8 @@ use crate::tenant::vectored_blob_io::{
     BlobFlag, BufView, StreamingVectoredReadPlanner, VectoredBlobReader, VectoredRead,
     VectoredReadPlanner,
 };
-use crate::virtual_file::TempVirtualFile;
 use crate::virtual_file::owned_buffers_io::io_buf_ext::{FullSlice, IoBufExt};
-use crate::virtual_file::owned_buffers_io::write::{Buffer, BufferedWriterShutdownMode};
-use crate::virtual_file::{self, IoBuffer, IoBufferMut, MaybeFatalIo, VirtualFile};
+use crate::virtual_file::{self, IoBufferMut, MaybeFatalIo, VirtualFile};
 use crate::{DELTA_FILE_MAGIC, STORAGE_FORMAT_VERSION, TEMP_FILE_SUFFIX};
 
 ///
@@ -114,15 +113,6 @@ impl From<&DeltaLayer> for Summary {
 }
 
 impl Summary {
-    /// Serializes the summary header into an aligned buffer of lenth `PAGE_SZ`.
-    pub fn ser_into_page(&self) -> Result<IoBuffer, SerializeError> {
-        let mut buf = IoBufferMut::with_capacity(PAGE_SZ);
-        Self::ser_into(self, &mut buf)?;
-        // Pad zeroes to the buffer so the length is a multiple of the alignment.
-        buf.extend_with(0, buf.capacity() - buf.len());
-        Ok(buf.freeze())
-    }
-
     pub(super) fn expected(
         tenant_id: TenantId,
         timeline_id: TimelineId,
@@ -298,20 +288,19 @@ impl DeltaLayer {
         key_start: Key,
         lsn_range: &Range<Lsn>,
     ) -> Utf8PathBuf {
-        // TempVirtualFile requires us to never reuse a filename while an old
-        // instance of TempVirtualFile created with that filename is not done dropping yet.
-        // So, we use a monotonic counter to disambiguate the filenames.
-        static NEXT_TEMP_DISAMBIGUATOR: AtomicU64 = AtomicU64::new(1);
-        let filename_disambiguator =
-            NEXT_TEMP_DISAMBIGUATOR.fetch_add(1, std::sync::atomic::Ordering::Relaxed);
+        let rand_string: String = rand::thread_rng()
+            .sample_iter(&Alphanumeric)
+            .take(8)
+            .map(char::from)
+            .collect();
 
         conf.timeline_path(tenant_shard_id, timeline_id)
             .join(format!(
-                "{}-XXX__{:016X}-{:016X}.{:x}.{}",
+                "{}-XXX__{:016X}-{:016X}.{}.{}",
                 key_start,
                 u64::from(lsn_range.start),
                 u64::from(lsn_range.end),
-                filename_disambiguator,
+                rand_string,
                 TEMP_FILE_SUFFIX,
             ))
     }
@@ -402,7 +391,7 @@ struct DeltaLayerWriterInner {
 
     tree: DiskBtreeBuilder<BlockBuf, DELTA_KEY_SIZE>,
 
-    blob_writer: BlobWriter<TempVirtualFile>,
+    blob_writer: BlobWriter<true>,
 
     // Number of key-lsns in the layer.
     num_keys: usize,
@@ -426,29 +415,16 @@ impl DeltaLayerWriterInner {
         // Create the file initially with a temporary filename. We don't know
         // the end key yet, so we cannot form the final filename yet. We will
         // rename it when we're done.
+        //
+        // Note: This overwrites any existing file. There shouldn't be any.
+        // FIXME: throw an error instead?
         let path =
             DeltaLayer::temp_path_for(conf, &tenant_shard_id, &timeline_id, key_start, &lsn_range);
-        let file = TempVirtualFile::new(
-            VirtualFile::open_with_options_v2(
-                &path,
-                virtual_file::OpenOptions::new()
-                    .create_new(true)
-                    .write(true),
-                ctx,
-            )
-            .await?,
-            gate.enter()?,
-        );
 
-        // Start at PAGE_SZ, make room for the header block
-        let blob_writer = BlobWriter::new(
-            file,
-            PAGE_SZ as u64,
-            gate,
-            cancel,
-            ctx,
-            info_span!(parent: None, "delta_layer_writer_flush_task", tenant_id=%tenant_shard_id.tenant_id, shard_id=%tenant_shard_id.shard_slug(), timeline_id=%timeline_id, path = %path),
-        )?;
+        let mut file = VirtualFile::create(&path, ctx).await?;
+        // make room for the header block
+        file.seek(SeekFrom::Start(PAGE_SZ as u64)).await?;
+        let blob_writer = BlobWriter::new(file, PAGE_SZ as u64, gate, cancel, ctx);
 
         // Initialize the b-tree index builder
         let block_buf = BlockBuf::new();
@@ -539,27 +515,34 @@ impl DeltaLayerWriterInner {
         self,
         key_end: Key,
         ctx: &RequestContext,
+    ) -> anyhow::Result<(PersistentLayerDesc, Utf8PathBuf)> {
+        let temp_path = self.path.clone();
+        let result = self.finish0(key_end, ctx).await;
+        if let Err(ref e) = result {
+            tracing::info!(%temp_path, "cleaning up temporary file after error during writing: {e}");
+            if let Err(e) = std::fs::remove_file(&temp_path) {
+                tracing::warn!(error=%e, %temp_path, "error cleaning up temporary layer file after error during writing");
+            }
+        }
+        result
+    }
+
+    async fn finish0(
+        self,
+        key_end: Key,
+        ctx: &RequestContext,
     ) -> anyhow::Result<(PersistentLayerDesc, Utf8PathBuf)> {
         let index_start_blk = self.blob_writer.size().div_ceil(PAGE_SZ as u64) as u32;
 
-        let file = self
-            .blob_writer
-            .shutdown(
-                BufferedWriterShutdownMode::ZeroPadToNextMultiple(PAGE_SZ),
-                ctx,
-            )
-            .await?;
+        let mut file = self.blob_writer.into_inner(ctx).await?;
 
         // Write out the index
         let (index_root_blk, block_buf) = self.tree.finish()?;
-        let mut offset = index_start_blk as u64 * PAGE_SZ as u64;
-
-        // TODO(yuchen): https://github.com/neondatabase/neon/issues/10092
-        // Should we just replace BlockBuf::blocks with one big buffer
+        file.seek(SeekFrom::Start(index_start_blk as u64 * PAGE_SZ as u64))
+            .await?;
         for buf in block_buf.blocks {
-            let (_buf, res) = file.write_all_at(buf.slice_len(), offset, ctx).await;
+            let (_buf, res) = file.write_all(buf.slice_len(), ctx).await;
             res?;
-            offset += PAGE_SZ as u64;
         }
         assert!(self.lsn_range.start < self.lsn_range.end);
         // Fill in the summary on blk 0
@@ -574,9 +557,11 @@ impl DeltaLayerWriterInner {
             index_root_blk,
         };
 
-        // Writes summary at the first block (offset 0).
-        let buf = summary.ser_into_page()?;
-        let (_buf, res) = file.write_all_at(buf.slice_len(), 0, ctx).await;
+        let mut buf = Vec::with_capacity(PAGE_SZ);
+        // TODO: could use smallvec here but it's a pain with Slice<T>
+        Summary::ser_into(&summary, &mut buf)?;
+        file.seek(SeekFrom::Start(0)).await?;
+        let (_buf, res) = file.write_all(buf.slice_len(), ctx).await;
         res?;
 
         let metadata = file
@@ -613,10 +598,6 @@ impl DeltaLayerWriterInner {
 
         trace!("created delta layer {}", self.path);
 
-        // The gate guard stored in `destination_file` is dropped. Callers (e.g.. flush loop or compaction)
-        // keep the gate open also, so that it's safe for them to rename the file to its final destination.
-        file.disarm_into_inner();
-
         Ok((desc, self.path))
     }
 }
@@ -745,6 +726,17 @@ impl DeltaLayerWriter {
     }
 }
 
+impl Drop for DeltaLayerWriter {
+    fn drop(&mut self) {
+        if let Some(inner) = self.inner.take() {
+            // We want to remove the virtual file here, so it's fine to not
+            // having completely flushed unwritten data.
+            let vfile = inner.blob_writer.into_inner_no_flush();
+            vfile.remove();
+        }
+    }
+}
+
 #[derive(thiserror::Error, Debug)]
 pub enum RewriteSummaryError {
     #[error("magic mismatch")]
@@ -768,7 +760,7 @@ impl DeltaLayer {
     where
         F: Fn(Summary) -> Summary,
     {
-        let file = VirtualFile::open_with_options_v2(
+        let mut file = VirtualFile::open_with_options(
             path,
             virtual_file::OpenOptions::new().read(true).write(true),
             ctx,
@@ -785,8 +777,11 @@ impl DeltaLayer {
 
         let new_summary = rewrite(actual_summary);
 
-        let buf = new_summary.ser_into_page().context("serialize")?;
-        let (_buf, res) = file.write_all_at(buf.slice_len(), 0, ctx).await;
+        let mut buf = Vec::with_capacity(PAGE_SZ);
+        // TODO: could use smallvec here, but it's a pain with Slice<T>
+        Summary::ser_into(&new_summary, &mut buf).context("serialize")?;
+        file.seek(SeekFrom::Start(0)).await?;
+        let (_buf, res) = file.write_all(buf.slice_len(), ctx).await;
         res?;
         Ok(())
     }
@@ -1442,19 +1437,6 @@ impl DeltaLayerInner {
     }
 
     pub fn iter<'a>(&'a self, ctx: &'a RequestContext) -> DeltaLayerIterator<'a> {
-        self.iter_with_options(
-            ctx,
-            1024 * 8192, // The default value. Unit tests might use a different value. 1024 * 8K = 8MB buffer.
-            1024,        // The default value. Unit tests might use a different value
-        )
-    }
-
-    pub fn iter_with_options<'a>(
-        &'a self,
-        ctx: &'a RequestContext,
-        max_read_size: u64,
-        max_batch_size: usize,
-    ) -> DeltaLayerIterator<'a> {
         let block_reader = FileBlockReader::new(&self.file, self.file_id);
         let tree_reader =
             DiskBtreeReader::new(self.index_start_blk, self.index_root_blk, block_reader);
@@ -1464,7 +1446,10 @@ impl DeltaLayerInner {
             index_iter: tree_reader.iter(&[0; DELTA_KEY_SIZE], ctx),
             key_values_batch: std::collections::VecDeque::new(),
             is_end: false,
-            planner: StreamingVectoredReadPlanner::new(max_read_size, max_batch_size),
+            planner: StreamingVectoredReadPlanner::new(
+                1024 * 8192, // The default value. Unit tests might use a different value. 1024 * 8K = 8MB buffer.
+                1024,        // The default value. Unit tests might use a different value
+            ),
         }
     }
 
@@ -1624,8 +1609,8 @@ pub(crate) mod test {
     use bytes::Bytes;
     use itertools::MinMaxResult;
     use pageserver_api::value::Value;
+    use rand::RngCore;
     use rand::prelude::{SeedableRng, SliceRandom, StdRng};
-    use rand::{Rng, RngCore};
 
     use super::*;
     use crate::DEFAULT_PG_VERSION;

pageserver/src/tenant/storage_layer/image_layer.rs

@@ -27,11 +27,11 @@
 //! actual page images are stored in the "values" part.
 use std::collections::{HashMap, VecDeque};
 use std::fs::File;
+use std::io::SeekFrom;
 use std::ops::Range;
 use std::os::unix::prelude::FileExt;
 use std::str::FromStr;
 use std::sync::Arc;
-use std::sync::atomic::AtomicU64;
 
 use anyhow::{Context, Result, bail, ensure};
 use bytes::Bytes;
@@ -43,13 +43,14 @@ use pageserver_api::key::{DBDIR_KEY, KEY_SIZE, Key};
 use pageserver_api::keyspace::KeySpace;
 use pageserver_api::shard::{ShardIdentity, TenantShardId};
 use pageserver_api::value::Value;
+use rand::Rng;
+use rand::distributions::Alphanumeric;
 use serde::{Deserialize, Serialize};
 use tokio::sync::OnceCell;
 use tokio_stream::StreamExt;
 use tokio_util::sync::CancellationToken;
 use tracing::*;
 use utils::bin_ser::BeSer;
-use utils::bin_ser::SerializeError;
 use utils::id::{TenantId, TimelineId};
 use utils::lsn::Lsn;
 
@@ -71,10 +72,8 @@ use crate::tenant::vectored_blob_io::{
     BlobFlag, BufView, StreamingVectoredReadPlanner, VectoredBlobReader, VectoredRead,
     VectoredReadPlanner,
 };
-use crate::virtual_file::TempVirtualFile;
 use crate::virtual_file::owned_buffers_io::io_buf_ext::IoBufExt;
-use crate::virtual_file::owned_buffers_io::write::{Buffer, BufferedWriterShutdownMode};
-use crate::virtual_file::{self, IoBuffer, IoBufferMut, MaybeFatalIo, VirtualFile};
+use crate::virtual_file::{self, IoBufferMut, MaybeFatalIo, VirtualFile};
 use crate::{IMAGE_FILE_MAGIC, STORAGE_FORMAT_VERSION, TEMP_FILE_SUFFIX};
 
 ///
@@ -113,15 +112,6 @@ impl From<&ImageLayer> for Summary {
 }
 
 impl Summary {
-    /// Serializes the summary header into an aligned buffer of lenth `PAGE_SZ`.
-    pub fn ser_into_page(&self) -> Result<IoBuffer, SerializeError> {
-        let mut buf = IoBufferMut::with_capacity(PAGE_SZ);
-        Self::ser_into(self, &mut buf)?;
-        // Pad zeroes to the buffer so the length is a multiple of the alignment.
-        buf.extend_with(0, buf.capacity() - buf.len());
-        Ok(buf.freeze())
-    }
-
     pub(super) fn expected(
         tenant_id: TenantId,
         timeline_id: TimelineId,
@@ -262,18 +252,14 @@ impl ImageLayer {
         tenant_shard_id: TenantShardId,
         fname: &ImageLayerName,
     ) -> Utf8PathBuf {
-        // TempVirtualFile requires us to never reuse a filename while an old
-        // instance of TempVirtualFile created with that filename is not done dropping yet.
-        // So, we use a monotonic counter to disambiguate the filenames.
-        static NEXT_TEMP_DISAMBIGUATOR: AtomicU64 = AtomicU64::new(1);
-        let filename_disambiguator =
-            NEXT_TEMP_DISAMBIGUATOR.fetch_add(1, std::sync::atomic::Ordering::Relaxed);
+        let rand_string: String = rand::thread_rng()
+            .sample_iter(&Alphanumeric)
+            .take(8)
+            .map(char::from)
+            .collect();
 
         conf.timeline_path(&tenant_shard_id, &timeline_id)
-            .join(format!(
-                "{fname}.{:x}.{TEMP_FILE_SUFFIX}",
-                filename_disambiguator
-            ))
+            .join(format!("{fname}.{rand_string}.{TEMP_FILE_SUFFIX}"))
     }
 
     ///
@@ -363,7 +349,7 @@ impl ImageLayer {
     where
         F: Fn(Summary) -> Summary,
     {
-        let file = VirtualFile::open_with_options_v2(
+        let mut file = VirtualFile::open_with_options(
             path,
             virtual_file::OpenOptions::new().read(true).write(true),
             ctx,
@@ -380,8 +366,11 @@ impl ImageLayer {
 
         let new_summary = rewrite(actual_summary);
 
-        let buf = new_summary.ser_into_page().context("serialize")?;
-        let (_buf, res) = file.write_all_at(buf.slice_len(), 0, ctx).await;
+        let mut buf = Vec::with_capacity(PAGE_SZ);
+        // TODO: could use smallvec here but it's a pain with Slice<T>
+        Summary::ser_into(&new_summary, &mut buf).context("serialize")?;
+        file.seek(SeekFrom::Start(0)).await?;
+        let (_buf, res) = file.write_all(buf.slice_len(), ctx).await;
         res?;
         Ok(())
     }
@@ -685,19 +674,6 @@ impl ImageLayerInner {
     }
 
     pub(crate) fn iter<'a>(&'a self, ctx: &'a RequestContext) -> ImageLayerIterator<'a> {
-        self.iter_with_options(
-            ctx,
-            1024 * 8192, // The default value. Unit tests might use a different value. 1024 * 8K = 8MB buffer.
-            1024,        // The default value. Unit tests might use a different value
-        )
-    }
-
-    pub(crate) fn iter_with_options<'a>(
-        &'a self,
-        ctx: &'a RequestContext,
-        max_read_size: u64,
-        max_batch_size: usize,
-    ) -> ImageLayerIterator<'a> {
         let block_reader = FileBlockReader::new(&self.file, self.file_id);
         let tree_reader =
             DiskBtreeReader::new(self.index_start_blk, self.index_root_blk, block_reader);
@@ -707,7 +683,10 @@ impl ImageLayerInner {
             index_iter: tree_reader.iter(&[0; KEY_SIZE], ctx),
             key_values_batch: VecDeque::new(),
             is_end: false,
-            planner: StreamingVectoredReadPlanner::new(max_read_size, max_batch_size),
+            planner: StreamingVectoredReadPlanner::new(
+                1024 * 8192, // The default value. Unit tests might use a different value. 1024 * 8K = 8MB buffer.
+                1024,        // The default value. Unit tests might use a different value
+            ),
         }
     }
 
@@ -760,7 +739,7 @@ struct ImageLayerWriterInner {
     // Number of keys in the layer.
     num_keys: usize,
 
-    blob_writer: BlobWriter<TempVirtualFile>,
+    blob_writer: BlobWriter<false>,
     tree: DiskBtreeBuilder<BlockBuf, KEY_SIZE>,
 
     #[cfg(feature = "testing")]
@@ -794,27 +773,19 @@ impl ImageLayerWriterInner {
             },
         );
         trace!("creating image layer {}", path);
-        let file = TempVirtualFile::new(
-            VirtualFile::open_with_options_v2(
+        let mut file = {
+            VirtualFile::open_with_options(
                 &path,
                 virtual_file::OpenOptions::new()
-                    .create_new(true)
-                    .write(true),
+                    .write(true)
+                    .create_new(true),
                 ctx,
             )
-            .await?,
-            gate.enter()?,
-        );
-
-        // Start at `PAGE_SZ` to make room for the header block.
-        let blob_writer = BlobWriter::new(
-            file,
-            PAGE_SZ as u64,
-            gate,
-            cancel,
-            ctx,
-            info_span!(parent: None, "image_layer_writer_flush_task", tenant_id=%tenant_shard_id.tenant_id, shard_id=%tenant_shard_id.shard_slug(), timeline_id=%timeline_id, path = %path),
-        )?;
+            .await?
+        };
+        // make room for the header block
+        file.seek(SeekFrom::Start(PAGE_SZ as u64)).await?;
+        let blob_writer = BlobWriter::new(file, PAGE_SZ as u64, gate, cancel, ctx);
 
         // Initialize the b-tree index builder
         let block_buf = BlockBuf::new();
@@ -925,6 +896,25 @@ impl ImageLayerWriterInner {
         self,
         ctx: &RequestContext,
         end_key: Option<Key>,
+    ) -> anyhow::Result<(PersistentLayerDesc, Utf8PathBuf)> {
+        let temp_path = self.path.clone();
+        let result = self.finish0(ctx, end_key).await;
+        if let Err(ref e) = result {
+            tracing::info!(%temp_path, "cleaning up temporary file after error during writing: {e}");
+            if let Err(e) = std::fs::remove_file(&temp_path) {
+                tracing::warn!(error=%e, %temp_path, "error cleaning up temporary layer file after error during writing");
+            }
+        }
+        result
+    }
+
+    ///
+    /// Finish writing the image layer.
+    ///
+    async fn finish0(
+        self,
+        ctx: &RequestContext,
+        end_key: Option<Key>,
     ) -> anyhow::Result<(PersistentLayerDesc, Utf8PathBuf)> {
         let index_start_blk = self.blob_writer.size().div_ceil(PAGE_SZ as u64) as u32;
 
@@ -942,24 +932,15 @@ impl ImageLayerWriterInner {
             crate::metrics::COMPRESSION_IMAGE_OUTPUT_BYTES.inc_by(compressed_size);
         };
 
-        let file = self
-            .blob_writer
-            .shutdown(
-                BufferedWriterShutdownMode::ZeroPadToNextMultiple(PAGE_SZ),
-                ctx,
-            )
-            .await?;
+        let mut file = self.blob_writer.into_inner();
 
         // Write out the index
-        let mut offset = index_start_blk as u64 * PAGE_SZ as u64;
+        file.seek(SeekFrom::Start(index_start_blk as u64 * PAGE_SZ as u64))
+            .await?;
         let (index_root_blk, block_buf) = self.tree.finish()?;
-
-        // TODO(yuchen): https://github.com/neondatabase/neon/issues/10092
-        // Should we just replace BlockBuf::blocks with one big buffer?
         for buf in block_buf.blocks {
-            let (_buf, res) = file.write_all_at(buf.slice_len(), offset, ctx).await;
+            let (_buf, res) = file.write_all(buf.slice_len(), ctx).await;
             res?;
-            offset += PAGE_SZ as u64;
         }
 
         let final_key_range = if let Some(end_key) = end_key {
@@ -980,9 +961,11 @@ impl ImageLayerWriterInner {
             index_root_blk,
         };
 
-        // Writes summary at the first block (offset 0).
-        let buf = summary.ser_into_page()?;
-        let (_buf, res) = file.write_all_at(buf.slice_len(), 0, ctx).await;
+        let mut buf = Vec::with_capacity(PAGE_SZ);
+        // TODO: could use smallvec here but it's a pain with Slice<T>
+        Summary::ser_into(&summary, &mut buf)?;
+        file.seek(SeekFrom::Start(0)).await?;
+        let (_buf, res) = file.write_all(buf.slice_len(), ctx).await;
         res?;
 
         let metadata = file
@@ -1017,10 +1000,6 @@ impl ImageLayerWriterInner {
 
         trace!("created image layer {}", self.path);
 
-        // The gate guard stored in `destination_file` is dropped. Callers (e.g.. flush loop or compaction)
-        // keep the gate open also, so that it's safe for them to rename the file to its final destination.
-        file.disarm_into_inner();
-
         Ok((desc, self.path))
     }
 }
@@ -1146,6 +1125,14 @@ impl ImageLayerWriter {
     }
 }
 
+impl Drop for ImageLayerWriter {
+    fn drop(&mut self) {
+        if let Some(inner) = self.inner.take() {
+            inner.blob_writer.into_inner().remove();
+        }
+    }
+}
+
 pub struct ImageLayerIterator<'a> {
     image_layer: &'a ImageLayerInner,
     ctx: &'a RequestContext,

pageserver/src/tenant/storage_layer/merge_iterator.rs

@@ -19,7 +19,6 @@ pub(crate) enum LayerRef<'a> {
 }
 
 impl<'a> LayerRef<'a> {
-    #[allow(dead_code)]
     fn iter(self, ctx: &'a RequestContext) -> LayerIterRef<'a> {
         match self {
             Self::Image(x) => LayerIterRef::Image(x.iter(ctx)),
@@ -27,22 +26,6 @@ impl<'a> LayerRef<'a> {
         }
     }
 
-    fn iter_with_options(
-        self,
-        ctx: &'a RequestContext,
-        max_read_size: u64,
-        max_batch_size: usize,
-    ) -> LayerIterRef<'a> {
-        match self {
-            Self::Image(x) => {
-                LayerIterRef::Image(x.iter_with_options(ctx, max_read_size, max_batch_size))
-            }
-            Self::Delta(x) => {
-                LayerIterRef::Delta(x.iter_with_options(ctx, max_read_size, max_batch_size))
-            }
-        }
-    }
-
     fn layer_dbg_info(&self) -> String {
         match self {
             Self::Image(x) => x.layer_dbg_info(),
@@ -83,8 +66,6 @@ pub(crate) enum IteratorWrapper<'a> {
         first_key_lower_bound: (Key, Lsn),
         layer: LayerRef<'a>,
         source_desc: Arc<PersistentLayerKey>,
-        max_read_size: u64,
-        max_batch_size: usize,
     },
     Loaded {
         iter: PeekableLayerIterRef<'a>,
@@ -165,8 +146,6 @@ impl<'a> IteratorWrapper<'a> {
     pub fn create_from_image_layer(
         image_layer: &'a ImageLayerInner,
         ctx: &'a RequestContext,
-        max_read_size: u64,
-        max_batch_size: usize,
     ) -> Self {
         Self::NotLoaded {
             layer: LayerRef::Image(image_layer),
@@ -178,16 +157,12 @@ impl<'a> IteratorWrapper<'a> {
                 is_delta: false,
             }
             .into(),
-            max_read_size,
-            max_batch_size,
         }
     }
 
     pub fn create_from_delta_layer(
         delta_layer: &'a DeltaLayerInner,
         ctx: &'a RequestContext,
-        max_read_size: u64,
-        max_batch_size: usize,
     ) -> Self {
         Self::NotLoaded {
             layer: LayerRef::Delta(delta_layer),
@@ -199,8 +174,6 @@ impl<'a> IteratorWrapper<'a> {
                 is_delta: true,
             }
             .into(),
-            max_read_size,
-            max_batch_size,
         }
     }
 
@@ -231,13 +204,11 @@ impl<'a> IteratorWrapper<'a> {
             first_key_lower_bound,
             layer,
             source_desc,
-            max_read_size,
-            max_batch_size,
         } = self
         else {
             unreachable!()
         };
-        let iter = layer.iter_with_options(ctx, *max_read_size, *max_batch_size);
+        let iter = layer.iter(ctx);
         let iter = PeekableLayerIterRef::create(iter).await?;
         if let Some((k1, l1, _)) = iter.peek() {
             let (k2, l2) = first_key_lower_bound;
@@ -322,41 +293,21 @@ impl MergeIteratorItem for ((Key, Lsn, Value), Arc<PersistentLayerKey>) {
 }
 
 impl<'a> MergeIterator<'a> {
-    pub fn create_with_options(
-        deltas: &[&'a DeltaLayerInner],
-        images: &[&'a ImageLayerInner],
-        ctx: &'a RequestContext,
-        max_read_size: u64,
-        max_batch_size: usize,
-    ) -> Self {
-        let mut heap = Vec::with_capacity(images.len() + deltas.len());
-        for image in images {
-            heap.push(IteratorWrapper::create_from_image_layer(
-                image,
-                ctx,
-                max_read_size,
-                max_batch_size,
-            ));
-        }
-        for delta in deltas {
-            heap.push(IteratorWrapper::create_from_delta_layer(
-                delta,
-                ctx,
-                max_read_size,
-                max_batch_size,
-            ));
-        }
-        Self {
-            heap: BinaryHeap::from(heap),
-        }
-    }
-
     pub fn create(
         deltas: &[&'a DeltaLayerInner],
         images: &[&'a ImageLayerInner],
         ctx: &'a RequestContext,
     ) -> Self {
-        Self::create_with_options(deltas, images, ctx, 1024 * 8192, 1024)
+        let mut heap = Vec::with_capacity(images.len() + deltas.len());
+        for image in images {
+            heap.push(IteratorWrapper::create_from_image_layer(image, ctx));
+        }
+        for delta in deltas {
+            heap.push(IteratorWrapper::create_from_delta_layer(delta, ctx));
+        }
+        Self {
+            heap: BinaryHeap::from(heap),
+        }
     }
 
     pub(crate) async fn next_inner<R: MergeIteratorItem>(&mut self) -> anyhow::Result<Option<R>> {

pageserver/src/tenant/timeline/compaction.rs

@@ -2828,41 +2828,6 @@ impl Timeline {
         Ok(())
     }
 
-    /// Check if the memory usage is within the limit.
-    async fn check_memory_usage(
-        self: &Arc<Self>,
-        layer_selection: &[Layer],
-    ) -> Result<(), CompactionError> {
-        let mut estimated_memory_usage_mb = 0.0;
-        let mut num_image_layers = 0;
-        let mut num_delta_layers = 0;
-        let target_layer_size_bytes = 256 * 1024 * 1024;
-        for layer in layer_selection {
-            let layer_desc = layer.layer_desc();
-            if layer_desc.is_delta() {
-                // Delta layers at most have 1MB buffer; 3x to make it safe (there're deltas as large as 16KB).
-                // Multiply the layer size so that tests can pass.
-                estimated_memory_usage_mb +=
-                    3.0 * (layer_desc.file_size / target_layer_size_bytes) as f64;
-                num_delta_layers += 1;
-            } else {
-                // Image layers at most have 1MB buffer but it might be compressed; assume 5x compression ratio.
-                estimated_memory_usage_mb +=
-                    5.0 * (layer_desc.file_size / target_layer_size_bytes) as f64;
-                num_image_layers += 1;
-            }
-        }
-        if estimated_memory_usage_mb > 1024.0 {
-            return Err(CompactionError::Other(anyhow!(
-                "estimated memory usage is too high: {}MB, giving up compaction; num_image_layers={}, num_delta_layers={}",
-                estimated_memory_usage_mb,
-                num_image_layers,
-                num_delta_layers
-            )));
-        }
-        Ok(())
-    }
-
     /// Get a watermark for gc-compaction, that is the lowest LSN that we can use as the `gc_horizon` for
     /// the compaction algorithm. It is min(space_cutoff, time_cutoff, latest_gc_cutoff, standby_horizon).
     /// Leases and retain_lsns are considered in the gc-compaction job itself so we don't need to account for them
@@ -3299,17 +3264,6 @@ impl Timeline {
         self.check_compaction_space(&job_desc.selected_layers)
             .await?;
 
-        self.check_memory_usage(&job_desc.selected_layers).await?;
-        if job_desc.selected_layers.len() > 100
-            && job_desc.rewrite_layers.len() as f64 >= job_desc.selected_layers.len() as f64 * 0.7
-        {
-            return Err(CompactionError::Other(anyhow!(
-                "too many layers to rewrite: {} / {}, giving up compaction",
-                job_desc.rewrite_layers.len(),
-                job_desc.selected_layers.len()
-            )));
-        }
-
         // Generate statistics for the compaction
         for layer in &job_desc.selected_layers {
             let desc = layer.layer_desc();
@@ -3405,13 +3359,7 @@ impl Timeline {
             .context("failed to collect gc compaction keyspace")
             .map_err(CompactionError::Other)?;
         let mut merge_iter = FilterIterator::create(
-            MergeIterator::create_with_options(
-                &delta_layers,
-                &image_layers,
-                ctx,
-                128 * 8192, /* 1MB buffer for each of the inner iterators */
-                128,
-            ),
+            MergeIterator::create(&delta_layers, &image_layers, ctx),
             dense_ks,
             sparse_ks,
         )

pageserver/src/tenant/timeline/import_pgdata.rs

@@ -1,21 +1,20 @@
 use std::sync::Arc;
 
 use anyhow::{Context, bail};
-use pageserver_api::models::ShardImportStatus;
 use remote_storage::RemotePath;
 use tokio_util::sync::CancellationToken;
-use tracing::info;
+use tracing::{Instrument, info, info_span};
 use utils::lsn::Lsn;
 
 use super::Timeline;
 use crate::context::RequestContext;
-use crate::controller_upcall_client::{StorageControllerUpcallApi, StorageControllerUpcallClient};
 use crate::tenant::metadata::TimelineMetadata;
 
 mod flow;
 mod importbucket_client;
 mod importbucket_format;
 pub(crate) mod index_part_format;
+pub(crate) mod upcall_api;
 
 pub async fn doit(
     timeline: &Arc<Timeline>,
@@ -35,6 +34,23 @@ pub async fn doit(
 
     let storage = importbucket_client::new(timeline.conf, &location, cancel.clone()).await?;
 
+    info!("get spec early so we know we'll be able to upcall when done");
+    let Some(spec) = storage.get_spec().await? else {
+        bail!("spec not found")
+    };
+
+    let upcall_client =
+        upcall_api::Client::new(timeline.conf, cancel.clone()).context("create upcall client")?;
+
+    //
+    // send an early progress update to clean up k8s job early and generate potentially useful logs
+    //
+    info!("send early progress update");
+    upcall_client
+        .send_progress_until_success(&spec)
+        .instrument(info_span!("early_progress_update"))
+        .await?;
+
     let status_prefix = RemotePath::from_string("status").unwrap();
 
     //
@@ -160,21 +176,7 @@ pub async fn doit(
 
         //
         // Communicate that shard is done.
-        // Ensure at-least-once delivery of the upcall to storage controller
-        // before we mark the task as done and never come here again.
         //
-        let storcon_client = StorageControllerUpcallClient::new(timeline.conf, &cancel)?
-            .expect("storcon configured");
-        storcon_client
-            .put_timeline_import_status(
-                timeline.tenant_shard_id,
-                timeline.timeline_id,
-                // TODO(vlad): What about import errors?
-                ShardImportStatus::Done,
-            )
-            .await
-            .map_err(|_err| anyhow::anyhow!("Shut down while putting timeline import status"))?;
-
         storage
             .put_json(
                 &shard_status_key,
@@ -184,6 +186,16 @@ pub async fn doit(
             .context("put shard status")?;
     }
 
+    //
+    // Ensure at-least-once deliver of the upcall to cplane
+    // before we mark the task as done and never come here again.
+    //
+    info!("send final progress update");
+    upcall_client
+        .send_progress_until_success(&spec)
+        .instrument(info_span!("final_progress_update"))
+        .await?;
+
     //
     // Mark as done in index_part.
     // This makes subsequent timeline loads enter the normal load code path

pageserver/src/tenant/timeline/import_pgdata/importbucket_client.rs

@@ -13,7 +13,7 @@ use tokio_util::sync::CancellationToken;
 use tracing::{debug, info, instrument};
 use utils::lsn::Lsn;
 
-use super::index_part_format;
+use super::{importbucket_format, index_part_format};
 use crate::assert_u64_eq_usize::U64IsUsize;
 use crate::config::PageServerConf;
 
@@ -173,6 +173,12 @@ impl RemoteStorageWrapper {
         res
     }
 
+    pub async fn get_spec(&self) -> Result<Option<importbucket_format::Spec>, anyhow::Error> {
+        self.get_json(&RemotePath::from_string("spec.json").unwrap())
+            .await
+            .context("get spec")
+    }
+
     #[instrument(level = tracing::Level::DEBUG, skip_all, fields(%path))]
     pub async fn get_json<T: DeserializeOwned>(
         &self,
@@ -238,8 +244,7 @@ impl RemoteStorageWrapper {
                             kind: DownloadKind::Large,
                             etag: None,
                             byte_start: Bound::Included(start_inclusive),
-                            byte_end: Bound::Excluded(end_exclusive),
-                            version_id: None,
+                            byte_end: Bound::Excluded(end_exclusive)
                         },
                         &self.cancel)
                     .await?;

pageserver/src/tenant/timeline/import_pgdata/importbucket_format.rs

@@ -11,3 +11,10 @@ pub struct ShardStatus {
     pub done: bool,
     // TODO: remaining fields
 }
+
+// TODO: dedupe with fast_import code
+#[derive(Deserialize, Serialize, Debug, Clone, PartialEq, Eq)]
+pub struct Spec {
+    pub project_id: String,
+    pub branch_id: String,
+}

pageserver/src/tenant/timeline/import_pgdata/upcall_api.rs

@@ -0,0 +1,124 @@
+//! FIXME: most of this is copy-paste from mgmt_api.rs ; dedupe into a `reqwest_utils::Client` crate.
+use pageserver_client::mgmt_api::{Error, ResponseErrorMessageExt};
+use reqwest::{Certificate, Method};
+use serde::{Deserialize, Serialize};
+use tokio_util::sync::CancellationToken;
+use tracing::error;
+
+use super::importbucket_format::Spec;
+use crate::config::PageServerConf;
+
+pub struct Client {
+    base_url: String,
+    authorization_header: Option<String>,
+    client: reqwest::Client,
+    cancel: CancellationToken,
+}
+
+pub type Result<T> = std::result::Result<T, Error>;
+
+#[derive(Serialize, Deserialize, Debug)]
+struct ImportProgressRequest {
+    // no fields yet, not sure if there every will be any
+}
+
+#[derive(Serialize, Deserialize, Debug)]
+struct ImportProgressResponse {
+    // we don't care
+}
+
+impl Client {
+    pub fn new(conf: &PageServerConf, cancel: CancellationToken) -> anyhow::Result<Self> {
+        let Some(ref base_url) = conf.import_pgdata_upcall_api else {
+            anyhow::bail!("import_pgdata_upcall_api is not configured")
+        };
+        let mut http_client = reqwest::Client::builder();
+        for cert in &conf.ssl_ca_certs {
+            http_client = http_client.add_root_certificate(Certificate::from_der(cert.contents())?);
+        }
+        let http_client = http_client.build()?;
+
+        Ok(Self {
+            base_url: base_url.to_string(),
+            client: http_client,
+            cancel,
+            authorization_header: conf
+                .import_pgdata_upcall_api_token
+                .as_ref()
+                .map(|secret_string| secret_string.get_contents())
+                .map(|jwt| format!("Bearer {jwt}")),
+        })
+    }
+
+    fn start_request<U: reqwest::IntoUrl>(
+        &self,
+        method: Method,
+        uri: U,
+    ) -> reqwest::RequestBuilder {
+        let req = self.client.request(method, uri);
+        if let Some(value) = &self.authorization_header {
+            req.header(reqwest::header::AUTHORIZATION, value)
+        } else {
+            req
+        }
+    }
+
+    async fn request_noerror<B: serde::Serialize, U: reqwest::IntoUrl>(
+        &self,
+        method: Method,
+        uri: U,
+        body: B,
+    ) -> Result<reqwest::Response> {
+        self.start_request(method, uri)
+            .json(&body)
+            .send()
+            .await
+            .map_err(Error::ReceiveBody)
+    }
+
+    async fn request<B: serde::Serialize, U: reqwest::IntoUrl>(
+        &self,
+        method: Method,
+        uri: U,
+        body: B,
+    ) -> Result<reqwest::Response> {
+        let res = self.request_noerror(method, uri, body).await?;
+        let response = res.error_from_body().await?;
+        Ok(response)
+    }
+
+    pub async fn send_progress_once(&self, spec: &Spec) -> Result<()> {
+        let url = format!(
+            "{}/projects/{}/branches/{}/import_progress",
+            self.base_url, spec.project_id, spec.branch_id
+        );
+        let ImportProgressResponse {} = self
+            .request(Method::POST, url, &ImportProgressRequest {})
+            .await?
+            .json()
+            .await
+            .map_err(Error::ReceiveBody)?;
+        Ok(())
+    }
+
+    pub async fn send_progress_until_success(&self, spec: &Spec) -> anyhow::Result<()> {
+        loop {
+            match self.send_progress_once(spec).await {
+                Ok(()) => return Ok(()),
+                Err(Error::Cancelled) => return Err(anyhow::anyhow!("cancelled")),
+                Err(err) => {
+                    error!(?err, "error sending progress, retrying");
+                    if tokio::time::timeout(
+                        std::time::Duration::from_secs(10),
+                        self.cancel.cancelled(),
+                    )
+                    .await
+                    .is_ok()
+                    {
+                        anyhow::bail!("cancelled while sending early progress update");
+                    }
+                }
+            }
+        }
+    }
+}

pageserver/src/tenant/vectored_blob_io.rs

@@ -507,9 +507,7 @@ impl<'a> VectoredBlobReader<'a> {
 
         for (blob_start, meta) in blobs_at.iter().copied() {
             let header_start = (blob_start - read.start) as usize;
-            let header = Header::decode(&buf[header_start..]).map_err(|anyhow_err| {
-                std::io::Error::new(std::io::ErrorKind::InvalidData, anyhow_err)
-            })?;
+            let header = Header::decode(&buf[header_start..])?;
             let data_start = header_start + header.header_len;
             let end = data_start + header.data_len;
             let compression_bits = header.compression_bits;
@@ -664,6 +662,7 @@ impl StreamingVectoredReadPlanner {
 
 #[cfg(test)]
 mod tests {
+    use anyhow::Error;
 
     use super::super::blob_io::tests::{random_array, write_maybe_compressed};
     use super::*;
@@ -946,16 +945,13 @@ mod tests {
         }
     }
 
-    async fn round_trip_test_compressed(
-        blobs: &[Vec<u8>],
-        compression: bool,
-    ) -> anyhow::Result<()> {
+    async fn round_trip_test_compressed(blobs: &[Vec<u8>], compression: bool) -> Result<(), Error> {
         let ctx =
             RequestContext::new(TaskKind::UnitTest, DownloadBehavior::Error).with_scope_unit_test();
         let (_temp_dir, pathbuf, offsets) =
-            write_maybe_compressed(blobs, compression, &ctx).await?;
+            write_maybe_compressed::<true>(blobs, compression, &ctx).await?;
 
-        let file = VirtualFile::open_v2(&pathbuf, &ctx).await?;
+        let file = VirtualFile::open(&pathbuf, &ctx).await?;
         let file_len = std::fs::metadata(&pathbuf)?.len();
 
         // Multiply by two (compressed data might need more space), and add a few bytes for the header
@@ -1001,7 +997,7 @@ mod tests {
     }
 
     #[tokio::test]
-    async fn test_really_big_array() -> anyhow::Result<()> {
+    async fn test_really_big_array() -> Result<(), Error> {
         let blobs = &[
             b"test".to_vec(),
             random_array(10 * PAGE_SZ),
@@ -1016,7 +1012,7 @@ mod tests {
     }
 
     #[tokio::test]
-    async fn test_arrays_inc() -> anyhow::Result<()> {
+    async fn test_arrays_inc() -> Result<(), Error> {
         let blobs = (0..PAGE_SZ / 8)
             .map(|v| random_array(v * 16))
             .collect::<Vec<_>>();

pageserver/src/virtual_file.rs

@@ -12,11 +12,10 @@
 //! src/backend/storage/file/fd.c
 //!
 use std::fs::File;
-use std::io::{Error, ErrorKind};
+use std::io::{Error, ErrorKind, Seek, SeekFrom};
 use std::os::fd::{AsRawFd, FromRawFd, IntoRawFd, OwnedFd, RawFd};
 #[cfg(target_os = "linux")]
 use std::os::unix::fs::OpenOptionsExt;
-use std::sync::LazyLock;
 use std::sync::atomic::{AtomicBool, AtomicU8, AtomicUsize, Ordering};
 
 use camino::{Utf8Path, Utf8PathBuf};
@@ -26,31 +25,29 @@ use owned_buffers_io::aligned_buffer::{AlignedBufferMut, AlignedSlice, ConstAlig
 use owned_buffers_io::io_buf_aligned::{IoBufAligned, IoBufAlignedMut};
 use owned_buffers_io::io_buf_ext::FullSlice;
 use pageserver_api::config::defaults::DEFAULT_IO_BUFFER_ALIGNMENT;
+pub use pageserver_api::models::virtual_file as api;
 use tokio::sync::{RwLock, RwLockReadGuard, RwLockWriteGuard};
 use tokio::time::Instant;
 use tokio_epoll_uring::{BoundedBuf, IoBuf, IoBufMut, Slice};
 
-use self::owned_buffers_io::write::OwnedAsyncWriter;
 use crate::assert_u64_eq_usize::UsizeIsU64;
 use crate::context::RequestContext;
 use crate::metrics::{STORAGE_IO_TIME_METRIC, StorageIoOperation};
 use crate::page_cache::{PAGE_SZ, PageWriteGuard};
-
-pub(crate) use api::IoMode;
-pub(crate) use io_engine::IoEngineKind;
+pub(crate) mod io_engine;
 pub use io_engine::{
     FeatureTestResult as IoEngineFeatureTestResult, feature_test as io_engine_feature_test,
     io_engine_for_bench,
 };
-pub(crate) use metadata::Metadata;
-pub(crate) use open_options::*;
-pub use pageserver_api::models::virtual_file as api;
-pub use temporary::TempVirtualFile;
-
-pub(crate) mod io_engine;
 mod metadata;
 mod open_options;
-mod temporary;
+pub(crate) use api::IoMode;
+pub(crate) use io_engine::IoEngineKind;
+pub(crate) use metadata::Metadata;
+pub(crate) use open_options::*;
+
+use self::owned_buffers_io::write::OwnedAsyncWriter;
+
 pub(crate) mod owned_buffers_io {
     //! Abstractions for IO with owned buffers.
     //!
@@ -97,38 +94,69 @@ impl VirtualFile {
         Self::open_with_options_v2(path.as_ref(), OpenOptions::new().read(true), ctx).await
     }
 
+    pub async fn create<P: AsRef<Utf8Path>>(
+        path: P,
+        ctx: &RequestContext,
+    ) -> Result<Self, std::io::Error> {
+        let inner = VirtualFileInner::create(path, ctx).await?;
+        Ok(VirtualFile {
+            inner,
+            _mode: IoMode::Buffered,
+        })
+    }
+
+    pub async fn create_v2<P: AsRef<Utf8Path>>(
+        path: P,
+        ctx: &RequestContext,
+    ) -> Result<Self, std::io::Error> {
+        VirtualFile::open_with_options_v2(
+            path.as_ref(),
+            OpenOptions::new().write(true).create(true).truncate(true),
+            ctx,
+        )
+        .await
+    }
+
+    pub async fn open_with_options<P: AsRef<Utf8Path>>(
+        path: P,
+        open_options: &OpenOptions,
+        ctx: &RequestContext,
+    ) -> Result<Self, std::io::Error> {
+        let inner = VirtualFileInner::open_with_options(path, open_options, ctx).await?;
+        Ok(VirtualFile {
+            inner,
+            _mode: IoMode::Buffered,
+        })
+    }
+
     pub async fn open_with_options_v2<P: AsRef<Utf8Path>>(
         path: P,
         open_options: &OpenOptions,
         ctx: &RequestContext,
     ) -> Result<Self, std::io::Error> {
-        let mode = get_io_mode();
-        let set_o_direct = match (mode, open_options.is_write()) {
-            (IoMode::Buffered, _) => false,
-            #[cfg(target_os = "linux")]
-            (IoMode::Direct, false) => true,
-            #[cfg(target_os = "linux")]
-            (IoMode::Direct, true) => false,
-            #[cfg(target_os = "linux")]
-            (IoMode::DirectRw, _) => true,
-        };
-        let open_options = open_options.clone();
-        let open_options = if set_o_direct {
-            #[cfg(target_os = "linux")]
-            {
-                let mut open_options = open_options;
-                open_options.custom_flags(nix::libc::O_DIRECT);
-                open_options
+        let file = match get_io_mode() {
+            IoMode::Buffered => {
+                let inner = VirtualFileInner::open_with_options(path, open_options, ctx).await?;
+                VirtualFile {
+                    inner,
+                    _mode: IoMode::Buffered,
+                }
+            }
+            #[cfg(target_os = "linux")]
+            IoMode::Direct => {
+                let inner = VirtualFileInner::open_with_options(
+                    path,
+                    open_options.clone().custom_flags(nix::libc::O_DIRECT),
+                    ctx,
+                )
+                .await?;
+                VirtualFile {
+                    inner,
+                    _mode: IoMode::Direct,
+                }
             }
-            #[cfg(not(target_os = "linux"))]
-            unreachable!(
-                "O_DIRECT is not supported on this platform, IoMode's that result in set_o_direct=true shouldn't even be defined"
-            );
-        } else {
-            open_options
         };
-        let inner = VirtualFileInner::open_with_options(path, open_options, ctx).await?;
-        Ok(VirtualFile { inner, _mode: mode })
+        Ok(file)
     }
 
     pub fn path(&self) -> &Utf8Path {
@@ -157,14 +185,18 @@ impl VirtualFile {
         self.inner.sync_data().await
     }
 
-    pub async fn set_len(&self, len: u64, ctx: &RequestContext) -> Result<(), Error> {
-        self.inner.set_len(len, ctx).await
-    }
-
     pub async fn metadata(&self) -> Result<Metadata, Error> {
         self.inner.metadata().await
     }
 
+    pub fn remove(self) {
+        self.inner.remove();
+    }
+
+    pub async fn seek(&mut self, pos: SeekFrom) -> Result<u64, Error> {
+        self.inner.seek(pos).await
+    }
+
     pub async fn read_exact_at<Buf>(
         &self,
         slice: Slice<Buf>,
@@ -195,31 +227,25 @@ impl VirtualFile {
         self.inner.write_all_at(buf, offset, ctx).await
     }
 
-    pub(crate) async fn read_to_string<P: AsRef<Utf8Path>>(
-        path: P,
+    pub async fn write_all<Buf: IoBuf + Send>(
+        &mut self,
+        buf: FullSlice<Buf>,
         ctx: &RequestContext,
-    ) -> std::io::Result<String> {
-        let file = VirtualFile::open(path, ctx).await?; // TODO: open_v2
+    ) -> (FullSlice<Buf>, Result<usize, Error>) {
+        self.inner.write_all(buf, ctx).await
+    }
+
+    async fn read_to_end(&mut self, buf: &mut Vec<u8>, ctx: &RequestContext) -> Result<(), Error> {
+        self.inner.read_to_end(buf, ctx).await
+    }
+
+    pub(crate) async fn read_to_string(
+        &mut self,
+        ctx: &RequestContext,
+    ) -> Result<String, anyhow::Error> {
         let mut buf = Vec::new();
-        let mut tmp = vec![0; 128];
-        let mut pos: u64 = 0;
-        loop {
-            let slice = tmp.slice(..128);
-            let (slice, res) = file.inner.read_at(slice, pos, ctx).await;
-            match res {
-                Ok(0) => break,
-                Ok(n) => {
-                    pos += n as u64;
-                    buf.extend_from_slice(&slice[..n]);
-                }
-                Err(ref e) if e.kind() == std::io::ErrorKind::Interrupted => {}
-                Err(e) => return Err(e),
-            }
-            tmp = slice.into_inner();
-        }
-        String::from_utf8(buf).map_err(|_| {
-            std::io::Error::new(ErrorKind::InvalidData, "file contents are not valid UTF-8")
-        })
+        self.read_to_end(&mut buf, ctx).await?;
+        Ok(String::from_utf8(buf)?)
     }
 }
 
@@ -266,6 +292,9 @@ pub struct VirtualFileInner {
     /// belongs to a different VirtualFile.
     handle: RwLock<SlotHandle>,
 
+    /// Current file position
+    pos: u64,
+
     /// File path and options to use to open it.
     ///
     /// Note: this only contains the options needed to re-open it. For example,
@@ -530,7 +559,21 @@ impl VirtualFileInner {
         path: P,
         ctx: &RequestContext,
     ) -> Result<VirtualFileInner, std::io::Error> {
-        Self::open_with_options(path.as_ref(), OpenOptions::new().read(true).clone(), ctx).await
+        Self::open_with_options(path.as_ref(), OpenOptions::new().read(true), ctx).await
+    }
+
+    /// Create a new file for writing. If the file exists, it will be truncated.
+    /// Like File::create.
+    pub async fn create<P: AsRef<Utf8Path>>(
+        path: P,
+        ctx: &RequestContext,
+    ) -> Result<VirtualFileInner, std::io::Error> {
+        Self::open_with_options(
+            path.as_ref(),
+            OpenOptions::new().write(true).create(true).truncate(true),
+            ctx,
+        )
+        .await
     }
 
     /// Open a file with given options.
@@ -540,7 +583,7 @@ impl VirtualFileInner {
     /// on the first time. Make sure that's sane!
     pub async fn open_with_options<P: AsRef<Utf8Path>>(
         path: P,
-        open_options: OpenOptions,
+        open_options: &OpenOptions,
         _ctx: &RequestContext,
     ) -> Result<VirtualFileInner, std::io::Error> {
         let path = path.as_ref();
@@ -565,6 +608,7 @@ impl VirtualFileInner {
 
         let vfile = VirtualFileInner {
             handle: RwLock::new(handle),
+            pos: 0,
             path: path.to_owned(),
             open_options: reopen_options,
         };
@@ -631,13 +675,6 @@ impl VirtualFileInner {
         })
     }
 
-    pub async fn set_len(&self, len: u64, _ctx: &RequestContext) -> Result<(), Error> {
-        with_file!(self, StorageIoOperation::SetLen, |file_guard| {
-            let (_file_guard, res) = io_engine::get().set_len(file_guard, len).await;
-            res.maybe_fatal_err("set_len")
-        })
-    }
-
     /// Helper function internal to `VirtualFile` that looks up the underlying File,
     /// opens it and evicts some other File if necessary. The passed parameter is
     /// assumed to be a function available for the physical `File`.
@@ -705,6 +742,38 @@ impl VirtualFileInner {
         })
     }
 
+    pub fn remove(self) {
+        let path = self.path.clone();
+        drop(self);
+        std::fs::remove_file(path).expect("failed to remove the virtual file");
+    }
+
+    pub async fn seek(&mut self, pos: SeekFrom) -> Result<u64, Error> {
+        match pos {
+            SeekFrom::Start(offset) => {
+                self.pos = offset;
+            }
+            SeekFrom::End(offset) => {
+                self.pos = with_file!(self, StorageIoOperation::Seek, |mut file_guard| file_guard
+                    .with_std_file_mut(|std_file| std_file.seek(SeekFrom::End(offset))))?
+            }
+            SeekFrom::Current(offset) => {
+                let pos = self.pos as i128 + offset as i128;
+                if pos < 0 {
+                    return Err(Error::new(
+                        ErrorKind::InvalidInput,
+                        "offset would be negative",
+                    ));
+                }
+                if pos > u64::MAX as i128 {
+                    return Err(Error::new(ErrorKind::InvalidInput, "offset overflow"));
+                }
+                self.pos = pos as u64;
+            }
+        }
+        Ok(self.pos)
+    }
+
     /// Read the file contents in range `offset..(offset + slice.bytes_total())` into `slice[0..slice.bytes_total()]`.
     ///
     /// The returned `Slice<Buf>` is equivalent to the input `slice`, i.e., it's the same view into the same buffer.
@@ -788,7 +857,59 @@ impl VirtualFileInner {
         (restore(buf), Ok(()))
     }
 
-    pub(super) async fn read_at<Buf>(
+    /// Writes `buf` to the file at the current offset.
+    ///
+    /// Panics if there is an uninitialized range in `buf`, as that is most likely a bug in the caller.
+    pub async fn write_all<Buf: IoBuf + Send>(
+        &mut self,
+        buf: FullSlice<Buf>,
+        ctx: &RequestContext,
+    ) -> (FullSlice<Buf>, Result<usize, Error>) {
+        let buf = buf.into_raw_slice();
+        let bounds = buf.bounds();
+        let restore =
+            |buf: Slice<_>| FullSlice::must_new(Slice::from_buf_bounds(buf.into_inner(), bounds));
+        let nbytes = buf.len();
+        let mut buf = buf;
+        while !buf.is_empty() {
+            let (tmp, res) = self.write(FullSlice::must_new(buf), ctx).await;
+            buf = tmp.into_raw_slice();
+            match res {
+                Ok(0) => {
+                    return (
+                        restore(buf),
+                        Err(Error::new(
+                            std::io::ErrorKind::WriteZero,
+                            "failed to write whole buffer",
+                        )),
+                    );
+                }
+                Ok(n) => {
+                    buf = buf.slice(n..);
+                }
+                Err(ref e) if e.kind() == std::io::ErrorKind::Interrupted => {}
+                Err(e) => return (restore(buf), Err(e)),
+            }
+        }
+        (restore(buf), Ok(nbytes))
+    }
+
+    async fn write<B: IoBuf + Send>(
+        &mut self,
+        buf: FullSlice<B>,
+        ctx: &RequestContext,
+    ) -> (FullSlice<B>, Result<usize, std::io::Error>) {
+        let pos = self.pos;
+        let (buf, res) = self.write_at(buf, pos, ctx).await;
+        let n = match res {
+            Ok(n) => n,
+            Err(e) => return (buf, Err(e)),
+        };
+        self.pos += n as u64;
+        (buf, Ok(n))
+    }
+
+    pub(crate) async fn read_at<Buf>(
         &self,
         buf: tokio_epoll_uring::Slice<Buf>,
         offset: u64,
@@ -816,11 +937,23 @@ impl VirtualFileInner {
         })
     }
 
+    /// The function aborts the process if the error is fatal.
     async fn write_at<B: IoBuf + Send>(
         &self,
         buf: FullSlice<B>,
         offset: u64,
         ctx: &RequestContext,
+    ) -> (FullSlice<B>, Result<usize, Error>) {
+        let (slice, result) = self.write_at_inner(buf, offset, ctx).await;
+        let result = result.maybe_fatal_err("write_at");
+        (slice, result)
+    }
+
+    async fn write_at_inner<B: IoBuf + Send>(
+        &self,
+        buf: FullSlice<B>,
+        offset: u64,
+        ctx: &RequestContext,
     ) -> (FullSlice<B>, Result<usize, Error>) {
         let file_guard = match self.lock_file().await {
             Ok(file_guard) => file_guard,
@@ -829,13 +962,30 @@ impl VirtualFileInner {
         observe_duration!(StorageIoOperation::Write, {
             let ((_file_guard, buf), result) =
                 io_engine::get().write_at(file_guard, offset, buf).await;
-            let result = result.maybe_fatal_err("write_at");
             if let Ok(size) = result {
                 ctx.io_size_metrics().write.add(size.into_u64());
             }
             (buf, result)
         })
     }
+
+    async fn read_to_end(&mut self, buf: &mut Vec<u8>, ctx: &RequestContext) -> Result<(), Error> {
+        let mut tmp = vec![0; 128];
+        loop {
+            let slice = tmp.slice(..128);
+            let (slice, res) = self.read_at(slice, self.pos, ctx).await;
+            match res {
+                Ok(0) => return Ok(()),
+                Ok(n) => {
+                    self.pos += n as u64;
+                    buf.extend_from_slice(&slice[..n]);
+                }
+                Err(ref e) if e.kind() == std::io::ErrorKind::Interrupted => {}
+                Err(e) => return Err(e),
+            }
+            tmp = slice.into_inner();
+        }
+    }
 }
 
 // Adapted from https://doc.rust-lang.org/1.72.0/src/std/os/unix/fs.rs.html#117-135
@@ -1050,6 +1200,19 @@ impl FileGuard {
         let _ = file.into_raw_fd();
         res
     }
+    /// Soft deprecation: we'll move VirtualFile to async APIs and remove this function eventually.
+    fn with_std_file_mut<F, R>(&mut self, with: F) -> R
+    where
+        F: FnOnce(&mut File) -> R,
+    {
+        // SAFETY:
+        // - lifetime of the fd: `file` doesn't outlive the OwnedFd stored in `self`.
+        // - &mut usage below: `self` is `&mut`, hence this call is the only task/thread that has control over the underlying fd
+        let mut file = unsafe { File::from_raw_fd(self.as_ref().as_raw_fd()) };
+        let res = with(&mut file);
+        let _ = file.into_raw_fd();
+        res
+    }
 }
 
 impl tokio_epoll_uring::IoFd for FileGuard {
@@ -1139,9 +1302,6 @@ impl OwnedAsyncWriter for VirtualFile {
     ) -> (FullSlice<Buf>, std::io::Result<()>) {
         VirtualFile::write_all_at(self, buf, offset, ctx).await
     }
-    async fn set_len(&self, len: u64, ctx: &RequestContext) -> std::io::Result<()> {
-        VirtualFile::set_len(self, len, ctx).await
-    }
 }
 
 impl OpenFiles {
@@ -1206,9 +1366,10 @@ pub(crate) type IoBuffer = AlignedBuffer<ConstAlign<{ get_io_buffer_alignment()
 pub(crate) type IoPageSlice<'a> =
     AlignedSlice<'a, PAGE_SZ, ConstAlign<{ get_io_buffer_alignment() }>>;
 
-static IO_MODE: LazyLock<AtomicU8> = LazyLock::new(|| AtomicU8::new(IoMode::preferred() as u8));
+static IO_MODE: once_cell::sync::Lazy<AtomicU8> =
+    once_cell::sync::Lazy::new(|| AtomicU8::new(IoMode::preferred() as u8));
 
-pub fn set_io_mode(mode: IoMode) {
+pub(crate) fn set_io_mode(mode: IoMode) {
     IO_MODE.store(mode as u8, std::sync::atomic::Ordering::Relaxed);
 }
 
@@ -1220,6 +1381,7 @@ static SYNC_MODE: AtomicU8 = AtomicU8::new(SyncMode::Sync as u8);
 
 #[cfg(test)]
 mod tests {
+    use std::io::Write;
     use std::os::unix::fs::FileExt;
     use std::sync::Arc;
 
@@ -1272,6 +1434,43 @@ mod tests {
                 MaybeVirtualFile::File(file) => file.write_all_at(&buf[..], offset),
             }
         }
+        async fn seek(&mut self, pos: SeekFrom) -> Result<u64, Error> {
+            match self {
+                MaybeVirtualFile::VirtualFile(file) => file.seek(pos).await,
+                MaybeVirtualFile::File(file) => file.seek(pos),
+            }
+        }
+        async fn write_all<Buf: IoBuf + Send>(
+            &mut self,
+            buf: FullSlice<Buf>,
+            ctx: &RequestContext,
+        ) -> Result<(), Error> {
+            match self {
+                MaybeVirtualFile::VirtualFile(file) => {
+                    let (_buf, res) = file.write_all(buf, ctx).await;
+                    res.map(|_| ())
+                }
+                MaybeVirtualFile::File(file) => file.write_all(&buf[..]),
+            }
+        }
+
+        // Helper function to slurp contents of a file, starting at the current position,
+        // into a string
+        async fn read_string(&mut self, ctx: &RequestContext) -> Result<String, Error> {
+            use std::io::Read;
+            let mut buf = String::new();
+            match self {
+                MaybeVirtualFile::VirtualFile(file) => {
+                    let mut buf = Vec::new();
+                    file.read_to_end(&mut buf, ctx).await?;
+                    return Ok(String::from_utf8(buf).unwrap());
+                }
+                MaybeVirtualFile::File(file) => {
+                    file.read_to_string(&mut buf)?;
+                }
+            }
+            Ok(buf)
+        }
 
         // Helper function to slurp a portion of a file into a string
         async fn read_string_at(
@@ -1307,7 +1506,7 @@ mod tests {
                 opts: OpenOptions,
                 ctx: &RequestContext,
             ) -> Result<MaybeVirtualFile, anyhow::Error> {
-                let vf = VirtualFile::open_with_options_v2(&path, &opts, ctx).await?;
+                let vf = VirtualFile::open_with_options(&path, &opts, ctx).await?;
                 Ok(MaybeVirtualFile::VirtualFile(vf))
             }
         }
@@ -1367,23 +1566,48 @@ mod tests {
         .await?;
 
         file_a
-            .write_all_at(IoBuffer::from(b"foobar").slice_len(), 0, &ctx)
+            .write_all(b"foobar".to_vec().slice_len(), &ctx)
             .await?;
 
         // cannot read from a file opened in write-only mode
-        let _ = file_a.read_string_at(0, 1, &ctx).await.unwrap_err();
+        let _ = file_a.read_string(&ctx).await.unwrap_err();
 
         // Close the file and re-open for reading
         let mut file_a = A::open(path_a, OpenOptions::new().read(true).to_owned(), &ctx).await?;
 
         // cannot write to a file opened in read-only mode
         let _ = file_a
-            .write_all_at(IoBuffer::from(b"bar").slice_len(), 0, &ctx)
+            .write_all(b"bar".to_vec().slice_len(), &ctx)
             .await
             .unwrap_err();
 
         // Try simple read
-        assert_eq!("foobar", file_a.read_string_at(0, 6, &ctx).await?);
+        assert_eq!("foobar", file_a.read_string(&ctx).await?);
+
+        // It's positioned at the EOF now.
+        assert_eq!("", file_a.read_string(&ctx).await?);
+
+        // Test seeks.
+        assert_eq!(file_a.seek(SeekFrom::Start(1)).await?, 1);
+        assert_eq!("oobar", file_a.read_string(&ctx).await?);
+
+        assert_eq!(file_a.seek(SeekFrom::End(-2)).await?, 4);
+        assert_eq!("ar", file_a.read_string(&ctx).await?);
+
+        assert_eq!(file_a.seek(SeekFrom::Start(1)).await?, 1);
+        assert_eq!(file_a.seek(SeekFrom::Current(2)).await?, 3);
+        assert_eq!("bar", file_a.read_string(&ctx).await?);
+
+        assert_eq!(file_a.seek(SeekFrom::Current(-5)).await?, 1);
+        assert_eq!("oobar", file_a.read_string(&ctx).await?);
+
+        // Test erroneous seeks to before byte 0
+        file_a.seek(SeekFrom::End(-7)).await.unwrap_err();
+        assert_eq!(file_a.seek(SeekFrom::Start(1)).await?, 1);
+        file_a.seek(SeekFrom::Current(-2)).await.unwrap_err();
+
+        // the erroneous seek should have left the position unchanged
+        assert_eq!("oobar", file_a.read_string(&ctx).await?);
 
         // Create another test file, and try FileExt functions on it.
         let path_b = testdir.join("file_b");
@@ -1409,6 +1633,9 @@ mod tests {
 
         // Open a lot of files, enough to cause some evictions. (Or to be precise,
         // open the same file many times. The effect is the same.)
+        //
+        // leave file_a positioned at offset 1 before we start
+        assert_eq!(file_a.seek(SeekFrom::Start(1)).await?, 1);
 
         let mut vfiles = Vec::new();
         for _ in 0..100 {
@@ -1418,7 +1645,7 @@ mod tests {
                 &ctx,
             )
             .await?;
-            assert_eq!("FOOBAR", vfile.read_string_at(0, 6, &ctx).await?);
+            assert_eq!("FOOBAR", vfile.read_string(&ctx).await?);
             vfiles.push(vfile);
         }
 
@@ -1426,8 +1653,8 @@ mod tests {
         assert!(vfiles.len() > TEST_MAX_FILE_DESCRIPTORS * 2);
 
         // The underlying file descriptor for 'file_a' should be closed now. Try to read
-        // from it again.
-        assert_eq!("foobar", file_a.read_string_at(0, 6, &ctx).await?);
+        // from it again. We left the file positioned at offset 1 above.
+        assert_eq!("oobar", file_a.read_string(&ctx).await?);
 
         // Check that all the other FDs still work too. Use them in random order for
         // good measure.
@@ -1466,7 +1693,7 @@ mod tests {
         for _ in 0..VIRTUAL_FILES {
             let f = VirtualFileInner::open_with_options(
                 &test_file_path,
-                OpenOptions::new().read(true).clone(),
+                OpenOptions::new().read(true),
                 &ctx,
             )
             .await?;
@@ -1521,7 +1748,7 @@ mod tests {
             .await
             .unwrap();
         let mut file = MaybeVirtualFile::from(VirtualFile::open(&path, &ctx).await.unwrap());
-        let post = file.read_string_at(0, 3, &ctx).await.unwrap();
+        let post = file.read_string(&ctx).await.unwrap();
         assert_eq!(post, "foo");
         assert!(!tmp_path.exists());
         drop(file);
@@ -1530,7 +1757,7 @@ mod tests {
             .await
             .unwrap();
         let mut file = MaybeVirtualFile::from(VirtualFile::open(&path, &ctx).await.unwrap());
-        let post = file.read_string_at(0, 3, &ctx).await.unwrap();
+        let post = file.read_string(&ctx).await.unwrap();
         assert_eq!(post, "bar");
         assert!(!tmp_path.exists());
         drop(file);
@@ -1555,7 +1782,7 @@ mod tests {
             .unwrap();
 
         let mut file = MaybeVirtualFile::from(VirtualFile::open(&path, &ctx).await.unwrap());
-        let post = file.read_string_at(0, 3, &ctx).await.unwrap();
+        let post = file.read_string(&ctx).await.unwrap();
         assert_eq!(post, "foo");
         assert!(!tmp_path.exists());
         drop(file);

pageserver/src/virtual_file/io_engine.rs

@@ -209,27 +209,6 @@ impl IoEngine {
             }
         }
     }
-
-    pub(super) async fn set_len(
-        &self,
-        file_guard: FileGuard,
-        len: u64,
-    ) -> (FileGuard, std::io::Result<()>) {
-        match self {
-            IoEngine::NotSet => panic!("not initialized"),
-            IoEngine::StdFs => {
-                let res = file_guard.with_std_file(|std_file| std_file.set_len(len));
-                (file_guard, res)
-            }
-            #[cfg(target_os = "linux")]
-            IoEngine::TokioEpollUring => {
-                // TODO: ftruncate op for tokio-epoll-uring
-                let res = file_guard.with_std_file(|std_file| std_file.set_len(len));
-                (file_guard, res)
-            }
-        }
-    }
-
     pub(super) async fn write_at<B: IoBuf + Send>(
         &self,
         file_guard: FileGuard,

pageserver/src/virtual_file/open_options.rs

@@ -6,12 +6,7 @@ use std::path::Path;
 use super::io_engine::IoEngine;
 
 #[derive(Debug, Clone)]
-pub struct OpenOptions {
-    write: bool,
-    inner: Inner,
-}
-#[derive(Debug, Clone)]
-enum Inner {
+pub enum OpenOptions {
     StdFs(std::fs::OpenOptions),
     #[cfg(target_os = "linux")]
     TokioEpollUring(tokio_epoll_uring::ops::open_at::OpenOptions),
@@ -19,17 +14,13 @@ enum Inner {
 
 impl Default for OpenOptions {
     fn default() -> Self {
-        let inner = match super::io_engine::get() {
+        match super::io_engine::get() {
             IoEngine::NotSet => panic!("io engine not set"),
-            IoEngine::StdFs => Inner::StdFs(std::fs::OpenOptions::new()),
+            IoEngine::StdFs => Self::StdFs(std::fs::OpenOptions::new()),
             #[cfg(target_os = "linux")]
             IoEngine::TokioEpollUring => {
-                Inner::TokioEpollUring(tokio_epoll_uring::ops::open_at::OpenOptions::new())
+                Self::TokioEpollUring(tokio_epoll_uring::ops::open_at::OpenOptions::new())
             }
-        };
-        Self {
-            write: false,
-            inner,
         }
     }
 }
@@ -39,17 +30,13 @@ impl OpenOptions {
         Self::default()
     }
 
-    pub(super) fn is_write(&self) -> bool {
-        self.write
-    }
-
     pub fn read(&mut self, read: bool) -> &mut OpenOptions {
-        match &mut self.inner {
-            Inner::StdFs(x) => {
+        match self {
+            OpenOptions::StdFs(x) => {
                 let _ = x.read(read);
             }
             #[cfg(target_os = "linux")]
-            Inner::TokioEpollUring(x) => {
+            OpenOptions::TokioEpollUring(x) => {
                 let _ = x.read(read);
             }
         }
@@ -57,13 +44,12 @@ impl OpenOptions {
     }
 
     pub fn write(&mut self, write: bool) -> &mut OpenOptions {
-        self.write = write;
-        match &mut self.inner {
-            Inner::StdFs(x) => {
+        match self {
+            OpenOptions::StdFs(x) => {
                 let _ = x.write(write);
             }
             #[cfg(target_os = "linux")]
-            Inner::TokioEpollUring(x) => {
+            OpenOptions::TokioEpollUring(x) => {
                 let _ = x.write(write);
             }
         }
@@ -71,12 +57,12 @@ impl OpenOptions {
     }
 
     pub fn create(&mut self, create: bool) -> &mut OpenOptions {
-        match &mut self.inner {
-            Inner::StdFs(x) => {
+        match self {
+            OpenOptions::StdFs(x) => {
                 let _ = x.create(create);
             }
             #[cfg(target_os = "linux")]
-            Inner::TokioEpollUring(x) => {
+            OpenOptions::TokioEpollUring(x) => {
                 let _ = x.create(create);
             }
         }
@@ -84,12 +70,12 @@ impl OpenOptions {
     }
 
     pub fn create_new(&mut self, create_new: bool) -> &mut OpenOptions {
-        match &mut self.inner {
-            Inner::StdFs(x) => {
+        match self {
+            OpenOptions::StdFs(x) => {
                 let _ = x.create_new(create_new);
             }
             #[cfg(target_os = "linux")]
-            Inner::TokioEpollUring(x) => {
+            OpenOptions::TokioEpollUring(x) => {
                 let _ = x.create_new(create_new);
             }
         }
@@ -97,12 +83,12 @@ impl OpenOptions {
     }
 
     pub fn truncate(&mut self, truncate: bool) -> &mut OpenOptions {
-        match &mut self.inner {
-            Inner::StdFs(x) => {
+        match self {
+            OpenOptions::StdFs(x) => {
                 let _ = x.truncate(truncate);
             }
             #[cfg(target_os = "linux")]
-            Inner::TokioEpollUring(x) => {
+            OpenOptions::TokioEpollUring(x) => {
                 let _ = x.truncate(truncate);
             }
         }
@@ -110,10 +96,10 @@ impl OpenOptions {
     }
 
     pub(in crate::virtual_file) async fn open(&self, path: &Path) -> std::io::Result<OwnedFd> {
-        match &self.inner {
-            Inner::StdFs(x) => x.open(path).map(|file| file.into()),
+        match self {
+            OpenOptions::StdFs(x) => x.open(path).map(|file| file.into()),
             #[cfg(target_os = "linux")]
-            Inner::TokioEpollUring(x) => {
+            OpenOptions::TokioEpollUring(x) => {
                 let system = super::io_engine::tokio_epoll_uring_ext::thread_local_system().await;
                 system.open(path, x).await.map_err(|e| match e {
                     tokio_epoll_uring::Error::Op(e) => e,
@@ -128,12 +114,12 @@ impl OpenOptions {
 
 impl std::os::unix::prelude::OpenOptionsExt for OpenOptions {
     fn mode(&mut self, mode: u32) -> &mut OpenOptions {
-        match &mut self.inner {
-            Inner::StdFs(x) => {
+        match self {
+            OpenOptions::StdFs(x) => {
                 let _ = x.mode(mode);
             }
             #[cfg(target_os = "linux")]
-            Inner::TokioEpollUring(x) => {
+            OpenOptions::TokioEpollUring(x) => {
                 let _ = x.mode(mode);
             }
         }
@@ -141,12 +127,12 @@ impl std::os::unix::prelude::OpenOptionsExt for OpenOptions {
     }
 
     fn custom_flags(&mut self, flags: i32) -> &mut OpenOptions {
-        match &mut self.inner {
-            Inner::StdFs(x) => {
+        match self {
+            OpenOptions::StdFs(x) => {
                 let _ = x.custom_flags(flags);
             }
             #[cfg(target_os = "linux")]
-            Inner::TokioEpollUring(x) => {
+            OpenOptions::TokioEpollUring(x) => {
                 let _ = x.custom_flags(flags);
             }
         }