temporary debugging of test_pgdata_import_smoke

## Problem
We have to test the extensions, shipped with Neon for compatibility
before the upgrade.
## Summary of changes
Added the test for compatibility with the upgraded extensions.

.github/actionlint.yml

@@ -4,6 +4,7 @@ self-hosted-runner:
     - large
     - large-arm64
     - small
+    - small-metal
     - small-arm64
     - us-east-2
 config-variables:

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

@@ -17,6 +17,31 @@ inputs:
   compute_units:
     description: '[Min, Max] compute units'
     default: '[1, 1]'
+  # settings below only needed if you want the project to be sharded from the beginning
+  shard_split_project:
+    description: 'by default new projects are not shard-split, specify true to shard-split'
+    required: false
+    default: 'false'
+  admin_api_key:
+    description: 'Admin API Key needed for shard-splitting. Must be specified if shard_split_project is true'
+    required: false
+  shard_count:
+    description: 'Number of shards to split the project into, only applies if shard_split_project is true'
+    required: false
+    default: '8'
+  stripe_size:
+    description: 'Stripe size, optional, in 8kiB pages.  e.g. set 2048 for 16MB stripes. Default is 128 MiB, only applies if shard_split_project is true'
+    required: false
+    default: '32768'
+  psql_path:
+    description: 'Path to psql binary - it is caller responsibility to provision the psql binary'
+    required: false
+    default: '/tmp/neon/pg_install/v16/bin/psql'
+  libpq_lib_path:
+    description: 'Path to directory containing libpq library - it is caller responsibility to provision the libpq library'
+    required: false
+    default: '/tmp/neon/pg_install/v16/lib'
+  
 
 outputs:
   dsn:
@@ -63,6 +88,23 @@ runs:
         echo "project_id=${project_id}" >> $GITHUB_OUTPUT
 
         echo "Project ${project_id} has been created"
+
+        if [ "${SHARD_SPLIT_PROJECT}" = "true" ]; then
+          # determine tenant ID
+          TENANT_ID=`${PSQL} ${dsn} -t -A -c "SHOW neon.tenant_id"`
+          
+          echo "Splitting project ${project_id} with tenant_id ${TENANT_ID} into $((SHARD_COUNT)) shards with stripe size $((STRIPE_SIZE))"
+
+          echo "Sending PUT request to https://${API_HOST}/regions/${REGION_ID}/api/v1/admin/storage/proxy/control/v1/tenant/${TENANT_ID}/shard_split"
+          echo "with body {\"new_shard_count\": $((SHARD_COUNT)), \"new_stripe_size\": $((STRIPE_SIZE))}"
+          
+          # we need an ADMIN API KEY to invoke storage controller API for shard splitting (bash -u above checks that the variable is set)
+          curl -X PUT \
+            "https://${API_HOST}/regions/${REGION_ID}/api/v1/admin/storage/proxy/control/v1/tenant/${TENANT_ID}/shard_split" \
+            -H "Accept: application/json" -H "Content-Type: application/json" -H "Authorization: Bearer ${ADMIN_API_KEY}" \
+            -d "{\"new_shard_count\": $SHARD_COUNT, \"new_stripe_size\": $STRIPE_SIZE}"
+        fi
+
       env:
         API_HOST: ${{ inputs.api_host }}
         API_KEY: ${{ inputs.api_key }}
@@ -70,3 +112,9 @@ runs:
         POSTGRES_VERSION: ${{ inputs.postgres_version }}
         MIN_CU: ${{ fromJSON(inputs.compute_units)[0] }}
         MAX_CU: ${{ fromJSON(inputs.compute_units)[1] }}
+        SHARD_SPLIT_PROJECT: ${{ inputs.shard_split_project }}
+        ADMIN_API_KEY: ${{ inputs.admin_api_key }}
+        SHARD_COUNT: ${{ inputs.shard_count }}
+        STRIPE_SIZE: ${{ inputs.stripe_size }}
+        PSQL: ${{ inputs.psql_path }}
+        LD_LIBRARY_PATH: ${{ inputs.libpq_lib_path }}

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

@@ -158,8 +158,6 @@ jobs:
 
       - name: Run cargo build
         run: |
-          PQ_LIB_DIR=$(pwd)/pg_install/v16/lib
-          export PQ_LIB_DIR
           ${cov_prefix} mold -run cargo build $CARGO_FLAGS $CARGO_FEATURES --bins --tests
 
       # Do install *before* running rust tests because they might recompile the
@@ -217,8 +215,6 @@ jobs:
         env:
           NEXTEST_RETRIES: 3
         run: |
-          PQ_LIB_DIR=$(pwd)/pg_install/v16/lib
-          export PQ_LIB_DIR
           LD_LIBRARY_PATH=$(pwd)/pg_install/v17/lib
           export LD_LIBRARY_PATH
 
@@ -229,8 +225,13 @@ jobs:
           ${cov_prefix} cargo nextest run $CARGO_FLAGS $CARGO_FEATURES -E '!package(pageserver)'
 
           # run pageserver tests with different settings
-          for io_engine in std-fs tokio-epoll-uring ; do
-            NEON_PAGESERVER_UNIT_TEST_VIRTUAL_FILE_IOENGINE=$io_engine ${cov_prefix} cargo nextest run $CARGO_FLAGS $CARGO_FEATURES  -E 'package(pageserver)'
+          for get_vectored_concurrent_io in sequential sidecar-task; do
+            for io_engine in std-fs tokio-epoll-uring ; do
+              NEON_PAGESERVER_UNIT_TEST_GET_VECTORED_CONCURRENT_IO=$get_vectored_concurrent_io \
+                NEON_PAGESERVER_UNIT_TEST_VIRTUAL_FILE_IOENGINE=$io_engine \
+                ${cov_prefix} \
+                cargo nextest run $CARGO_FLAGS $CARGO_FEATURES  -E 'package(pageserver)'
+            done
           done
 
           # Run separate tests for real S3
@@ -314,6 +315,7 @@ jobs:
           CHECK_ONDISK_DATA_COMPATIBILITY: nonempty
           BUILD_TAG: ${{ inputs.build-tag }}
           PAGESERVER_VIRTUAL_FILE_IO_ENGINE: tokio-epoll-uring
+          PAGESERVER_GET_VECTORED_CONCURRENT_IO: sidecar-task
           USE_LFC: ${{ matrix.lfc_state == 'with-lfc' && 'true' || 'false' }}
 
       # Temporary disable this step until we figure out why it's so flaky

.github/workflows/build-macos.yml

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

.github/workflows/build_and_test.yml

@@ -242,7 +242,7 @@ jobs:
       statuses: write
       contents: write
       pull-requests: write
-    runs-on: [ self-hosted, small ]
+    runs-on: [ self-hosted, small-metal ]
     container:
       image: ${{ needs.build-build-tools-image.outputs.image }}-bookworm
       credentials:
@@ -786,6 +786,17 @@ jobs:
           username: ${{ secrets.NEON_DOCKERHUB_USERNAME }}
           password: ${{ secrets.NEON_DOCKERHUB_PASSWORD }}
 
+      - name: Get the last compute release tag
+        id: get-last-compute-release-tag
+        env:
+          GH_TOKEN: ${{ secrets.CI_ACCESS_TOKEN }}
+        run: |
+          tag=$(gh api -q '[.[].tag_name | select(startswith("release-compute"))][0]'\
+            -H "Accept: application/vnd.github+json" \
+            -H "X-GitHub-Api-Version: 2022-11-28" \
+            "/repos/${{ github.repository }}/releases")
+          echo tag=${tag} >> ${GITHUB_OUTPUT}
+
       # `neondatabase/neon` contains multiple binaries, all of them use the same input for the version into the same version formatting library.
       # Pick pageserver as currently the only binary with extra "version" features printed in the string to verify.
       # Regular pageserver version string looks like
@@ -817,11 +828,25 @@ jobs:
           TEST_VERSION_ONLY: ${{ matrix.pg_version }}
         run: ./docker-compose/docker_compose_test.sh
 
+      - name: Print logs and clean up docker-compose test
+        if: always()
+        run: |
+          docker compose --profile test-extensions -f ./docker-compose/docker-compose.yml logs || true
+          docker compose --profile test-extensions -f ./docker-compose/docker-compose.yml down
+
+      - name: Test extension upgrade
+        timeout-minutes: 20
+        if: ${{ needs.tag.outputs.build-tag == github.run_id }}
+        env:
+          NEWTAG: ${{ needs.tag.outputs.build-tag }}
+          OLDTAG: ${{ steps.get-last-compute-release-tag.outputs.tag }}
+        run: ./docker-compose/test_extensions_upgrade.sh
+
       - name: Print logs and clean up
         if: always()
         run: |
-          docker compose -f ./docker-compose/docker-compose.yml logs || 0
-          docker compose -f ./docker-compose/docker-compose.yml down
+          docker compose --profile test-extensions -f ./docker-compose/docker-compose.yml logs || true
+          docker compose --profile test-extensions -f ./docker-compose/docker-compose.yml down
 
   promote-images-dev:
     needs: [ check-permissions, tag, vm-compute-node-image, neon-image ]
@@ -859,7 +884,7 @@ jobs:
           done
 
   promote-images-prod:
-    needs: [ check-permissions, tag, test-images, vm-compute-node-image ]
+    needs: [ check-permissions, tag, test-images, promote-images-dev ]
     runs-on: ubuntu-22.04
     if: github.ref_name == 'main' || github.ref_name == 'release' || github.ref_name == 'release-proxy' || github.ref_name == 'release-compute'
 
@@ -1050,6 +1075,7 @@ jobs:
           retries: 5
           script: |
             const tag = "${{ needs.tag.outputs.build-tag }}";
+            const branch = "${{ github.ref_name }}";
 
             try {
               const existingRef = await github.rest.git.getRef({
@@ -1078,12 +1104,6 @@ jobs:
               console.log(`Tag ${tag} created successfully.`);
             }
 
-            // TODO: check how GitHub releases looks for proxy/compute releases and enable them if they're ok
-            if (context.ref !== 'refs/heads/release') {
-              console.log(`GitHub release skipped for ${context.ref}.`);
-              return;
-            }
-
             try {
               const existingRelease = await github.rest.repos.getReleaseByTag({
                 owner: context.repo.owner,
@@ -1098,11 +1118,48 @@ jobs:
               }
 
               console.log(`Release for tag ${tag} does not exist. Creating it...`);
+
+              // Find the PR number using the commit SHA
+              const pullRequests = await github.rest.pulls.list({
+                owner: context.repo.owner,
+                repo: context.repo.repo,
+                state: 'closed',
+                base: branch,
+              });
+
+              const pr = pullRequests.data.find(pr => pr.merge_commit_sha === context.sha);
+              const prNumber = pr ? pr.number : null;
+
+              // Find the previous release on the branch
+              const releases = await github.rest.repos.listReleases({
+                owner: context.repo.owner,
+                repo: context.repo.repo,
+                per_page: 100,
+              });
+
+              const branchReleases = releases.data
+                .filter((release) => {
+                  const regex = new RegExp(`^${branch}-\\d+$`);
+                  return regex.test(release.tag_name) && !release.draft && !release.prerelease;
+                })
+                .sort((a, b) => new Date(b.created_at) - new Date(a.created_at));
+
+              const previousTag = branchReleases.length > 0 ? branchReleases[0].tag_name : null;
+
+              const releaseNotes = [
+                prNumber
+                  ? `Release PR https://github.com/${context.repo.owner}/${context.repo.repo}/pull/${prNumber}.`
+                  : 'Release PR not found.',
+                previousTag
+                  ? `Diff with the previous release https://github.com/${context.repo.owner}/${context.repo.repo}/compare/${previousTag}...${tag}.`
+                  : `No previous release found on branch ${branch}.`,
+              ].join('\n\n');
+
               await github.rest.repos.createRelease({
                 owner: context.repo.owner,
                 repo: context.repo.repo,
                 tag_name: tag,
-                generate_release_notes: true,
+                body: releaseNotes,
               });
               console.log(`Release for tag ${tag} created successfully.`);
             }

.github/workflows/ingest_benchmark.yml

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

.github/workflows/neon_extra_builds.yml

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

.github/workflows/release.yml

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

Cargo.lock

@@ -179,7 +179,7 @@ dependencies = [
  "nom",
  "num-traits",
  "rusticata-macros",
- "thiserror",
+ "thiserror 1.0.69",
  "time",
 ]
 
@@ -718,14 +718,14 @@ dependencies = [
 
 [[package]]
 name = "axum"
-version = "0.7.9"
+version = "0.8.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "edca88bc138befd0323b20752846e6587272d3b03b0343c8ea28a6f819e6e71f"
+checksum = "6d6fd624c75e18b3b4c6b9caf42b1afe24437daaee904069137d8bab077be8b8"
 dependencies = [
- "async-trait",
  "axum-core",
  "base64 0.22.1",
  "bytes",
+ "form_urlencoded",
  "futures-util",
  "http 1.1.0",
  "http-body 1.0.0",
@@ -733,7 +733,7 @@ dependencies = [
  "hyper 1.4.1",
  "hyper-util",
  "itoa",
- "matchit 0.7.0",
+ "matchit",
  "memchr",
  "mime",
  "percent-encoding",
@@ -746,7 +746,7 @@ dependencies = [
  "sha1",
  "sync_wrapper 1.0.1",
  "tokio",
- "tokio-tungstenite 0.24.0",
+ "tokio-tungstenite 0.26.1",
  "tower 0.5.2",
  "tower-layer",
  "tower-service",
@@ -755,11 +755,10 @@ dependencies = [
 
 [[package]]
 name = "axum-core"
-version = "0.4.5"
+version = "0.5.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "09f2bd6146b97ae3359fa0cc6d6b376d9539582c7b4220f041a33ec24c226199"
+checksum = "df1362f362fd16024ae199c1970ce98f9661bf5ef94b9808fee734bc3698b733"
 dependencies = [
- "async-trait",
  "bytes",
  "futures-util",
  "http 1.1.0",
@@ -942,6 +941,18 @@ version = "1.6.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "8c3c1a368f70d6cf7302d78f8f7093da241fb8e8807c05cc9e51a125895a6d5b"
 
+[[package]]
+name = "bb8"
+version = "0.8.6"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d89aabfae550a5c44b43ab941844ffcd2e993cb6900b342debf59e9ea74acdb8"
+dependencies = [
+ "async-trait",
+ "futures-util",
+ "parking_lot 0.12.1",
+ "tokio",
+]
+
 [[package]]
 name = "bcder"
 version = "0.7.4"
@@ -1118,7 +1129,7 @@ dependencies = [
  "log",
  "nix 0.25.1",
  "regex",
- "thiserror",
+ "thiserror 1.0.69",
 ]
 
 [[package]]
@@ -1299,9 +1310,9 @@ dependencies = [
  "serde_with",
  "signal-hook",
  "tar",
- "thiserror",
+ "thiserror 1.0.69",
  "tokio",
- "tokio-postgres",
+ "tokio-postgres 0.7.7",
  "tokio-stream",
  "tokio-util",
  "tower 0.5.2",
@@ -1408,9 +1419,9 @@ dependencies = [
  "serde",
  "serde_json",
  "storage_broker",
- "thiserror",
+ "thiserror 1.0.69",
  "tokio",
- "tokio-postgres",
+ "tokio-postgres 0.7.7",
  "tokio-util",
  "toml",
  "toml_edit",
@@ -1786,11 +1797,24 @@ dependencies = [
  "chrono",
  "diesel_derives",
  "itoa",
- "pq-sys",
- "r2d2",
  "serde_json",
 ]
 
+[[package]]
+name = "diesel-async"
+version = "0.5.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "51a307ac00f7c23f526a04a77761a0519b9f0eb2838ebf5b905a58580095bdcb"
+dependencies = [
+ "async-trait",
+ "bb8",
+ "diesel",
+ "futures-util",
+ "scoped-futures",
+ "tokio",
+ "tokio-postgres 0.7.12",
+]
+
 [[package]]
 name = "diesel_derives"
 version = "2.2.1"
@@ -2239,7 +2263,7 @@ dependencies = [
  "pin-project",
  "rand 0.8.5",
  "sha1",
- "thiserror",
+ "thiserror 1.0.69",
  "tokio",
  "tokio-util",
 ]
@@ -3365,12 +3389,6 @@ dependencies = [
  "regex-automata 0.1.10",
 ]
 
-[[package]]
-name = "matchit"
-version = "0.7.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "b87248edafb776e59e6ee64a79086f65890d3510f2c656c000bf2a7e8a0aea40"
-
 [[package]]
 name = "matchit"
 version = "0.8.4"
@@ -3761,7 +3779,7 @@ dependencies = [
  "serde_json",
  "serde_path_to_error",
  "sha2",
- "thiserror",
+ "thiserror 1.0.69",
  "url",
 ]
 
@@ -3811,7 +3829,7 @@ dependencies = [
  "futures-sink",
  "js-sys",
  "pin-project-lite",
- "thiserror",
+ "thiserror 1.0.69",
  "tracing",
 ]
 
@@ -3843,7 +3861,7 @@ dependencies = [
  "opentelemetry_sdk",
  "prost",
  "reqwest",
- "thiserror",
+ "thiserror 1.0.69",
 ]
 
 [[package]]
@@ -3879,7 +3897,7 @@ dependencies = [
  "percent-encoding",
  "rand 0.8.5",
  "serde_json",
- "thiserror",
+ "thiserror 1.0.69",
  "tokio",
  "tokio-stream",
  "tracing",
@@ -3993,7 +4011,7 @@ dependencies = [
  "remote_storage",
  "serde_json",
  "svg_fmt",
- "thiserror",
+ "thiserror 1.0.69",
  "tokio",
  "tokio-util",
  "utils",
@@ -4042,8 +4060,8 @@ dependencies = [
  "pageserver_compaction",
  "pin-project-lite",
  "postgres",
- "postgres-protocol",
- "postgres-types",
+ "postgres-protocol 0.6.4",
+ "postgres-types 0.2.4",
  "postgres_backend",
  "postgres_connection",
  "postgres_ffi",
@@ -4069,12 +4087,12 @@ dependencies = [
  "strum_macros",
  "sysinfo",
  "tenant_size_model",
- "thiserror",
+ "thiserror 1.0.69",
  "tikv-jemallocator",
  "tokio",
  "tokio-epoll-uring",
  "tokio-io-timeout",
- "tokio-postgres",
+ "tokio-postgres 0.7.7",
  "tokio-stream",
  "tokio-tar",
  "tokio-util",
@@ -4115,7 +4133,7 @@ dependencies = [
  "storage_broker",
  "strum",
  "strum_macros",
- "thiserror",
+ "thiserror 1.0.69",
  "utils",
 ]
 
@@ -4130,9 +4148,9 @@ dependencies = [
  "postgres",
  "reqwest",
  "serde",
- "thiserror",
+ "thiserror 1.0.69",
  "tokio",
- "tokio-postgres",
+ "tokio-postgres 0.7.7",
  "tokio-stream",
  "tokio-util",
  "utils",
@@ -4438,7 +4456,7 @@ dependencies = [
  "futures-util",
  "log",
  "tokio",
- "tokio-postgres",
+ "tokio-postgres 0.7.7",
 ]
 
 [[package]]
@@ -4459,6 +4477,24 @@ dependencies = [
  "stringprep",
 ]
 
+[[package]]
+name = "postgres-protocol"
+version = "0.6.7"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "acda0ebdebc28befa84bee35e651e4c5f09073d668c7aed4cf7e23c3cda84b23"
+dependencies = [
+ "base64 0.22.1",
+ "byteorder",
+ "bytes",
+ "fallible-iterator",
+ "hmac",
+ "md-5",
+ "memchr",
+ "rand 0.8.5",
+ "sha2",
+ "stringprep",
+]
+
 [[package]]
 name = "postgres-protocol2"
 version = "0.1.0"
@@ -4482,7 +4518,18 @@ source = "git+https://github.com/neondatabase/rust-postgres.git?branch=neon#511f
 dependencies = [
  "bytes",
  "fallible-iterator",
- "postgres-protocol",
+ "postgres-protocol 0.6.4",
+]
+
+[[package]]
+name = "postgres-types"
+version = "0.2.8"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f66ea23a2d0e5734297357705193335e0a957696f34bed2f2faefacb2fec336f"
+dependencies = [
+ "bytes",
+ "fallible-iterator",
+ "postgres-protocol 0.6.7",
 ]
 
 [[package]]
@@ -4505,9 +4552,9 @@ dependencies = [
  "rustls 0.23.18",
  "rustls-pemfile 2.1.1",
  "serde",
- "thiserror",
+ "thiserror 1.0.69",
  "tokio",
- "tokio-postgres",
+ "tokio-postgres 0.7.7",
  "tokio-postgres-rustls",
  "tokio-rustls 0.26.0",
  "tokio-util",
@@ -4522,7 +4569,7 @@ dependencies = [
  "itertools 0.10.5",
  "once_cell",
  "postgres",
- "tokio-postgres",
+ "tokio-postgres 0.7.7",
  "url",
 ]
 
@@ -4543,7 +4590,7 @@ dependencies = [
  "pprof",
  "regex",
  "serde",
- "thiserror",
+ "thiserror 1.0.69",
  "tracing",
  "utils",
 ]
@@ -4554,7 +4601,7 @@ version = "0.1.0"
 dependencies = [
  "anyhow",
  "camino",
- "thiserror",
+ "thiserror 1.0.69",
  "tokio",
  "workspace_hack",
 ]
@@ -4587,7 +4634,7 @@ dependencies = [
  "smallvec",
  "symbolic-demangle",
  "tempfile",
- "thiserror",
+ "thiserror 1.0.69",
 ]
 
 [[package]]
@@ -4609,15 +4656,6 @@ version = "0.2.17"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "5b40af805b3121feab8a3c29f04d8ad262fa8e0561883e7653e024ae4479e6de"
 
-[[package]]
-name = "pq-sys"
-version = "0.6.3"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "f6cc05d7ea95200187117196eee9edd0644424911821aeb28a18ce60ea0b8793"
-dependencies = [
- "vcpkg",
-]
-
 [[package]]
 name = "pq_proto"
 version = "0.1.0"
@@ -4625,10 +4663,10 @@ dependencies = [
  "byteorder",
  "bytes",
  "itertools 0.10.5",
- "postgres-protocol",
+ "postgres-protocol 0.6.4",
  "rand 0.8.5",
  "serde",
- "thiserror",
+ "thiserror 1.0.69",
  "tokio",
 ]
 
@@ -4699,7 +4737,7 @@ dependencies = [
  "memchr",
  "parking_lot 0.12.1",
  "procfs",
- "thiserror",
+ "thiserror 1.0.69",
 ]
 
 [[package]]
@@ -4869,11 +4907,11 @@ dependencies = [
  "strum",
  "strum_macros",
  "subtle",
- "thiserror",
+ "thiserror 1.0.69",
  "tikv-jemalloc-ctl",
  "tikv-jemallocator",
  "tokio",
- "tokio-postgres",
+ "tokio-postgres 0.7.7",
  "tokio-postgres2",
  "tokio-rustls 0.26.0",
  "tokio-tungstenite 0.21.0",
@@ -4930,17 +4968,6 @@ dependencies = [
  "proc-macro2",
 ]
 
-[[package]]
-name = "r2d2"
-version = "0.8.10"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "51de85fb3fb6524929c8a2eb85e6b6d363de4e8c48f9e2c2eac4944abc181c93"
-dependencies = [
- "log",
- "parking_lot 0.12.1",
- "scheduled-thread-pool",
-]
-
 [[package]]
 name = "rand"
 version = "0.7.3"
@@ -5277,7 +5304,7 @@ dependencies = [
  "http 1.1.0",
  "reqwest",
  "serde",
- "thiserror",
+ "thiserror 1.0.69",
  "tower-service",
 ]
 
@@ -5297,7 +5324,7 @@ dependencies = [
  "reqwest",
  "reqwest-middleware",
  "retry-policies",
- "thiserror",
+ "thiserror 1.0.69",
  "tokio",
  "tracing",
  "wasm-timer",
@@ -5313,7 +5340,7 @@ dependencies = [
  "async-trait",
  "getrandom 0.2.11",
  "http 1.1.0",
- "matchit 0.8.4",
+ "matchit",
  "opentelemetry",
  "reqwest",
  "reqwest-middleware",
@@ -5672,7 +5699,7 @@ dependencies = [
  "pageserver_api",
  "parking_lot 0.12.1",
  "postgres",
- "postgres-protocol",
+ "postgres-protocol 0.6.4",
  "postgres_backend",
  "postgres_ffi",
  "pprof",
@@ -5692,11 +5719,11 @@ dependencies = [
  "storage_broker",
  "strum",
  "strum_macros",
- "thiserror",
+ "thiserror 1.0.69",
  "tikv-jemallocator",
  "tokio",
  "tokio-io-timeout",
- "tokio-postgres",
+ "tokio-postgres 0.7.7",
  "tokio-stream",
  "tokio-tar",
  "tokio-util",
@@ -5731,7 +5758,7 @@ dependencies = [
  "reqwest",
  "safekeeper_api",
  "serde",
- "thiserror",
+ "thiserror 1.0.69",
  "utils",
  "workspace_hack",
 ]
@@ -5755,12 +5782,12 @@ dependencies = [
 ]
 
 [[package]]
-name = "scheduled-thread-pool"
-version = "0.2.7"
+name = "scoped-futures"
+version = "0.1.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "3cbc66816425a074528352f5789333ecff06ca41b36b0b0efdfbb29edc391a19"
+checksum = "1b24aae2d0636530f359e9d5ef0c04669d11c5e756699b27a6a6d845d8329091"
 dependencies = [
- "parking_lot 0.12.1",
+ "pin-project-lite",
 ]
 
 [[package]]
@@ -5940,7 +5967,7 @@ dependencies = [
  "rand 0.8.5",
  "serde",
  "serde_json",
- "thiserror",
+ "thiserror 1.0.69",
  "time",
  "url",
  "uuid",
@@ -6012,7 +6039,7 @@ checksum = "c7715380eec75f029a4ef7de39a9200e0a63823176b759d055b613f5a87df6a6"
 dependencies = [
  "percent-encoding",
  "serde",
- "thiserror",
+ "thiserror 1.0.69",
 ]
 
 [[package]]
@@ -6174,7 +6201,7 @@ checksum = "adc4e5204eb1910f40f9cfa375f6f05b68c3abac4b6fd879c8ff5e7ae8a0a085"
 dependencies = [
  "num-bigint",
  "num-traits",
- "thiserror",
+ "thiserror 1.0.69",
  "time",
 ]
 
@@ -6295,6 +6322,7 @@ dependencies = [
  "clap",
  "control_plane",
  "diesel",
+ "diesel-async",
  "diesel_migrations",
  "fail",
  "futures",
@@ -6309,16 +6337,16 @@ dependencies = [
  "pageserver_api",
  "pageserver_client",
  "postgres_connection",
- "r2d2",
  "rand 0.8.5",
  "reqwest",
  "routerify",
+ "scoped-futures",
  "scopeguard",
  "serde",
  "serde_json",
  "strum",
  "strum_macros",
- "thiserror",
+ "thiserror 1.0.69",
  "tokio",
  "tokio-util",
  "tracing",
@@ -6365,7 +6393,7 @@ dependencies = [
  "serde_json",
  "storage_controller_client",
  "tokio",
- "tokio-postgres",
+ "tokio-postgres 0.7.7",
  "tokio-postgres-rustls",
  "tokio-stream",
  "tokio-util",
@@ -6610,7 +6638,16 @@ version = "1.0.69"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "b6aaf5339b578ea85b50e080feb250a3e8ae8cfcdff9a461c9ec2904bc923f52"
 dependencies = [
- "thiserror-impl",
+ "thiserror-impl 1.0.69",
+]
+
+[[package]]
+name = "thiserror"
+version = "2.0.11"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d452f284b73e6d76dd36758a0c8684b1d5be31f92b89d07fd5822175732206fc"
+dependencies = [
+ "thiserror-impl 2.0.11",
 ]
 
 [[package]]
@@ -6624,6 +6661,17 @@ dependencies = [
  "syn 2.0.90",
 ]
 
+[[package]]
+name = "thiserror-impl"
+version = "2.0.11"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "26afc1baea8a989337eeb52b6e72a039780ce45c3edfcc9c5b9d112feeb173c2"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn 2.0.90",
+]
+
 [[package]]
 name = "thread_local"
 version = "1.1.7"
@@ -6774,13 +6822,13 @@ dependencies = [
 [[package]]
 name = "tokio-epoll-uring"
 version = "0.1.0"
-source = "git+https://github.com/neondatabase/tokio-epoll-uring.git?branch=main#33e00106a268644d02ba0461bbd64476073b0ee1"
+source = "git+https://github.com/neondatabase/tokio-epoll-uring.git?branch=main#781989bb540a1408b0b93daa1e9d1fa452195497"
 dependencies = [
  "futures",
  "nix 0.26.4",
  "once_cell",
  "scopeguard",
- "thiserror",
+ "thiserror 1.0.69",
  "tokio",
  "tokio-util",
  "tracing",
@@ -6824,13 +6872,39 @@ dependencies = [
  "percent-encoding",
  "phf",
  "pin-project-lite",
- "postgres-protocol",
- "postgres-types",
+ "postgres-protocol 0.6.4",
+ "postgres-types 0.2.4",
  "socket2",
  "tokio",
  "tokio-util",
 ]
 
+[[package]]
+name = "tokio-postgres"
+version = "0.7.12"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3b5d3742945bc7d7f210693b0c58ae542c6fd47b17adbbda0885f3dcb34a6bdb"
+dependencies = [
+ "async-trait",
+ "byteorder",
+ "bytes",
+ "fallible-iterator",
+ "futures-channel",
+ "futures-util",
+ "log",
+ "parking_lot 0.12.1",
+ "percent-encoding",
+ "phf",
+ "pin-project-lite",
+ "postgres-protocol 0.6.7",
+ "postgres-types 0.2.8",
+ "rand 0.8.5",
+ "socket2",
+ "tokio",
+ "tokio-util",
+ "whoami",
+]
+
 [[package]]
 name = "tokio-postgres-rustls"
 version = "0.12.0"
@@ -6840,7 +6914,7 @@ dependencies = [
  "ring",
  "rustls 0.23.18",
  "tokio",
- "tokio-postgres",
+ "tokio-postgres 0.7.7",
  "tokio-rustls 0.26.0",
  "x509-certificate",
 ]
@@ -6937,14 +7011,14 @@ dependencies = [
 
 [[package]]
 name = "tokio-tungstenite"
-version = "0.24.0"
+version = "0.26.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "edc5f74e248dc973e0dbb7b74c7e0d6fcc301c694ff50049504004ef4d0cdcd9"
+checksum = "be4bf6fecd69fcdede0ec680aaf474cdab988f9de6bc73d3758f0160e3b7025a"
 dependencies = [
  "futures-util",
  "log",
  "tokio",
- "tungstenite 0.24.0",
+ "tungstenite 0.26.1",
 ]
 
 [[package]]
@@ -7004,12 +7078,9 @@ version = "0.12.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "877c5b330756d856ffcc4553ab34a5684481ade925ecc54bcd1bf02b1d0d4d52"
 dependencies = [
- "async-stream",
  "async-trait",
- "axum",
  "base64 0.22.1",
  "bytes",
- "h2 0.4.4",
  "http 1.1.0",
  "http-body 1.0.0",
  "http-body-util",
@@ -7021,7 +7092,6 @@ dependencies = [
  "prost",
  "rustls-native-certs 0.8.0",
  "rustls-pemfile 2.1.1",
- "socket2",
  "tokio",
  "tokio-rustls 0.26.0",
  "tokio-stream",
@@ -7258,16 +7328,16 @@ dependencies = [
  "log",
  "rand 0.8.5",
  "sha1",
- "thiserror",
+ "thiserror 1.0.69",
  "url",
  "utf-8",
 ]
 
 [[package]]
 name = "tungstenite"
-version = "0.24.0"
+version = "0.26.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "18e5b8366ee7a95b16d32197d0b2604b43a0be89dc5fac9f8e96ccafbaedda8a"
+checksum = "413083a99c579593656008130e29255e54dcaae495be556cc26888f211648c24"
 dependencies = [
  "byteorder",
  "bytes",
@@ -7277,7 +7347,7 @@ dependencies = [
  "log",
  "rand 0.8.5",
  "sha1",
- "thiserror",
+ "thiserror 2.0.11",
  "utf-8",
 ]
 
@@ -7373,7 +7443,7 @@ dependencies = [
 [[package]]
 name = "uring-common"
 version = "0.1.0"
-source = "git+https://github.com/neondatabase/tokio-epoll-uring.git?branch=main#33e00106a268644d02ba0461bbd64476073b0ee1"
+source = "git+https://github.com/neondatabase/tokio-epoll-uring.git?branch=main#781989bb540a1408b0b93daa1e9d1fa452195497"
 dependencies = [
  "bytes",
  "io-uring",
@@ -7472,7 +7542,7 @@ dependencies = [
  "signal-hook",
  "strum",
  "strum_macros",
- "thiserror",
+ "thiserror 1.0.69",
  "tokio",
  "tokio-stream",
  "tokio-tar",
@@ -7502,12 +7572,6 @@ version = "0.1.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "830b7e5d4d90034032940e4ace0d9a9a057e7a45cd94e6c007832e39edb82f6d"
 
-[[package]]
-name = "vcpkg"
-version = "0.2.15"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "accd4ea62f7bb7a82fe23066fb0957d48ef677f6eeb8215f372f52e48bb32426"
-
 [[package]]
 name = "version_check"
 version = "0.9.4"
@@ -7527,7 +7591,7 @@ dependencies = [
  "serde_json",
  "sysinfo",
  "tokio",
- "tokio-postgres",
+ "tokio-postgres 0.7.7",
  "tokio-util",
  "tracing",
  "tracing-subscriber",
@@ -7578,11 +7642,10 @@ dependencies = [
  "remote_storage",
  "serde",
  "serde_json",
- "thiserror",
+ "thiserror 1.0.69",
  "tikv-jemallocator",
  "tokio",
  "tokio-util",
- "tonic",
  "tonic-build",
  "tracing",
  "utils",
@@ -7991,8 +8054,6 @@ version = "0.1.0"
 dependencies = [
  "ahash",
  "anyhow",
- "axum",
- "axum-core",
  "base64 0.13.1",
  "base64 0.21.1",
  "base64ct",
@@ -8073,7 +8134,6 @@ dependencies = [
  "toml_edit",
  "tonic",
  "tower 0.4.13",
- "tower 0.5.2",
  "tracing",
  "tracing-core",
  "url",
@@ -8111,7 +8171,7 @@ dependencies = [
  "ring",
  "signature 2.2.0",
  "spki 0.7.3",
- "thiserror",
+ "thiserror 1.0.69",
  "zeroize",
 ]
 
@@ -8128,7 +8188,7 @@ dependencies = [
  "nom",
  "oid-registry",
  "rusticata-macros",
- "thiserror",
+ "thiserror 1.0.69",
  "time",
 ]
 

Cargo.toml

@@ -65,7 +65,7 @@ aws-smithy-types = "1.2"
 aws-credential-types = "1.2.0"
 aws-sigv4 = { version = "1.2", features = ["sign-http"] }
 aws-types = "1.3"
-axum = { version = "0.7.9", features = ["ws"] }
+axum = { version = "0.8.1", features = ["ws"] }
 base64 = "0.13.0"
 bincode = "1.3"
 bindgen = "0.70"
@@ -187,7 +187,7 @@ tokio-tar = "0.3"
 tokio-util = { version = "0.7.10", features = ["io", "rt"] }
 toml = "0.8"
 toml_edit = "0.22"
-tonic = {version = "0.12.3", features = ["tls", "tls-roots"]}
+tonic = {version = "0.12.3", default-features = false, features = ["channel", "tls", "tls-roots"]}
 tower = { version = "0.5.2", default-features = false }
 tower-http = { version = "0.6.2", features = ["request-id", "trace"] }
 tower-service = "0.3.3"

Dockerfile

@@ -45,7 +45,7 @@ COPY --chown=nonroot . .
 
 ARG ADDITIONAL_RUSTFLAGS
 RUN set -e \
-    && PQ_LIB_DIR=$(pwd)/pg_install/v${STABLE_PG_VERSION}/lib RUSTFLAGS="-Clinker=clang -Clink-arg=-fuse-ld=mold -Clink-arg=-Wl,--no-rosegment -Cforce-frame-pointers=yes ${ADDITIONAL_RUSTFLAGS}" cargo build \
+    && RUSTFLAGS="-Clinker=clang -Clink-arg=-fuse-ld=mold -Clink-arg=-Wl,--no-rosegment -Cforce-frame-pointers=yes ${ADDITIONAL_RUSTFLAGS}" cargo build \
       --bin pg_sni_router  \
       --bin pageserver  \
       --bin pagectl  \

Makefile

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

README.md

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

compute/compute-node.Dockerfile

@@ -67,6 +67,9 @@ RUN cd postgres && \
     # Enable some of contrib extensions
     echo 'trusted = true' >> /usr/local/pgsql/share/extension/autoinc.control && \
     echo 'trusted = true' >> /usr/local/pgsql/share/extension/dblink.control && \
+    echo 'trusted = true' >> /usr/local/pgsql/share/extension/postgres_fdw.control && \
+    file=/usr/local/pgsql/share/extension/postgres_fdw--1.0.sql && [ -e $file ] && \
+    echo 'GRANT USAGE ON FOREIGN DATA WRAPPER postgres_fdw TO neon_superuser;' >> $file && \
     echo 'trusted = true' >> /usr/local/pgsql/share/extension/bloom.control && \
     echo 'trusted = true' >> /usr/local/pgsql/share/extension/earthdistance.control && \
     echo 'trusted = true' >> /usr/local/pgsql/share/extension/insert_username.control && \
@@ -360,6 +363,8 @@ COPY compute/patches/pgvector.patch /pgvector.patch
 RUN wget https://github.com/pgvector/pgvector/archive/refs/tags/v0.8.0.tar.gz -O pgvector.tar.gz && \
     echo "867a2c328d4928a5a9d6f052cd3bc78c7d60228a9b914ad32aa3db88e9de27b0 pgvector.tar.gz" | sha256sum --check && \
     mkdir pgvector-src && cd pgvector-src && tar xzf ../pgvector.tar.gz --strip-components=1 -C . && \
+    wget https://github.com/pgvector/pgvector/raw/refs/tags/v0.7.4/sql/vector.sql -O ./sql/vector--0.7.4.sql && \
+    echo "10218d05dc02299562252a9484775178b14a1d8edb92a2d1672ef488530f7778 ./sql/vector--0.7.4.sql" | sha256sum --check && \
     patch -p1 < /pgvector.patch && \
     make -j $(getconf _NPROCESSORS_ONLN) OPTFLAGS="" && \
     make -j $(getconf _NPROCESSORS_ONLN) OPTFLAGS="" install && \
@@ -1261,11 +1266,12 @@ RUN set -e \
 
 #########################################################################################
 #
-# Layers "postgres-exporter" and "sql-exporter"
+# Layers "postgres-exporter", "pgbouncer-exporter", and "sql-exporter"
 #
 #########################################################################################
 
 FROM quay.io/prometheuscommunity/postgres-exporter:v0.16.0 AS postgres-exporter
+FROM quay.io/prometheuscommunity/pgbouncer-exporter:v0.10.2 AS pgbouncer-exporter
 
 # Keep the version the same as in build-tools.Dockerfile and
 # test_runner/regress/test_compute_metrics.py.
@@ -1347,9 +1353,6 @@ COPY --from=pg-roaringbitmap-pg-build /pg_roaringbitmap.tar.gz /ext-src
 COPY --from=pg-semver-pg-build /pg_semver.tar.gz /ext-src
 #COPY --from=pg-embedding-pg-build /home/nonroot/pg_embedding-src/ /ext-src
 #COPY --from=wal2json-pg-build /wal2json_2_5.tar.gz /ext-src
-#pg_anon is not supported yet for pg v17 so, don't fail if nothing found
-COPY --from=pg-anon-pg-build /pg_anon.tar.g? /ext-src
-COPY compute/patches/pg_anon.patch /ext-src
 COPY --from=pg-ivm-build /pg_ivm.tar.gz /ext-src
 COPY --from=pg-partman-build /pg_partman.tar.gz /ext-src
 RUN cd /ext-src/ && for f in *.tar.gz; \
@@ -1360,9 +1363,6 @@ RUN cd /ext-src/rum-src && patch -p1 <../rum.patch
 RUN cd /ext-src/pgvector-src && patch -p1 <../pgvector.patch
 RUN cd /ext-src/pg_hint_plan-src && patch -p1 < /ext-src/pg_hint_plan_${PG_VERSION}.patch
 COPY --chmod=755 docker-compose/run-tests.sh /run-tests.sh
-RUN case "${PG_VERSION}" in "v17") \
-    echo "postgresql_anonymizer does not yet support PG17" && exit 0;; \
-    esac && patch -p1 </ext-src/pg_anon.patch
 RUN patch -p1 </ext-src/pg_cron.patch
 ENV PATH=/usr/local/pgsql/bin:$PATH
 ENV PGHOST=compute
@@ -1403,6 +1403,7 @@ RUN mkdir -p /etc/local_proxy && chown postgres:postgres /etc/local_proxy
 
 # Metrics exporter binaries and  configuration files
 COPY --from=postgres-exporter /bin/postgres_exporter /bin/postgres_exporter
+COPY --from=pgbouncer-exporter /bin/pgbouncer_exporter /bin/pgbouncer_exporter
 COPY --from=sql-exporter      /bin/sql_exporter      /bin/sql_exporter
 
 COPY --chown=postgres compute/etc/postgres_exporter.yml /etc/postgres_exporter.yml

compute/etc/pgbouncer.ini

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

compute/patches/pgvector.patch

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

compute/vm-image-spec-bookworm.yaml

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

compute/vm-image-spec-bullseye.yaml

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

compute_tools/src/bin/fast_import.rs

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

compute_tools/src/compute.rs

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

compute_tools/src/config.rs

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

compute_tools/src/extension_server.rs

@@ -85,6 +85,8 @@ use tracing::info;
 use tracing::log::warn;
 use zstd::stream::read::Decoder;
 
+use crate::metrics::{REMOTE_EXT_REQUESTS_FAILED, REMOTE_EXT_REQUESTS_TOTAL, UNKNOWN_HTTP_STATUS};
+
 fn get_pg_config(argument: &str, pgbin: &str) -> String {
     // gives the result of `pg_config [argument]`
     // where argument is a flag like `--version` or `--sharedir`
@@ -258,21 +260,60 @@ async fn download_extension_tar(ext_remote_storage: &str, ext_path: &str) -> Res
 
     info!("Download extension {:?} from uri {:?}", ext_path, uri);
 
-    let resp = reqwest::get(uri).await?;
+    REMOTE_EXT_REQUESTS_TOTAL.with_label_values(&[]).inc();
 
-    match resp.status() {
+    match do_extension_server_request(&uri).await {
+        Ok(resp) => {
+            info!(
+                "Successfully downloaded remote extension data {:?}",
+                ext_path
+            );
+            Ok(resp)
+        }
+        Err((msg, status)) => {
+            let status_str = status
+                .map(|s| s.to_string())
+                .unwrap_or(UNKNOWN_HTTP_STATUS.to_string());
+            REMOTE_EXT_REQUESTS_FAILED
+                .with_label_values(&[&status_str])
+                .inc();
+            bail!(msg);
+        }
+    }
+}
+
+// Do a single remote extensions server request.
+// Return result or (error message + status code) in case of any failures.
+async fn do_extension_server_request(uri: &str) -> Result<Bytes, (String, Option<StatusCode>)> {
+    let resp = reqwest::get(uri).await.map_err(|e| {
+        (
+            format!("could not perform remote extensions server request: {}", e),
+            None,
+        )
+    })?;
+    let status = resp.status();
+
+    match status {
         StatusCode::OK => match resp.bytes().await {
-            Ok(resp) => {
-                info!("Download extension {:?} completed successfully", ext_path);
-                Ok(resp)
-            }
-            Err(e) => bail!("could not deserialize remote extension response: {}", e),
+            Ok(resp) => Ok(resp),
+            Err(e) => Err((
+                format!("could not read remote extensions server response: {}", e),
+                // It's fine to return and report error with status as 200 OK,
+                // because we still failed to read the response.
+                Some(status),
+            )),
         },
-        StatusCode::SERVICE_UNAVAILABLE => bail!("remote extension is temporarily unavailable"),
-        _ => bail!(
-            "unexpected remote extension response status code: {}",
-            resp.status()
-        ),
+        StatusCode::SERVICE_UNAVAILABLE => Err((
+            "remote extensions server is temporarily unavailable".to_string(),
+            Some(status),
+        )),
+        _ => Err((
+            format!(
+                "unexpected remote extensions server response status code: {}",
+                status
+            ),
+            Some(status),
+        )),
     }
 }
 

compute_tools/src/http/extract/json.rs

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

compute_tools/src/http/extract/path.rs

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

compute_tools/src/http/extract/query.rs

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

compute_tools/src/http/routes/metrics.rs

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

compute_tools/src/http/server.rs

@@ -55,7 +55,7 @@ async fn serve(port: u16, compute: Arc<ComputeNode>) {
         .route("/database_schema", get(database_schema::get_schema_dump))
         .route("/dbs_and_roles", get(dbs_and_roles::get_catalog_objects))
         .route(
-            "/extension_server/*filename",
+            "/extension_server/{*filename}",
             post(extension_server::download_extension),
         )
         .route("/extensions", post(extensions::install_extension))

compute_tools/src/installed_extensions.rs

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

compute_tools/src/lib.rs

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

compute_tools/src/metrics.rs

@@ -0,0 +1,90 @@
+use metrics::core::Collector;
+use metrics::proto::MetricFamily;
+use metrics::{register_int_counter_vec, register_uint_gauge_vec, IntCounterVec, UIntGaugeVec};
+use once_cell::sync::Lazy;
+
+pub(crate) static INSTALLED_EXTENSIONS: Lazy<UIntGaugeVec> = Lazy::new(|| {
+    register_uint_gauge_vec!(
+        "compute_installed_extensions",
+        "Number of databases where the version of extension is installed",
+        &["extension_name", "version", "owned_by_superuser"]
+    )
+    .expect("failed to define a metric")
+});
+
+// Normally, any HTTP API request is described by METHOD (e.g. GET, POST, etc.) + PATH,
+// but for all our APIs we defined a 'slug'/method/operationId in the OpenAPI spec.
+// And it's fair to call it a 'RPC' (Remote Procedure Call).
+pub enum CPlaneRequestRPC {
+    GetSpec,
+}
+
+impl CPlaneRequestRPC {
+    pub fn as_str(&self) -> &str {
+        match self {
+            CPlaneRequestRPC::GetSpec => "GetSpec",
+        }
+    }
+}
+
+pub const UNKNOWN_HTTP_STATUS: &str = "unknown";
+
+pub(crate) static CPLANE_REQUESTS_TOTAL: Lazy<IntCounterVec> = Lazy::new(|| {
+    register_int_counter_vec!(
+        "compute_ctl_cplane_requests_total",
+        "Total number of control plane requests made by compute_ctl",
+        &["rpc"]
+    )
+    .expect("failed to define a metric")
+});
+
+pub(crate) static CPLANE_REQUESTS_FAILED: Lazy<IntCounterVec> = Lazy::new(|| {
+    register_int_counter_vec!(
+        "compute_ctl_cplane_requests_failed_total",
+        "Total number of failed control plane requests made by compute_ctl",
+        &["rpc", "http_status"]
+    )
+    .expect("failed to define a metric")
+});
+
+/// Total number of failed database migrations. Per-compute, this is actually a boolean metric,
+/// either empty or with a single value (1, migration_id) because we stop at the first failure.
+/// Yet, the sum over the fleet will provide the total number of failures.
+pub(crate) static DB_MIGRATION_FAILED: Lazy<IntCounterVec> = Lazy::new(|| {
+    register_int_counter_vec!(
+        "compute_ctl_db_migration_failed_total",
+        "Total number of failed database migrations",
+        &["migration_id"]
+    )
+    .expect("failed to define a metric")
+});
+
+pub(crate) static REMOTE_EXT_REQUESTS_TOTAL: Lazy<IntCounterVec> = Lazy::new(|| {
+    register_int_counter_vec!(
+        "compute_ctl_remote_ext_requests_total",
+        "Total number of requests made by compute_ctl to download extensions from S3 proxy",
+        // Do not use any labels like extension name yet.
+        // We can add them later if needed.
+        &[]
+    )
+    .expect("failed to define a metric")
+});
+
+pub(crate) static REMOTE_EXT_REQUESTS_FAILED: Lazy<IntCounterVec> = Lazy::new(|| {
+    register_int_counter_vec!(
+        "compute_ctl_remote_ext_requests_failed_total",
+        "Total number of failed requests to S3 proxy",
+        &["http_status"]
+    )
+    .expect("failed to define a metric")
+});
+
+pub fn collect() -> Vec<MetricFamily> {
+    let mut metrics = INSTALLED_EXTENSIONS.collect();
+    metrics.extend(CPLANE_REQUESTS_TOTAL.collect());
+    metrics.extend(CPLANE_REQUESTS_FAILED.collect());
+    metrics.extend(DB_MIGRATION_FAILED.collect());
+    metrics.extend(REMOTE_EXT_REQUESTS_TOTAL.collect());
+    metrics.extend(REMOTE_EXT_REQUESTS_FAILED.collect());
+    metrics
+}

compute_tools/src/migration.rs

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

compute_tools/src/spec.rs

@@ -6,6 +6,9 @@ use std::path::Path;
 use tracing::{error, info, instrument, warn};
 
 use crate::config;
+use crate::metrics::{
+    CPlaneRequestRPC, CPLANE_REQUESTS_FAILED, CPLANE_REQUESTS_TOTAL, UNKNOWN_HTTP_STATUS,
+};
 use crate::migration::MigrationRunner;
 use crate::params::PG_HBA_ALL_MD5;
 use crate::pg_helpers::*;
@@ -19,7 +22,7 @@ use compute_api::spec::ComputeSpec;
 fn do_control_plane_request(
     uri: &str,
     jwt: &str,
-) -> Result<ControlPlaneSpecResponse, (bool, String)> {
+) -> Result<ControlPlaneSpecResponse, (bool, String, Option<StatusCode>)> {
     let resp = reqwest::blocking::Client::new()
         .get(uri)
         .header("Authorization", format!("Bearer {}", jwt))
@@ -28,34 +31,41 @@ fn do_control_plane_request(
             (
                 true,
                 format!("could not perform spec request to control plane: {}", e),
+                None,
             )
         })?;
 
-    match resp.status() {
+    let status = resp.status();
+    match status {
         StatusCode::OK => match resp.json::<ControlPlaneSpecResponse>() {
             Ok(spec_resp) => Ok(spec_resp),
             Err(e) => Err((
                 true,
                 format!("could not deserialize control plane response: {}", e),
+                Some(status),
             )),
         },
-        StatusCode::SERVICE_UNAVAILABLE => {
-            Err((true, "control plane is temporarily unavailable".to_string()))
-        }
+        StatusCode::SERVICE_UNAVAILABLE => Err((
+            true,
+            "control plane is temporarily unavailable".to_string(),
+            Some(status),
+        )),
         StatusCode::BAD_GATEWAY => {
             // We have a problem with intermittent 502 errors now
             // https://github.com/neondatabase/cloud/issues/2353
             // It's fine to retry GET request in this case.
-            Err((true, "control plane request failed with 502".to_string()))
+            Err((
+                true,
+                "control plane request failed with 502".to_string(),
+                Some(status),
+            ))
         }
         // Another code, likely 500 or 404, means that compute is unknown to the control plane
         // or some internal failure happened. Doesn't make much sense to retry in this case.
         _ => Err((
             false,
-            format!(
-                "unexpected control plane response status code: {}",
-                resp.status()
-            ),
+            format!("unexpected control plane response status code: {}", status),
+            Some(status),
         )),
     }
 }
@@ -82,6 +92,9 @@ pub fn get_spec_from_control_plane(
     // - no spec for compute yet (Empty state) -> return Ok(None)
     // - got spec -> return Ok(Some(spec))
     while attempt < 4 {
+        CPLANE_REQUESTS_TOTAL
+            .with_label_values(&[CPlaneRequestRPC::GetSpec.as_str()])
+            .inc();
         spec = match do_control_plane_request(&cp_uri, &jwt) {
             Ok(spec_resp) => match spec_resp.status {
                 ControlPlaneComputeStatus::Empty => Ok(None),
@@ -93,7 +106,13 @@ pub fn get_spec_from_control_plane(
                     }
                 }
             },
-            Err((retry, msg)) => {
+            Err((retry, msg, status)) => {
+                let status_str = status
+                    .map(|s| s.to_string())
+                    .unwrap_or(UNKNOWN_HTTP_STATUS.to_string());
+                CPLANE_REQUESTS_FAILED
+                    .with_label_values(&[CPlaneRequestRPC::GetSpec.as_str(), &status_str])
+                    .inc();
                 if retry {
                     Err(anyhow!(msg))
                 } else {

compute_tools/src/spec_apply.rs

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

compute_tools/src/sql/finalize_drop_subscriptions.sql

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

control_plane/README.md

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

control_plane/src/bin/neon_local.rs

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

control_plane/src/endpoint.rs

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

control_plane/src/pageserver.rs

@@ -347,11 +347,31 @@ impl PageServerNode {
                 .map(|x| x.parse::<usize>())
                 .transpose()
                 .context("Failed to parse 'compaction_threshold' as an integer")?,
+            compaction_upper_limit: settings
+                .remove("compaction_upper_limit")
+                .map(|x| x.parse::<usize>())
+                .transpose()
+                .context("Failed to parse 'compaction_upper_limit' as an integer")?,
             compaction_algorithm: settings
                 .remove("compaction_algorithm")
                 .map(serde_json::from_str)
                 .transpose()
                 .context("Failed to parse 'compaction_algorithm' json")?,
+            l0_flush_delay_threshold: settings
+                .remove("l0_flush_delay_threshold")
+                .map(|x| x.parse::<usize>())
+                .transpose()
+                .context("Failed to parse 'l0_flush_delay_threshold' as an integer")?,
+            l0_flush_wait_upload: settings
+                .remove("l0_flush_wait_upload")
+                .map(|x| x.parse::<bool>())
+                .transpose()
+                .context("Failed to parse 'l0_flush_wait_upload' as a boolean")?,
+            l0_flush_stall_threshold: settings
+                .remove("l0_flush_stall_threshold")
+                .map(|x| x.parse::<usize>())
+                .transpose()
+                .context("Failed to parse 'l0_flush_stall_threshold' as an integer")?,
             gc_horizon: settings
                 .remove("gc_horizon")
                 .map(|x| x.parse::<u64>())
@@ -423,6 +443,21 @@ impl PageServerNode {
                 .map(|x| x.parse::<bool>())
                 .transpose()
                 .context("Failed to parse 'rel_size_v2_enabled' as bool")?,
+            gc_compaction_enabled: settings
+                .remove("gc_compaction_enabled")
+                .map(|x| x.parse::<bool>())
+                .transpose()
+                .context("Failed to parse 'gc_compaction_enabled' as bool")?,
+            gc_compaction_initial_threshold_kb: settings
+                .remove("gc_compaction_initial_threshold_kb")
+                .map(|x| x.parse::<u64>())
+                .transpose()
+                .context("Failed to parse 'gc_compaction_initial_threshold_kb' as integer")?,
+            gc_compaction_ratio_percent: settings
+                .remove("gc_compaction_ratio_percent")
+                .map(|x| x.parse::<u64>())
+                .transpose()
+                .context("Failed to parse 'gc_compaction_ratio_percent' as integer")?,
         };
         if !settings.is_empty() {
             bail!("Unrecognized tenant settings: {settings:?}")

deny.toml

@@ -41,8 +41,8 @@ allow = [
     "MIT",
     "MPL-2.0",
     "OpenSSL",
-    "Unicode-DFS-2016",
     "Unicode-3.0",
+    "Zlib",
 ]
 confidence-threshold = 0.8
 exceptions = [

docker-compose/compute_wrapper/Dockerfile

@@ -10,10 +10,7 @@ USER root
 RUN apt-get update &&       \
     apt-get install -y curl \
                        jq   \
-                       python3-pip \
                        netcat-openbsd
-#Faker is required for the pg_anon test
-RUN case $COMPUTE_IMAGE in compute-node-v17) OPT="--break-system-packages";; *) OPT= ;; esac && pip3 install $OPT Faker
 #This is required for the pg_hintplan test
 RUN mkdir -p /ext-src/pg_hint_plan-src && chown postgres /ext-src/pg_hint_plan-src 
 

docker-compose/compute_wrapper/shell/compute.sh

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

docker-compose/docker-compose.yml

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

docker-compose/docker_compose_test.sh

@@ -18,14 +18,10 @@ cd $(dirname $0)
 COMPUTE_CONTAINER_NAME=docker-compose-compute-1
 TEST_CONTAINER_NAME=docker-compose-neon-test-extensions-1
 PSQL_OPTION="-h localhost -U cloud_admin -p 55433 -d postgres"
-: ${http_proxy:=}
-: ${https_proxy:=}
-export http_proxy https_proxy
 
 cleanup() {
     echo "show container information"
     docker ps
-    docker compose --profile test-extensions -f $COMPOSE_FILE logs
     echo "stop containers..."
     docker compose --profile test-extensions -f $COMPOSE_FILE down
 }
@@ -35,12 +31,6 @@ for pg_version in ${TEST_VERSION_ONLY-14 15 16 17}; do
     echo "clean up containers if exists"
     cleanup
     PG_TEST_VERSION=$((pg_version < 16 ? 16 : pg_version))
-    # The support of pg_anon not yet added to PG17, so we have to add the corresponding option for other PG versions
-    if [ "${pg_version}" -ne 17 ]; then
-      SPEC_PATH="compute_wrapper/var/db/postgres/specs"
-      mv $SPEC_PATH/spec.json $SPEC_PATH/spec.bak
-      jq '.cluster.settings += [{"name": "session_preload_libraries","value": "anon","vartype": "string"}]' "${SPEC_PATH}/spec.bak" > "${SPEC_PATH}/spec.json"
-    fi
     PG_VERSION=$pg_version PG_TEST_VERSION=$PG_TEST_VERSION docker compose --profile test-extensions -f $COMPOSE_FILE up --build -d
 
     echo "wait until the compute is ready. timeout after 60s. "
@@ -50,7 +40,6 @@ for pg_version in ${TEST_VERSION_ONLY-14 15 16 17}; do
         cnt=`expr $cnt + 3`
         if [ $cnt -gt 60 ]; then
             echo "timeout before the compute is ready."
-            cleanup
             exit 1
         fi
         if docker compose --profile test-extensions -f $COMPOSE_FILE logs "compute_is_ready" | grep -q "accepting connections"; then
@@ -62,36 +51,19 @@ for pg_version in ${TEST_VERSION_ONLY-14 15 16 17}; do
     done
 
     if [ $pg_version -ge 16 ]; then
-        echo Enabling trust connection
-        docker exec $COMPUTE_CONTAINER_NAME bash -c "sed -i '\$d' /var/db/postgres/compute/pg_hba.conf && echo -e 'host\t all\t all\t all\t trust' >> /var/db/postgres/compute/pg_hba.conf && psql $PSQL_OPTION -c 'select pg_reload_conf()' "
-        echo Adding postgres role
-        docker exec $COMPUTE_CONTAINER_NAME psql $PSQL_OPTION -c "CREATE ROLE postgres SUPERUSER LOGIN"
         # This is required for the pg_hint_plan test, to prevent flaky log message causing the test to fail
         # It cannot be moved to Dockerfile now because the database directory is created after the start of the container
         echo Adding dummy config
         docker exec $COMPUTE_CONTAINER_NAME touch /var/db/postgres/compute/compute_ctl_temp_override.conf
-        # This block is required for the pg_anon extension test.
-        # The test assumes that it is running on the same host with the postgres engine.
-        # In our case it's not true, that's why we are copying files to the compute node
+        # The following block copies the files for the pg_hintplan test to the compute node for the extension test in an isolated docker-compose environment
         TMPDIR=$(mktemp -d)
-        # Add support for pg_anon for pg_v16
-        if [ $pg_version -ne 17 ]; then
-          docker cp $TEST_CONTAINER_NAME:/ext-src/pg_anon-src/data $TMPDIR/data
-          echo -e '1\t too \t many \t tabs' > $TMPDIR/data/bad.csv
-          docker cp $TMPDIR/data $COMPUTE_CONTAINER_NAME:/tmp/tmp_anon_alternate_data
-        rm -rf $TMPDIR
-        fi
-        TMPDIR=$(mktemp -d)
-        # The following block does the same for the pg_hintplan test
         docker cp $TEST_CONTAINER_NAME:/ext-src/pg_hint_plan-src/data $TMPDIR/data
         docker cp $TMPDIR/data $COMPUTE_CONTAINER_NAME:/ext-src/pg_hint_plan-src/
         rm -rf $TMPDIR
         # We are running tests now
-        if docker exec -e SKIP=timescaledb-src,rdkit-src,postgis-src,pgx_ulid-src,pgtap-src,pg_tiktoken-src,pg_jsonschema-src,pg_graphql-src,kq_imcx-src,wal2json_2_5-src \
+        if ! docker exec -e SKIP=timescaledb-src,rdkit-src,postgis-src,pgx_ulid-src,pgtap-src,pg_tiktoken-src,pg_jsonschema-src,pg_graphql-src,kq_imcx-src,wal2json_2_5-src \
             $TEST_CONTAINER_NAME /run-tests.sh | tee testout.txt
         then
-            cleanup
-        else
             FAILED=$(tail -1 testout.txt)
             for d in $FAILED
             do
@@ -101,13 +73,7 @@ for pg_version in ${TEST_VERSION_ONLY-14 15 16 17}; do
                 cat $d/regression.out $d/regression.diffs || true
             done
         rm -rf $FAILED
-        cleanup
         exit 1
         fi
     fi
-    cleanup
-    # Restore the original spec.json
-    if [ "$pg_version" -ne 17 ]; then
-      mv "$SPEC_PATH/spec.bak" "$SPEC_PATH/spec.json"
-    fi
 done

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

docker-compose/run-tests.sh

@@ -7,7 +7,10 @@ LIST=$( (echo -e "${SKIP//","/"\n"}"; ls -d -- *-src) | sort | uniq -u)
 for d in ${LIST}
 do
        [ -d "${d}" ] || continue
-    psql -c "select 1" >/dev/null || break
+       if ! psql -w -c "select 1" >/dev/null; then
+          FAILED="${d} ${FAILED}"
+          break
+       fi
        USE_PGXS=1 make -C "${d}" installcheck || FAILED="${d} ${FAILED}"
 done
 [ -z "${FAILED}" ] && exit 0

docker-compose/test_extensions_upgrade.sh

@@ -0,0 +1,93 @@
+#!/bin/bash
+set -eux -o pipefail
+cd "$(dirname "${0}")"
+# Takes a variable name as argument. The result is stored in that variable.
+generate_id() {
+    local -n resvar=$1
+    printf -v resvar '%08x%08x%08x%08x' $SRANDOM $SRANDOM $SRANDOM $SRANDOM
+}
+if [ -z ${OLDTAG+x} ] || [ -z ${NEWTAG+x} ] || [ -z "${OLDTAG}" ] || [ -z "${NEWTAG}" ]; then
+  echo OLDTAG and NEWTAG must be defined
+  exit 1
+fi
+export PG_VERSION=${PG_VERSION:-16}
+function wait_for_ready {
+  TIME=0
+  while ! docker compose logs compute_is_ready | grep -q "accepting connections" && [ ${TIME} -le 300 ] ; do
+    ((TIME += 1 ))
+    sleep 1
+  done
+  if [ ${TIME} -gt 300 ]; then
+    echo Time is out.
+    exit 2
+  fi
+}
+function create_extensions() {
+  for ext in ${1}; do
+    docker compose exec neon-test-extensions psql -X -v ON_ERROR_STOP=1 -d contrib_regression -c "CREATE EXTENSION IF NOT EXISTS ${ext}"
+  done
+}
+EXTENSIONS='[
+{"extname": "plv8", "extdir": "plv8-src"},
+{"extname": "vector", "extdir": "pgvector-src"},
+{"extname": "unit", "extdir": "postgresql-unit-src"},
+{"extname": "hypopg", "extdir": "hypopg-src"},
+{"extname": "rum", "extdir": "rum-src"},
+{"extname": "ip4r", "extdir": "ip4r-src"},
+{"extname": "prefix", "extdir": "prefix-src"},
+{"extname": "hll", "extdir": "hll-src"},
+{"extname": "pg_cron", "extdir": "pg_cron-src"},
+{"extname": "pg_uuidv7", "extdir": "pg_uuidv7-src"},
+{"extname": "roaringbitmap", "extdir": "pg_roaringbitmap-src"},
+{"extname": "semver", "extdir": "pg_semver-src"},
+{"extname": "pg_ivm", "extdir": "pg_ivm-src"}
+]'
+EXTNAMES=$(echo ${EXTENSIONS} | jq -r '.[].extname' | paste -sd ' ' -)
+TAG=${NEWTAG} docker compose --profile test-extensions up --quiet-pull --build -d
+wait_for_ready
+docker compose exec neon-test-extensions psql -c "DROP DATABASE IF EXISTS contrib_regression"
+docker compose exec neon-test-extensions psql -c "CREATE DATABASE contrib_regression"
+create_extensions "${EXTNAMES}"
+query="select json_object_agg(extname,extversion) from pg_extension where extname in ('${EXTNAMES// /\',\'}')"
+new_vers=$(docker compose exec neon-test-extensions psql -Aqt -d contrib_regression -c "$query")
+docker compose --profile test-extensions down
+TAG=${OLDTAG} docker compose --profile test-extensions up --quiet-pull --build -d --force-recreate
+wait_for_ready
+docker compose cp  ext-src neon-test-extensions:/
+docker compose exec neon-test-extensions psql -c "DROP DATABASE IF EXISTS contrib_regression"
+docker compose exec neon-test-extensions psql -c "CREATE DATABASE contrib_regression"
+create_extensions "${EXTNAMES}"
+query="select pge.extname from pg_extension pge join (select key as extname, value as extversion from json_each_text('${new_vers}')) x on pge.extname=x.extname and pge.extversion <> x.extversion"
+exts=$(docker compose exec neon-test-extensions psql -Aqt -d contrib_regression -c "$query")
+if [ -z "${exts}" ]; then
+  echo "No extensions were upgraded"
+else
+  tenant_id=$(docker compose exec neon-test-extensions psql -Aqt -c "SHOW neon.tenant_id")
+  timeline_id=$(docker compose exec neon-test-extensions psql -Aqt -c "SHOW neon.timeline_id")
+  for ext in ${exts}; do
+    echo Testing ${ext}...
+    EXTDIR=$(echo ${EXTENSIONS} | jq -r '.[] | select(.extname=="'${ext}'") | .extdir')
+    generate_id new_timeline_id
+    PARAMS=(
+        -sbf
+        -X POST
+        -H "Content-Type: application/json"
+        -d "{\"new_timeline_id\": \"${new_timeline_id}\", \"pg_version\": ${PG_VERSION}, \"ancestor_timeline_id\": \"${timeline_id}\"}"
+        "http://127.0.0.1:9898/v1/tenant/${tenant_id}/timeline/"
+    )
+    result=$(curl "${PARAMS[@]}")
+    echo $result | jq .
+    TENANT_ID=${tenant_id} TIMELINE_ID=${new_timeline_id} TAG=${OLDTAG} docker compose down compute compute_is_ready
+    COMPUTE_TAG=${NEWTAG} TAG=${OLDTAG} TENANT_ID=${tenant_id} TIMELINE_ID=${new_timeline_id} docker compose up --quiet-pull -d --build compute compute_is_ready
+    wait_for_ready
+    TID=$(docker compose exec neon-test-extensions psql -Aqt -c "SHOW neon.timeline_id")
+    if [ ${TID} != ${new_timeline_id} ]; then
+      echo Timeline mismatch
+      exit 1
+    fi
+    docker compose exec neon-test-extensions psql -d contrib_regression -c "\dx ${ext}"
+    docker compose exec neon-test-extensions sh -c /ext-src/${EXTDIR}/test-upgrade.sh
+    docker compose exec neon-test-extensions psql -d contrib_regression -c "alter extension ${ext} update"
+    docker compose exec neon-test-extensions psql -d contrib_regression -c "\dx ${ext}"
+  done
+fi

libs/compute_api/src/spec.rs

@@ -138,6 +138,13 @@ pub struct ComputeSpec {
     /// enough spare connections for reconfiguration process to succeed.
     #[serde(default = "default_reconfigure_concurrency")]
     pub reconfigure_concurrency: usize,
+
+    /// If set to true, the compute_ctl will drop all subscriptions before starting the
+    /// compute. This is needed when we start an endpoint on a branch, so that child
+    /// would not compete with parent branch subscriptions
+    /// over the same replication content from publisher.
+    #[serde(default)] // Default false
+    pub drop_subscriptions_before_start: bool,
 }
 
 /// Feature flag to signal `compute_ctl` to enable certain experimental functionality.

libs/pageserver_api/src/config.rs

@@ -120,6 +120,7 @@ pub struct ConfigToml {
     pub no_sync: Option<bool>,
     pub wal_receiver_protocol: PostgresClientProtocol,
     pub page_service_pipelining: PageServicePipeliningConfig,
+    pub get_vectored_concurrent_io: GetVectoredConcurrentIo,
 }
 
 #[derive(Debug, Clone, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
@@ -158,6 +159,25 @@ pub enum PageServiceProtocolPipelinedExecutionStrategy {
     Tasks,
 }
 
+#[derive(Debug, Clone, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
+#[serde(tag = "mode", rename_all = "kebab-case")]
+#[serde(deny_unknown_fields)]
+pub enum GetVectoredConcurrentIo {
+    /// The read path is fully sequential: layers are visited
+    /// one after the other and IOs are issued and waited upon
+    /// from the same task that traverses the layers.
+    Sequential,
+    /// The read path still traverses layers sequentially, and
+    /// index blocks will be read into the PS PageCache from
+    /// that task, with waiting.
+    /// But data IOs are dispatched and waited upon from a sidecar
+    /// task so that the traversing task can continue to traverse
+    /// layers while the IOs are in flight.
+    /// If the PS PageCache miss rate is low, this improves
+    /// throughput dramatically.
+    SidecarTask,
+}
+
 pub mod statvfs {
     pub mod mock {
         #[derive(Debug, Clone, PartialEq, Eq, serde::Serialize, serde::Deserialize)]
@@ -234,9 +254,26 @@ pub struct TenantConfigToml {
     // Duration::ZERO means automatic compaction is disabled.
     #[serde(with = "humantime_serde")]
     pub compaction_period: Duration,
-    // Level0 delta layer threshold for compaction.
+    /// Level0 delta layer threshold for compaction.
     pub compaction_threshold: usize,
+    /// Controls the amount of L0 included in a single compaction iteration.
+    /// The unit is `checkpoint_distance`, i.e., a size.
+    /// We add L0s to the set of layers to compact until their cumulative
+    /// size exceeds `compaction_upper_limit * checkpoint_distance`.
+    pub compaction_upper_limit: usize,
     pub compaction_algorithm: crate::models::CompactionAlgorithmSettings,
+    /// Level0 delta layer threshold at which to delay layer flushes for compaction backpressure,
+    /// such that they take 2x as long, and start waiting for layer flushes during ephemeral layer
+    /// rolls. This helps compaction keep up with WAL ingestion, and avoids read amplification
+    /// blowing up. Should be >compaction_threshold. 0 to disable. Disabled by default.
+    pub l0_flush_delay_threshold: Option<usize>,
+    /// Level0 delta layer threshold at which to stall layer flushes. Must be >compaction_threshold
+    /// to avoid deadlock. 0 to disable. Disabled by default.
+    pub l0_flush_stall_threshold: Option<usize>,
+    /// If true, Level0 delta layer flushes will wait for S3 upload before flushing the next
+    /// layer. This is a temporary backpressure mechanism which should be removed once
+    /// l0_flush_{delay,stall}_threshold is fully enabled.
+    pub l0_flush_wait_upload: bool,
     // Determines how much history is retained, to allow
     // branching and read replicas at an older point in time.
     // The unit is #of bytes of WAL.
@@ -305,6 +342,16 @@ pub struct TenantConfigToml {
     /// Enable rel_size_v2 for this tenant. Once enabled, the tenant will persist this information into
     /// `index_part.json`, and it cannot be reversed.
     pub rel_size_v2_enabled: Option<bool>,
+
+    // gc-compaction related configs
+    /// Enable automatic gc-compaction trigger on this tenant.
+    pub gc_compaction_enabled: bool,
+    /// The initial threshold for gc-compaction in KB. Once the total size of layers below the gc-horizon is above this threshold,
+    /// gc-compaction will be triggered.
+    pub gc_compaction_initial_threshold_kb: u64,
+    /// The ratio that triggers the auto gc-compaction. If (the total size of layers between L2 LSN and gc-horizon) / (size below the L2 LSN)
+    /// is above this ratio, gc-compaction will be triggered.
+    pub gc_compaction_ratio_percent: u64,
 }
 
 pub mod defaults {
@@ -454,6 +501,11 @@ impl Default for ConfigToml {
                     execution: PageServiceProtocolPipelinedExecutionStrategy::ConcurrentFutures,
                 })
             },
+            get_vectored_concurrent_io: if !cfg!(test) {
+                GetVectoredConcurrentIo::Sequential
+            } else {
+                GetVectoredConcurrentIo::SidecarTask
+            },
         }
     }
 }
@@ -476,9 +528,17 @@ pub mod tenant_conf_defaults {
 
     pub const DEFAULT_COMPACTION_PERIOD: &str = "20 s";
     pub const DEFAULT_COMPACTION_THRESHOLD: usize = 10;
+
+    // This value needs to be tuned to avoid OOM. We have 3/4 of the total CPU threads to do background works, that's 16*3/4=9 on
+    // most of our pageservers. Compaction ~50 layers requires about 2GB memory (could be reduced later by optimizing L0 hole
+    // calculation to avoid loading all keys into the memory). So with this config, we can get a maximum peak compaction usage of 18GB.
+    pub const DEFAULT_COMPACTION_UPPER_LIMIT: usize = 50;
+
     pub const DEFAULT_COMPACTION_ALGORITHM: crate::models::CompactionAlgorithm =
         crate::models::CompactionAlgorithm::Legacy;
 
+    pub const DEFAULT_L0_FLUSH_WAIT_UPLOAD: bool = true;
+
     pub const DEFAULT_GC_HORIZON: u64 = 64 * 1024 * 1024;
 
     // Large DEFAULT_GC_PERIOD is fine as long as PITR_INTERVAL is larger.
@@ -498,6 +558,9 @@ pub mod tenant_conf_defaults {
     // By default ingest enough WAL for two new L0 layers before checking if new image
     // image layers should be created.
     pub const DEFAULT_IMAGE_LAYER_CREATION_CHECK_THRESHOLD: u8 = 2;
+    pub const DEFAULT_GC_COMPACTION_ENABLED: bool = false;
+    pub const DEFAULT_GC_COMPACTION_INITIAL_THRESHOLD_KB: u64 = 10240000;
+    pub const DEFAULT_GC_COMPACTION_RATIO_PERCENT: u64 = 100;
 }
 
 impl Default for TenantConfigToml {
@@ -511,9 +574,13 @@ impl Default for TenantConfigToml {
             compaction_period: humantime::parse_duration(DEFAULT_COMPACTION_PERIOD)
                 .expect("cannot parse default compaction period"),
             compaction_threshold: DEFAULT_COMPACTION_THRESHOLD,
+            compaction_upper_limit: DEFAULT_COMPACTION_UPPER_LIMIT,
             compaction_algorithm: crate::models::CompactionAlgorithmSettings {
                 kind: DEFAULT_COMPACTION_ALGORITHM,
             },
+            l0_flush_delay_threshold: None,
+            l0_flush_stall_threshold: None,
+            l0_flush_wait_upload: DEFAULT_L0_FLUSH_WAIT_UPLOAD,
             gc_horizon: DEFAULT_GC_HORIZON,
             gc_period: humantime::parse_duration(DEFAULT_GC_PERIOD)
                 .expect("cannot parse default gc period"),
@@ -543,6 +610,9 @@ impl Default for TenantConfigToml {
             timeline_offloading: false,
             wal_receiver_protocol_override: None,
             rel_size_v2_enabled: None,
+            gc_compaction_enabled: DEFAULT_GC_COMPACTION_ENABLED,
+            gc_compaction_initial_threshold_kb: DEFAULT_GC_COMPACTION_INITIAL_THRESHOLD_KB,
+            gc_compaction_ratio_percent: DEFAULT_GC_COMPACTION_RATIO_PERCENT,
         }
     }
 }

libs/pageserver_api/src/models.rs

@@ -282,7 +282,7 @@ pub struct TimelineCreateRequest {
     pub mode: TimelineCreateRequestMode,
 }
 
-#[derive(Serialize, Deserialize, Clone)]
+#[derive(Serialize, Deserialize, Clone, Debug)]
 #[serde(untagged)]
 pub enum TimelineCreateRequestMode {
     Branch {
@@ -307,7 +307,7 @@ pub enum TimelineCreateRequestMode {
     },
 }
 
-#[derive(Serialize, Deserialize, Clone)]
+#[derive(Serialize, Deserialize, Clone, Debug)]
 pub struct TimelineCreateRequestModeImportPgdata {
     pub location: ImportPgdataLocation,
     pub idempotency_key: ImportPgdataIdempotencyKey,
@@ -326,7 +326,7 @@ pub enum ImportPgdataLocation {
     },
 }
 
-#[derive(Serialize, Deserialize, Clone)]
+#[derive(Serialize, Deserialize, Clone, Debug)]
 #[serde(transparent)]
 pub struct ImportPgdataIdempotencyKey(pub String);
 
@@ -458,10 +458,18 @@ pub struct TenantConfigPatch {
     pub compaction_period: FieldPatch<String>,
     #[serde(skip_serializing_if = "FieldPatch::is_noop")]
     pub compaction_threshold: FieldPatch<usize>,
+    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
+    pub compaction_upper_limit: FieldPatch<usize>,
     // defer parsing compaction_algorithm, like eviction_policy
     #[serde(skip_serializing_if = "FieldPatch::is_noop")]
     pub compaction_algorithm: FieldPatch<CompactionAlgorithmSettings>,
     #[serde(skip_serializing_if = "FieldPatch::is_noop")]
+    pub l0_flush_delay_threshold: FieldPatch<usize>,
+    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
+    pub l0_flush_stall_threshold: FieldPatch<usize>,
+    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
+    pub l0_flush_wait_upload: FieldPatch<bool>,
+    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
     pub gc_horizon: FieldPatch<u64>,
     #[serde(skip_serializing_if = "FieldPatch::is_noop")]
     pub gc_period: FieldPatch<String>,
@@ -499,6 +507,12 @@ pub struct TenantConfigPatch {
     pub wal_receiver_protocol_override: FieldPatch<PostgresClientProtocol>,
     #[serde(skip_serializing_if = "FieldPatch::is_noop")]
     pub rel_size_v2_enabled: FieldPatch<bool>,
+    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
+    pub gc_compaction_enabled: FieldPatch<bool>,
+    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
+    pub gc_compaction_initial_threshold_kb: FieldPatch<u64>,
+    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
+    pub gc_compaction_ratio_percent: FieldPatch<u64>,
 }
 
 /// An alternative representation of `pageserver::tenant::TenantConf` with
@@ -510,8 +524,12 @@ pub struct TenantConfig {
     pub compaction_target_size: Option<u64>,
     pub compaction_period: Option<String>,
     pub compaction_threshold: Option<usize>,
+    pub compaction_upper_limit: Option<usize>,
     // defer parsing compaction_algorithm, like eviction_policy
     pub compaction_algorithm: Option<CompactionAlgorithmSettings>,
+    pub l0_flush_delay_threshold: Option<usize>,
+    pub l0_flush_stall_threshold: Option<usize>,
+    pub l0_flush_wait_upload: Option<bool>,
     pub gc_horizon: Option<u64>,
     pub gc_period: Option<String>,
     pub image_creation_threshold: Option<usize>,
@@ -531,6 +549,9 @@ pub struct TenantConfig {
     pub timeline_offloading: Option<bool>,
     pub wal_receiver_protocol_override: Option<PostgresClientProtocol>,
     pub rel_size_v2_enabled: Option<bool>,
+    pub gc_compaction_enabled: Option<bool>,
+    pub gc_compaction_initial_threshold_kb: Option<u64>,
+    pub gc_compaction_ratio_percent: Option<u64>,
 }
 
 impl TenantConfig {
@@ -541,7 +562,11 @@ impl TenantConfig {
             mut compaction_target_size,
             mut compaction_period,
             mut compaction_threshold,
+            mut compaction_upper_limit,
             mut compaction_algorithm,
+            mut l0_flush_delay_threshold,
+            mut l0_flush_stall_threshold,
+            mut l0_flush_wait_upload,
             mut gc_horizon,
             mut gc_period,
             mut image_creation_threshold,
@@ -561,6 +586,9 @@ impl TenantConfig {
             mut timeline_offloading,
             mut wal_receiver_protocol_override,
             mut rel_size_v2_enabled,
+            mut gc_compaction_enabled,
+            mut gc_compaction_initial_threshold_kb,
+            mut gc_compaction_ratio_percent,
         } = self;
 
         patch.checkpoint_distance.apply(&mut checkpoint_distance);
@@ -570,7 +598,17 @@ impl TenantConfig {
             .apply(&mut compaction_target_size);
         patch.compaction_period.apply(&mut compaction_period);
         patch.compaction_threshold.apply(&mut compaction_threshold);
+        patch
+            .compaction_upper_limit
+            .apply(&mut compaction_upper_limit);
         patch.compaction_algorithm.apply(&mut compaction_algorithm);
+        patch
+            .l0_flush_delay_threshold
+            .apply(&mut l0_flush_delay_threshold);
+        patch
+            .l0_flush_stall_threshold
+            .apply(&mut l0_flush_stall_threshold);
+        patch.l0_flush_wait_upload.apply(&mut l0_flush_wait_upload);
         patch.gc_horizon.apply(&mut gc_horizon);
         patch.gc_period.apply(&mut gc_period);
         patch
@@ -606,6 +644,15 @@ impl TenantConfig {
             .wal_receiver_protocol_override
             .apply(&mut wal_receiver_protocol_override);
         patch.rel_size_v2_enabled.apply(&mut rel_size_v2_enabled);
+        patch
+            .gc_compaction_enabled
+            .apply(&mut gc_compaction_enabled);
+        patch
+            .gc_compaction_initial_threshold_kb
+            .apply(&mut gc_compaction_initial_threshold_kb);
+        patch
+            .gc_compaction_ratio_percent
+            .apply(&mut gc_compaction_ratio_percent);
 
         Self {
             checkpoint_distance,
@@ -613,7 +660,11 @@ impl TenantConfig {
             compaction_target_size,
             compaction_period,
             compaction_threshold,
+            compaction_upper_limit,
             compaction_algorithm,
+            l0_flush_delay_threshold,
+            l0_flush_stall_threshold,
+            l0_flush_wait_upload,
             gc_horizon,
             gc_period,
             image_creation_threshold,
@@ -633,6 +684,9 @@ impl TenantConfig {
             timeline_offloading,
             wal_receiver_protocol_override,
             rel_size_v2_enabled,
+            gc_compaction_enabled,
+            gc_compaction_initial_threshold_kb,
+            gc_compaction_ratio_percent,
         }
     }
 }
@@ -975,6 +1029,13 @@ pub struct TenantConfigPatchRequest {
     pub config: TenantConfigPatch, // as we have a flattened field, we should reject all unknown fields in it
 }
 
+#[derive(Serialize, Deserialize, Debug)]
+pub struct TenantWaitLsnRequest {
+    #[serde(flatten)]
+    pub timelines: HashMap<TimelineId, Lsn>,
+    pub timeout: Duration,
+}
+
 /// See [`TenantState::attachment_status`] and the OpenAPI docs for context.
 #[derive(Serialize, Deserialize, Clone)]
 #[serde(tag = "slug", content = "data", rename_all = "snake_case")]

libs/remote_storage/src/azure_blob.rs

@@ -377,7 +377,8 @@ impl RemoteStorage for AzureBlobStorage {
 
                 let next_item = next_item?;
 
-                if timeout_try_cnt >= 2 {
+                // Log a warning if we saw two timeouts in a row before a successful request
+                if timeout_try_cnt > 2 {
                     tracing::warn!("Azure Blob Storage list timed out and succeeded after {} tries", timeout_try_cnt);
                 }
                 timeout_try_cnt = 1;

libs/utils/src/env.rs

@@ -2,6 +2,7 @@
 
 use std::{fmt::Display, str::FromStr};
 
+/// For types `V` that implement [`FromStr`].
 pub fn var<V, E>(varname: &str) -> Option<V>
 where
     V: FromStr<Err = E>,
@@ -10,7 +11,9 @@ where
     match std::env::var(varname) {
         Ok(s) => Some(
             s.parse()
-                .map_err(|e| format!("failed to parse env var {varname}: {e:#}"))
+                .map_err(|e| {
+                    format!("failed to parse env var {varname} using FromStr::parse: {e:#}")
+                })
                 .unwrap(),
         ),
         Err(std::env::VarError::NotPresent) => None,
@@ -19,3 +22,24 @@ where
         }
     }
 }
+
+/// For types `V` that implement [`serde::de::DeserializeOwned`].
+pub fn var_serde_json_string<V>(varname: &str) -> Option<V>
+where
+    V: serde::de::DeserializeOwned,
+{
+    match std::env::var(varname) {
+        Ok(s) => Some({
+            let value = serde_json::Value::String(s);
+            serde_json::from_value(value)
+                .map_err(|e| {
+                    format!("failed to parse env var {varname} as a serde_json json string: {e:#}")
+                })
+                .unwrap()
+        }),
+        Err(std::env::VarError::NotPresent) => None,
+        Err(std::env::VarError::NotUnicode(_)) => {
+            panic!("env var {varname} is not unicode")
+        }
+    }
+}

libs/utils/src/failpoint_support.rs

@@ -11,31 +11,55 @@ use tracing::*;
 
 /// Declare a failpoint that can use to `pause` failpoint action.
 /// We don't want to block the executor thread, hence, spawn_blocking + await.
+///
+/// Optionally pass a cancellation token, and this failpoint will drop out of
+/// its pause when the cancellation token fires. This is useful for testing
+/// cases where we would like to block something, but test its clean shutdown behavior.
+/// The macro evaluates to a Result in that case, where Ok(()) is the case
+/// where the failpoint was not paused, and Err() is the case where cancellation
+/// token fired while evaluating the failpoint.
+///
+/// Remember to unpause the failpoint in the test; until that happens, one of the
+/// limited number of spawn_blocking thread pool threads is leaked.
 #[macro_export]
 macro_rules! pausable_failpoint {
-    ($name:literal) => {
+    ($name:literal) => {{
         if cfg!(feature = "testing") {
-            tokio::task::spawn_blocking({
-                let current = tracing::Span::current();
+            let cancel = ::tokio_util::sync::CancellationToken::new();
+            let _ = $crate::pausable_failpoint!($name, &cancel);
+        }
+    }};
+    ($name:literal, $cancel:expr) => {{
+        if cfg!(feature = "testing") {
+            let failpoint_fut = ::tokio::task::spawn_blocking({
+                let current = ::tracing::Span::current();
                 move || {
                     let _entered = current.entered();
-                    tracing::info!("at failpoint {}", $name);
-                    fail::fail_point!($name);
+                    ::tracing::info!("at failpoint {}", $name);
+                    ::fail::fail_point!($name);
+                }
+            });
+            let cancel_fut = async move {
+                $cancel.cancelled().await;
+            };
+            ::tokio::select! {
+                res = failpoint_fut => {
+                    res.expect("spawn_blocking");
+                    // continue with execution
+                    Ok(())
+                },
+                _ = cancel_fut => {
+                    Err(())
                 }
-            })
-            .await
-            .expect("spawn_blocking");
-        }
-    };
-    ($name:literal, $cond:expr) => {
-        if cfg!(feature = "testing") {
-            if $cond {
-                pausable_failpoint!($name)
             }
+        } else {
+            Ok(())
         }
-    };
+    }};
 }
 
+pub use pausable_failpoint;
+
 /// use with fail::cfg("$name", "return(2000)")
 ///
 /// The effect is similar to a "sleep(2000)" action, i.e. we sleep for the

libs/utils/src/sync/gate.rs

@@ -64,6 +64,12 @@ pub struct GateGuard {
     gate: Arc<GateInner>,
 }
 
+impl GateGuard {
+    pub fn try_clone(&self) -> Result<Self, GateError> {
+        Gate::enter_impl(self.gate.clone())
+    }
+}
+
 impl Drop for GateGuard {
     fn drop(&mut self) {
         if self.gate.closing.load(Ordering::Relaxed) {
@@ -107,11 +113,11 @@ impl Gate {
     /// to avoid blocking close() indefinitely: typically types that contain a Gate will
     /// also contain a CancellationToken.
     pub fn enter(&self) -> Result<GateGuard, GateError> {
-        let permit = self
-            .inner
-            .sem
-            .try_acquire()
-            .map_err(|_| GateError::GateClosed)?;
+        Self::enter_impl(self.inner.clone())
+    }
+
+    fn enter_impl(gate: Arc<GateInner>) -> Result<GateGuard, GateError> {
+        let permit = gate.sem.try_acquire().map_err(|_| GateError::GateClosed)?;
 
         // we now have the permit, let's disable the normal raii functionality and leave
         // "returning" the permit to our GateGuard::drop.
@@ -122,7 +128,7 @@ impl Gate {
 
         Ok(GateGuard {
             span_at_enter: tracing::Span::current(),
-            gate: self.inner.clone(),
+            gate,
         })
     }
 
@@ -252,4 +258,39 @@ mod tests {
         // Attempting to enter() is still forbidden
         gate.enter().expect_err("enter should fail finishing close");
     }
+
+    #[tokio::test(start_paused = true)]
+    async fn clone_gate_guard() {
+        let gate = Gate::default();
+        let forever = Duration::from_secs(24 * 7 * 365);
+
+        let guard1 = gate.enter().expect("gate isn't closed");
+
+        let guard2 = guard1.try_clone().expect("gate isn't clsoed");
+
+        let mut close_fut = std::pin::pin!(gate.close());
+
+        tokio::time::timeout(forever, &mut close_fut)
+            .await
+            .unwrap_err();
+
+        // we polled close_fut once, that should prevent all later enters and clones
+        gate.enter().unwrap_err();
+        guard1.try_clone().unwrap_err();
+        guard2.try_clone().unwrap_err();
+
+        // guard2 keeps gate open even if guard1 is closed
+        drop(guard1);
+        tokio::time::timeout(forever, &mut close_fut)
+            .await
+            .unwrap_err();
+
+        drop(guard2);
+
+        // now that the last guard is dropped, closing should complete
+        close_fut.await;
+
+        // entering is still forbidden
+        gate.enter().expect_err("enter should stilll fail");
+    }
 }

libs/vm_monitor/src/dispatcher.rs

@@ -7,7 +7,7 @@
 //! (notifying it of upscale).
 
 use anyhow::{bail, Context};
-use axum::extract::ws::{Message, WebSocket};
+use axum::extract::ws::{Message, Utf8Bytes, WebSocket};
 use futures::{
     stream::{SplitSink, SplitStream},
     SinkExt, StreamExt,
@@ -82,21 +82,21 @@ impl Dispatcher {
 
         let highest_shared_version = match monitor_range.highest_shared_version(&agent_range) {
             Ok(version) => {
-                sink.send(Message::Text(
+                sink.send(Message::Text(Utf8Bytes::from(
                     serde_json::to_string(&ProtocolResponse::Version(version)).unwrap(),
-                ))
+                )))
                 .await
                 .context("failed to notify agent of negotiated protocol version")?;
                 version
             }
             Err(e) => {
-                sink.send(Message::Text(
+                sink.send(Message::Text(Utf8Bytes::from(
                     serde_json::to_string(&ProtocolResponse::Error(format!(
                         "Received protocol version range {} which does not overlap with {}",
                         agent_range, monitor_range
                     )))
                     .unwrap(),
-                ))
+                )))
                 .await
                 .context("failed to notify agent of no overlap between protocol version ranges")?;
                 Err(e).context("error determining suitable protocol version range")?
@@ -126,7 +126,7 @@ impl Dispatcher {
 
         let json = serde_json::to_string(&message).context("failed to serialize message")?;
         self.sink
-            .send(Message::Text(json))
+            .send(Message::Text(Utf8Bytes::from(json)))
             .await
             .context("stream error sending message")
     }

libs/wal_decoder/Cargo.toml

@@ -17,7 +17,6 @@ postgres_ffi.workspace = true
 serde.workspace = true
 thiserror.workspace = true
 tokio = { workspace = true, features = ["io-util"] }
-tonic.workspace = true
 tracing.workspace = true
 utils.workspace = true
 workspace_hack = { version = "0.1", path = "../../workspace_hack" }

libs/wal_decoder/src/models.rs

@@ -45,7 +45,7 @@ pub mod proto {
     #![allow(clippy::derive_partial_eq_without_eq)]
     // The generated ValueMeta has a `len` method generate for its `len` field.
     #![allow(clippy::len_without_is_empty)]
-    tonic::include_proto!("interpreted_wal");
+    include!(concat!(env!("OUT_DIR"), concat!("/interpreted_wal.rs")));
 }
 
 #[derive(Copy, Clone, Serialize, Deserialize)]

pageserver/client/src/mgmt_api.rs

@@ -763,4 +763,19 @@ impl Client {
             .await
             .map_err(Error::ReceiveBody)
     }
+
+    pub async fn wait_lsn(
+        &self,
+        tenant_shard_id: TenantShardId,
+        request: TenantWaitLsnRequest,
+    ) -> Result<StatusCode> {
+        let uri = format!(
+            "{}/v1/tenant/{tenant_shard_id}/wait_lsn",
+            self.mgmt_api_endpoint,
+        );
+
+        self.request_noerror(Method::POST, uri, request)
+            .await
+            .map(|resp| resp.status())
+    }
 }

pageserver/compaction/src/simulator/draw.rs

@@ -160,9 +160,12 @@ pub fn draw_history<W: std::io::Write>(history: &[LayerTraceEvent], mut output:
 
         // Fill in and thicken rectangle if it's an
         // image layer so that we can see it.
-        let mut style = Style::default();
-        style.fill = Fill::Color(rgb(0x80, 0x80, 0x80));
-        style.stroke = Stroke::Color(rgb(0, 0, 0), 0.5);
+        let mut style = Style {
+            fill: Fill::Color(rgb(0x80, 0x80, 0x80)),
+            stroke: Stroke::Color(rgb(0, 0, 0), 0.5),
+            opacity: 1.0,
+            stroke_opacity: 1.0,
+        };
 
         let y_start = lsn_max - lsn_start;
         let y_end = lsn_max - lsn_end;
@@ -214,10 +217,6 @@ pub fn draw_history<W: std::io::Write>(history: &[LayerTraceEvent], mut output:
         files_seen.insert(f);
     }
 
-    let mut record_style = Style::default();
-    record_style.fill = Fill::Color(rgb(0x80, 0x80, 0x80));
-    record_style.stroke = Stroke::None;
-
     writeln!(svg, "{}", EndSvg)?;
 
     let mut layer_events_str = String::new();

pageserver/pagebench/src/cmd/getpage_latest_lsn.rs

@@ -13,7 +13,7 @@ use rand::prelude::*;
 use tokio::task::JoinSet;
 use tracing::info;
 
-use std::collections::HashSet;
+use std::collections::{HashSet, VecDeque};
 use std::future::Future;
 use std::num::NonZeroUsize;
 use std::pin::Pin;
@@ -63,6 +63,10 @@ pub(crate) struct Args {
     #[clap(long)]
     set_io_mode: Option<pageserver_api::models::virtual_file::IoMode>,
 
+    /// Queue depth generated in each client.
+    #[clap(long, default_value = "1")]
+    queue_depth: NonZeroUsize,
+
     targets: Option<Vec<TenantTimelineId>>,
 }
 
@@ -298,6 +302,7 @@ async fn main_impl(
             start_work_barrier.wait().await;
             let client_start = Instant::now();
             let mut ticks_processed = 0;
+            let mut inflight = VecDeque::new();
             while !cancel.is_cancelled() {
                 // Detect if a request took longer than the RPS rate
                 if let Some(period) = &rps_period {
@@ -311,31 +316,37 @@ async fn main_impl(
                     ticks_processed = periods_passed_until_now;
                 }
 
-                let start = Instant::now();
-                let req = {
-                    let mut rng = rand::thread_rng();
-                    let r = &ranges[weights.sample(&mut rng)];
-                    let key: i128 = rng.gen_range(r.start..r.end);
-                    let key = Key::from_i128(key);
-                    assert!(key.is_rel_block_key());
-                    let (rel_tag, block_no) = key
-                        .to_rel_block()
-                        .expect("we filter non-rel-block keys out above");
-                    PagestreamGetPageRequest {
-                        hdr: PagestreamRequest {
-                            reqid: 0,
-                            request_lsn: if rng.gen_bool(args.req_latest_probability) {
-                                Lsn::MAX
-                            } else {
-                                r.timeline_lsn
+                while inflight.len() < args.queue_depth.get() {
+                    let start = Instant::now();
+                    let req = {
+                        let mut rng = rand::thread_rng();
+                        let r = &ranges[weights.sample(&mut rng)];
+                        let key: i128 = rng.gen_range(r.start..r.end);
+                        let key = Key::from_i128(key);
+                        assert!(key.is_rel_block_key());
+                        let (rel_tag, block_no) = key
+                            .to_rel_block()
+                            .expect("we filter non-rel-block keys out above");
+                        PagestreamGetPageRequest {
+                            hdr: PagestreamRequest {
+                                reqid: 0,
+                                request_lsn: if rng.gen_bool(args.req_latest_probability) {
+                                    Lsn::MAX
+                                } else {
+                                    r.timeline_lsn
+                                },
+                                not_modified_since: r.timeline_lsn,
                             },
-                            not_modified_since: r.timeline_lsn,
-                        },
-                        rel: rel_tag,
-                        blkno: block_no,
-                    }
-                };
-                client.getpage(req).await.unwrap();
+                            rel: rel_tag,
+                            blkno: block_no,
+                        }
+                    };
+                    client.getpage_send(req).await.unwrap();
+                    inflight.push_back(start);
+                }
+
+                let start = inflight.pop_front().unwrap();
+                client.getpage_recv().await.unwrap();
                 let end = Instant::now();
                 live_stats.request_done();
                 ticks_processed += 1;

pageserver/src/basebackup.rs

@@ -25,6 +25,7 @@ use tokio_tar::{Builder, EntryType, Header};
 
 use crate::context::RequestContext;
 use crate::pgdatadir_mapping::Version;
+use crate::tenant::storage_layer::IoConcurrency;
 use crate::tenant::Timeline;
 use pageserver_api::reltag::{RelTag, SlruKind};
 
@@ -123,6 +124,13 @@ where
         full_backup,
         replica,
         ctx,
+        io_concurrency: IoConcurrency::spawn_from_conf(
+            timeline.conf,
+            timeline
+                .gate
+                .enter()
+                .map_err(|e| BasebackupError::Server(e.into()))?,
+        ),
     };
     basebackup
         .send_tarball()
@@ -144,6 +152,7 @@ where
     full_backup: bool,
     replica: bool,
     ctx: &'a RequestContext,
+    io_concurrency: IoConcurrency,
 }
 
 /// A sink that accepts SLRU blocks ordered by key and forwards
@@ -303,7 +312,7 @@ where
             for part in slru_partitions.parts {
                 let blocks = self
                     .timeline
-                    .get_vectored(part, self.lsn, self.ctx)
+                    .get_vectored(part, self.lsn, self.io_concurrency.clone(), self.ctx)
                     .await
                     .map_err(|e| BasebackupError::Server(e.into()))?;
 
@@ -358,7 +367,7 @@ where
         let start_time = Instant::now();
         let aux_files = self
             .timeline
-            .list_aux_files(self.lsn, self.ctx)
+            .list_aux_files(self.lsn, self.ctx, self.io_concurrency.clone())
             .await
             .map_err(|e| BasebackupError::Server(e.into()))?;
         let aux_scan_time = start_time.elapsed();
@@ -422,7 +431,7 @@ where
         }
         let repl_origins = self
             .timeline
-            .get_replorigins(self.lsn, self.ctx)
+            .get_replorigins(self.lsn, self.ctx, self.io_concurrency.clone())
             .await
             .map_err(|e| BasebackupError::Server(e.into()))?;
         let n_origins = repl_origins.len();
@@ -489,7 +498,13 @@ where
             for blknum in startblk..endblk {
                 let img = self
                     .timeline
-                    .get_rel_page_at_lsn(src, blknum, Version::Lsn(self.lsn), self.ctx)
+                    .get_rel_page_at_lsn(
+                        src,
+                        blknum,
+                        Version::Lsn(self.lsn),
+                        self.ctx,
+                        self.io_concurrency.clone(),
+                    )
                     .await
                     .map_err(|e| BasebackupError::Server(e.into()))?;
                 segment_data.extend_from_slice(&img[..]);

pageserver/src/bin/pageserver.rs

@@ -135,6 +135,7 @@ fn main() -> anyhow::Result<()> {
     info!(?conf.virtual_file_io_mode, "starting with virtual_file IO mode");
     info!(?conf.wal_receiver_protocol, "starting with WAL receiver protocol");
     info!(?conf.page_service_pipelining, "starting with page service pipelining config");
+    info!(?conf.get_vectored_concurrent_io, "starting with get_vectored IO concurrency config");
 
     // The tenants directory contains all the pageserver local disk state.
     // Create if not exists and make sure all the contents are durable before proceeding.

pageserver/src/config.rs

@@ -191,6 +191,8 @@ pub struct PageServerConf {
     pub wal_receiver_protocol: PostgresClientProtocol,
 
     pub page_service_pipelining: pageserver_api::config::PageServicePipeliningConfig,
+
+    pub get_vectored_concurrent_io: pageserver_api::config::GetVectoredConcurrentIo,
 }
 
 /// Token for authentication to safekeepers
@@ -352,6 +354,7 @@ impl PageServerConf {
             no_sync,
             wal_receiver_protocol,
             page_service_pipelining,
+            get_vectored_concurrent_io,
         } = config_toml;
 
         let mut conf = PageServerConf {
@@ -396,6 +399,7 @@ impl PageServerConf {
             import_pgdata_aws_endpoint_url,
             wal_receiver_protocol,
             page_service_pipelining,
+            get_vectored_concurrent_io,
 
             // ------------------------------------------------------------
             // fields that require additional validation or custom handling

pageserver/src/http/openapi_spec.yml

@@ -984,6 +984,8 @@ components:
           type: string
         compaction_threshold:
           type: string
+        compaction_upper_limit:
+          type: string
         image_creation_threshold:
           type: integer
         walreceiver_connect_timeout:

pageserver/src/http/routes.rs

@@ -10,6 +10,7 @@ use std::time::Duration;
 
 use anyhow::{anyhow, Context, Result};
 use enumset::EnumSet;
+use futures::future::join_all;
 use futures::StreamExt;
 use futures::TryFutureExt;
 use humantime::format_rfc3339;
@@ -40,6 +41,7 @@ use pageserver_api::models::TenantShardSplitRequest;
 use pageserver_api::models::TenantShardSplitResponse;
 use pageserver_api::models::TenantSorting;
 use pageserver_api::models::TenantState;
+use pageserver_api::models::TenantWaitLsnRequest;
 use pageserver_api::models::TimelineArchivalConfigRequest;
 use pageserver_api::models::TimelineCreateRequestMode;
 use pageserver_api::models::TimelineCreateRequestModeImportPgdata;
@@ -84,6 +86,7 @@ use crate::tenant::remote_timeline_client::list_remote_tenant_shards;
 use crate::tenant::remote_timeline_client::list_remote_timelines;
 use crate::tenant::secondary::SecondaryController;
 use crate::tenant::size::ModelInputs;
+use crate::tenant::storage_layer::IoConcurrency;
 use crate::tenant::storage_layer::LayerAccessStatsReset;
 use crate::tenant::storage_layer::LayerName;
 use crate::tenant::timeline::import_pgdata;
@@ -94,6 +97,8 @@ use crate::tenant::timeline::CompactOptions;
 use crate::tenant::timeline::CompactRequest;
 use crate::tenant::timeline::CompactionError;
 use crate::tenant::timeline::Timeline;
+use crate::tenant::timeline::WaitLsnTimeout;
+use crate::tenant::timeline::WaitLsnWaiter;
 use crate::tenant::GetTimelineError;
 use crate::tenant::OffloadedTimeline;
 use crate::tenant::{LogicalSizeCalculationCause, PageReconstructError};
@@ -2789,6 +2794,63 @@ async fn secondary_download_handler(
     json_response(status, progress)
 }
 
+async fn wait_lsn_handler(
+    mut request: Request<Body>,
+    cancel: CancellationToken,
+) -> Result<Response<Body>, ApiError> {
+    let tenant_shard_id: TenantShardId = parse_request_param(&request, "tenant_shard_id")?;
+    let wait_lsn_request: TenantWaitLsnRequest = json_request(&mut request).await?;
+
+    let state = get_state(&request);
+    let tenant = state
+        .tenant_manager
+        .get_attached_tenant_shard(tenant_shard_id)?;
+
+    let mut wait_futures = Vec::default();
+    for timeline in tenant.list_timelines() {
+        let Some(lsn) = wait_lsn_request.timelines.get(&timeline.timeline_id) else {
+            continue;
+        };
+
+        let fut = {
+            let timeline = timeline.clone();
+            let ctx = RequestContext::new(TaskKind::MgmtRequest, DownloadBehavior::Error);
+            async move {
+                timeline
+                    .wait_lsn(
+                        *lsn,
+                        WaitLsnWaiter::HttpEndpoint,
+                        WaitLsnTimeout::Custom(wait_lsn_request.timeout),
+                        &ctx,
+                    )
+                    .await
+            }
+        };
+        wait_futures.push(fut);
+    }
+
+    if wait_futures.is_empty() {
+        return json_response(StatusCode::NOT_FOUND, ());
+    }
+
+    let all_done = tokio::select! {
+        results = join_all(wait_futures) => {
+            results.iter().all(|res| res.is_ok())
+        },
+        _ = cancel.cancelled() => {
+            return Err(ApiError::Cancelled);
+        }
+    };
+
+    let status = if all_done {
+        StatusCode::OK
+    } else {
+        StatusCode::ACCEPTED
+    };
+
+    json_response(status, ())
+}
+
 async fn secondary_status_handler(
     request: Request<Body>,
     _cancel: CancellationToken,
@@ -2938,8 +3000,15 @@ async fn list_aux_files(
         active_timeline_of_active_tenant(&state.tenant_manager, tenant_shard_id, timeline_id)
             .await?;
 
+    let io_concurrency = IoConcurrency::spawn_from_conf(
+        state.conf,
+        timeline.gate.enter().map_err(|_| ApiError::Cancelled)?,
+    );
+
     let ctx = RequestContext::new(TaskKind::MgmtRequest, DownloadBehavior::Download);
-    let files = timeline.list_aux_files(body.lsn, &ctx).await?;
+    let files = timeline
+        .list_aux_files(body.lsn, &ctx, io_concurrency)
+        .await?;
     json_response(StatusCode::OK, files)
 }
 
@@ -3569,6 +3638,9 @@ pub fn make_router(
         .post("/v1/tenant/:tenant_shard_id/secondary/download", |r| {
             api_handler(r, secondary_download_handler)
         })
+        .post("/v1/tenant/:tenant_shard_id/wait_lsn", |r| {
+            api_handler(r, wait_lsn_handler)
+        })
         .put("/v1/tenant/:tenant_shard_id/break", |r| {
             testing_api_handler("set tenant state to broken", r, handle_tenant_break)
         })

pageserver/src/metrics.rs

@@ -38,6 +38,9 @@ pub(crate) enum StorageTimeOperation {
     #[strum(serialize = "layer flush")]
     LayerFlush,
 
+    #[strum(serialize = "layer flush delay")]
+    LayerFlushDelay,
+
     #[strum(serialize = "compact")]
     Compact,
 
@@ -126,73 +129,6 @@ pub(crate) static INITDB_RUN_TIME: Lazy<Histogram> = Lazy::new(|| {
     .expect("failed to define metric")
 });
 
-// Metrics collected on operations on the storage repository.
-#[derive(
-    Clone, Copy, enum_map::Enum, strum_macros::EnumString, strum_macros::Display, IntoStaticStr,
-)]
-pub(crate) enum GetKind {
-    Singular,
-    Vectored,
-}
-
-pub(crate) struct ReconstructTimeMetrics {
-    singular: Histogram,
-    vectored: Histogram,
-}
-
-pub(crate) static RECONSTRUCT_TIME: Lazy<ReconstructTimeMetrics> = Lazy::new(|| {
-    let inner = register_histogram_vec!(
-        "pageserver_getpage_reconstruct_seconds",
-        "Time spent in reconstruct_value (reconstruct a page from deltas)",
-        &["get_kind"],
-        CRITICAL_OP_BUCKETS.into(),
-    )
-    .expect("failed to define a metric");
-
-    ReconstructTimeMetrics {
-        singular: inner.with_label_values(&[GetKind::Singular.into()]),
-        vectored: inner.with_label_values(&[GetKind::Vectored.into()]),
-    }
-});
-
-impl ReconstructTimeMetrics {
-    pub(crate) fn for_get_kind(&self, get_kind: GetKind) -> &Histogram {
-        match get_kind {
-            GetKind::Singular => &self.singular,
-            GetKind::Vectored => &self.vectored,
-        }
-    }
-}
-
-pub(crate) struct ReconstructDataTimeMetrics {
-    singular: Histogram,
-    vectored: Histogram,
-}
-
-impl ReconstructDataTimeMetrics {
-    pub(crate) fn for_get_kind(&self, get_kind: GetKind) -> &Histogram {
-        match get_kind {
-            GetKind::Singular => &self.singular,
-            GetKind::Vectored => &self.vectored,
-        }
-    }
-}
-
-pub(crate) static GET_RECONSTRUCT_DATA_TIME: Lazy<ReconstructDataTimeMetrics> = Lazy::new(|| {
-    let inner = register_histogram_vec!(
-        "pageserver_getpage_get_reconstruct_data_seconds",
-        "Time spent in get_reconstruct_value_data",
-        &["get_kind"],
-        CRITICAL_OP_BUCKETS.into(),
-    )
-    .expect("failed to define a metric");
-
-    ReconstructDataTimeMetrics {
-        singular: inner.with_label_values(&[GetKind::Singular.into()]),
-        vectored: inner.with_label_values(&[GetKind::Vectored.into()]),
-    }
-});
-
 pub(crate) struct GetVectoredLatency {
     map: EnumMap<TaskKind, Option<Histogram>>,
 }
@@ -2550,12 +2486,19 @@ impl StorageTimeMetricsTimer {
         }
     }
 
-    /// Record the time from creation to now.
-    pub fn stop_and_record(self) {
-        let duration = self.start.elapsed().as_secs_f64();
-        self.metrics.timeline_sum.inc_by(duration);
+    /// Returns the elapsed duration of the timer.
+    pub fn elapsed(&self) -> Duration {
+        self.start.elapsed()
+    }
+
+    /// Record the time from creation to now and return it.
+    pub fn stop_and_record(self) -> Duration {
+        let duration = self.elapsed();
+        let seconds = duration.as_secs_f64();
+        self.metrics.timeline_sum.inc_by(seconds);
         self.metrics.timeline_count.inc();
-        self.metrics.global_histogram.observe(duration);
+        self.metrics.global_histogram.observe(seconds);
+        duration
     }
 
     /// Turns this timer into a timer, which will always record -- usually this means recording
@@ -2575,6 +2518,13 @@ impl Drop for AlwaysRecordingStorageTimeMetricsTimer {
     }
 }
 
+impl AlwaysRecordingStorageTimeMetricsTimer {
+    /// Returns the elapsed duration of the timer.
+    pub fn elapsed(&self) -> Duration {
+        self.0.as_ref().expect("not dropped yet").elapsed()
+    }
+}
+
 /// Timing facilities for an globally histogrammed metric, which is supported by per tenant and
 /// timeline total sum and count.
 #[derive(Clone, Debug)]
@@ -2627,6 +2577,7 @@ pub(crate) struct TimelineMetrics {
     shard_id: String,
     timeline_id: String,
     pub flush_time_histo: StorageTimeMetrics,
+    pub flush_delay_histo: StorageTimeMetrics,
     pub flush_wait_upload_time_gauge: Gauge,
     pub compact_time_histo: StorageTimeMetrics,
     pub create_images_time_histo: StorageTimeMetrics,
@@ -2673,6 +2624,12 @@ impl TimelineMetrics {
             &shard_id,
             &timeline_id,
         );
+        let flush_delay_histo = StorageTimeMetrics::new(
+            StorageTimeOperation::LayerFlushDelay,
+            &tenant_id,
+            &shard_id,
+            &timeline_id,
+        );
         let flush_wait_upload_time_gauge = FLUSH_WAIT_UPLOAD_TIME
             .get_metric_with_label_values(&[&tenant_id, &shard_id, &timeline_id])
             .unwrap();
@@ -2821,6 +2778,7 @@ impl TimelineMetrics {
             shard_id,
             timeline_id,
             flush_time_histo,
+            flush_delay_histo,
             flush_wait_upload_time_gauge,
             compact_time_histo,
             create_images_time_histo,
@@ -3919,7 +3877,6 @@ pub fn preinitialize_metrics(conf: &'static PageServerConf) {
     });
 
     // Custom
-    Lazy::force(&RECONSTRUCT_TIME);
     Lazy::force(&BASEBACKUP_QUERY_TIME);
     Lazy::force(&COMPUTE_COMMANDS_COUNTERS);
     Lazy::force(&tokio_epoll_uring::THREAD_LOCAL_METRICS_STORAGE);

pageserver/src/page_service.rs

@@ -39,6 +39,7 @@ use tokio::io::{AsyncWriteExt, BufWriter};
 use tokio::task::JoinHandle;
 use tokio_util::sync::CancellationToken;
 use tracing::*;
+use utils::sync::gate::{Gate, GateGuard};
 use utils::sync::spsc_fold;
 use utils::{
     auth::{Claims, Scope, SwappableJwtAuth},
@@ -61,6 +62,7 @@ use crate::task_mgr::{self, COMPUTE_REQUEST_RUNTIME};
 use crate::tenant::mgr::ShardSelector;
 use crate::tenant::mgr::TenantManager;
 use crate::tenant::mgr::{GetActiveTenantError, GetTenantError, ShardResolveResult};
+use crate::tenant::storage_layer::IoConcurrency;
 use crate::tenant::timeline::{self, WaitLsnError};
 use crate::tenant::GetTimelineError;
 use crate::tenant::PageReconstructError;
@@ -90,6 +92,7 @@ pub struct Listener {
 pub struct Connections {
     cancel: CancellationToken,
     tasks: tokio::task::JoinSet<ConnectionHandlerResult>,
+    gate: Gate,
 }
 
 pub fn spawn(
@@ -110,6 +113,7 @@ pub fn spawn(
     let task = COMPUTE_REQUEST_RUNTIME.spawn(task_mgr::exit_on_panic_or_error(
         "libpq listener",
         libpq_listener_main(
+            conf,
             tenant_manager,
             pg_auth,
             tcp_listener,
@@ -134,11 +138,16 @@ impl Listener {
 }
 impl Connections {
     pub(crate) async fn shutdown(self) {
-        let Self { cancel, mut tasks } = self;
+        let Self {
+            cancel,
+            mut tasks,
+            gate,
+        } = self;
         cancel.cancel();
         while let Some(res) = tasks.join_next().await {
             Self::handle_connection_completion(res);
         }
+        gate.close().await;
     }
 
     fn handle_connection_completion(res: Result<anyhow::Result<()>, tokio::task::JoinError>) {
@@ -158,7 +167,9 @@ impl Connections {
 /// Returns Ok(()) upon cancellation via `cancel`, returning the set of
 /// open connections.
 ///
+#[allow(clippy::too_many_arguments)]
 pub async fn libpq_listener_main(
+    conf: &'static PageServerConf,
     tenant_manager: Arc<TenantManager>,
     auth: Option<Arc<SwappableJwtAuth>>,
     listener: tokio::net::TcpListener,
@@ -168,9 +179,15 @@ pub async fn libpq_listener_main(
     listener_cancel: CancellationToken,
 ) -> Connections {
     let connections_cancel = CancellationToken::new();
+    let connections_gate = Gate::default();
     let mut connection_handler_tasks = tokio::task::JoinSet::default();
 
     loop {
+        let gate_guard = match connections_gate.enter() {
+            Ok(guard) => guard,
+            Err(_) => break,
+        };
+
         let accepted = tokio::select! {
             biased;
             _ = listener_cancel.cancelled() => break,
@@ -190,6 +207,7 @@ pub async fn libpq_listener_main(
                 let connection_ctx = listener_ctx
                     .detached_child(TaskKind::PageRequestHandler, DownloadBehavior::Download);
                 connection_handler_tasks.spawn(page_service_conn_main(
+                    conf,
                     tenant_manager.clone(),
                     local_auth,
                     socket,
@@ -197,6 +215,7 @@ pub async fn libpq_listener_main(
                     pipelining_config.clone(),
                     connection_ctx,
                     connections_cancel.child_token(),
+                    gate_guard,
                 ));
             }
             Err(err) => {
@@ -211,13 +230,16 @@ pub async fn libpq_listener_main(
     Connections {
         cancel: connections_cancel,
         tasks: connection_handler_tasks,
+        gate: connections_gate,
     }
 }
 
 type ConnectionHandlerResult = anyhow::Result<()>;
 
 #[instrument(skip_all, fields(peer_addr))]
+#[allow(clippy::too_many_arguments)]
 async fn page_service_conn_main(
+    conf: &'static PageServerConf,
     tenant_manager: Arc<TenantManager>,
     auth: Option<Arc<SwappableJwtAuth>>,
     socket: tokio::net::TcpStream,
@@ -225,6 +247,7 @@ async fn page_service_conn_main(
     pipelining_config: PageServicePipeliningConfig,
     connection_ctx: RequestContext,
     cancel: CancellationToken,
+    gate_guard: GateGuard,
 ) -> ConnectionHandlerResult {
     let _guard = LIVE_CONNECTIONS
         .with_label_values(&["page_service"])
@@ -274,11 +297,13 @@ async fn page_service_conn_main(
     // But it's in a shared crate, so, we store connection_ctx inside PageServerHandler
     // and create the per-query context in process_query ourselves.
     let mut conn_handler = PageServerHandler::new(
+        conf,
         tenant_manager,
         auth,
         pipelining_config,
         connection_ctx,
         cancel.clone(),
+        gate_guard,
     );
     let pgbackend = PostgresBackend::new_from_io(socket, peer_addr, auth_type, None)?;
 
@@ -310,6 +335,7 @@ async fn page_service_conn_main(
 }
 
 struct PageServerHandler {
+    conf: &'static PageServerConf,
     auth: Option<Arc<SwappableJwtAuth>>,
     claims: Option<Claims>,
 
@@ -325,6 +351,8 @@ struct PageServerHandler {
     timeline_handles: Option<TimelineHandles>,
 
     pipelining_config: PageServicePipeliningConfig,
+
+    gate_guard: GateGuard,
 }
 
 struct TimelineHandles {
@@ -634,19 +662,23 @@ impl BatchedFeMessage {
 
 impl PageServerHandler {
     pub fn new(
+        conf: &'static PageServerConf,
         tenant_manager: Arc<TenantManager>,
         auth: Option<Arc<SwappableJwtAuth>>,
         pipelining_config: PageServicePipeliningConfig,
         connection_ctx: RequestContext,
         cancel: CancellationToken,
+        gate_guard: GateGuard,
     ) -> Self {
         PageServerHandler {
+            conf,
             auth,
             claims: None,
             connection_ctx,
             timeline_handles: Some(TimelineHandles::new(tenant_manager)),
             cancel,
             pipelining_config,
+            gate_guard,
         }
     }
 
@@ -1015,6 +1047,7 @@ impl PageServerHandler {
         &mut self,
         pgb_writer: &mut PostgresBackend<IO>,
         batch: BatchedFeMessage,
+        io_concurrency: IoConcurrency,
         cancel: &CancellationToken,
         protocol_version: PagestreamProtocolVersion,
         ctx: &RequestContext,
@@ -1084,6 +1117,7 @@ impl PageServerHandler {
                                 &*shard.upgrade()?,
                                 effective_request_lsn,
                                 pages,
+                                io_concurrency,
                                 ctx,
                             )
                             .instrument(span.clone())
@@ -1288,6 +1322,17 @@ impl PageServerHandler {
             }
         }
 
+        let io_concurrency = IoConcurrency::spawn_from_conf(
+            self.conf,
+            match self.gate_guard.try_clone() {
+                Ok(guard) => guard,
+                Err(_) => {
+                    info!("shutdown request received in page handler");
+                    return Err(QueryError::Shutdown);
+                }
+            },
+        );
+
         let pgb_reader = pgb
             .split()
             .context("implementation error: split pgb into reader and writer")?;
@@ -1309,6 +1354,7 @@ impl PageServerHandler {
                     request_span,
                     pipelining_config,
                     protocol_version,
+                    io_concurrency,
                     &ctx,
                 )
                 .await
@@ -1322,6 +1368,7 @@ impl PageServerHandler {
                     timeline_handles,
                     request_span,
                     protocol_version,
+                    io_concurrency,
                     &ctx,
                 )
                 .await
@@ -1349,6 +1396,7 @@ impl PageServerHandler {
         mut timeline_handles: TimelineHandles,
         request_span: Span,
         protocol_version: PagestreamProtocolVersion,
+        io_concurrency: IoConcurrency,
         ctx: &RequestContext,
     ) -> (
         (PostgresBackendReader<IO>, TimelineHandles),
@@ -1383,7 +1431,14 @@ impl PageServerHandler {
             };
 
             let err = self
-                .pagesteam_handle_batched_message(pgb_writer, msg, &cancel, protocol_version, ctx)
+                .pagesteam_handle_batched_message(
+                    pgb_writer,
+                    msg,
+                    io_concurrency.clone(),
+                    &cancel,
+                    protocol_version,
+                    ctx,
+                )
                 .await;
             match err {
                 Ok(()) => {}
@@ -1407,6 +1462,7 @@ impl PageServerHandler {
         request_span: Span,
         pipelining_config: PageServicePipeliningConfigPipelined,
         protocol_version: PagestreamProtocolVersion,
+        io_concurrency: IoConcurrency,
         ctx: &RequestContext,
     ) -> (
         (PostgresBackendReader<IO>, TimelineHandles),
@@ -1550,6 +1606,7 @@ impl PageServerHandler {
                     self.pagesteam_handle_batched_message(
                         pgb_writer,
                         batch,
+                        io_concurrency.clone(),
                         &cancel,
                         protocol_version,
                         &ctx,
@@ -1651,6 +1708,7 @@ impl PageServerHandler {
                 .wait_lsn(
                     not_modified_since,
                     crate::tenant::timeline::WaitLsnWaiter::PageService,
+                    timeline::WaitLsnTimeout::Default,
                     ctx,
                 )
                 .await?;
@@ -1806,6 +1864,7 @@ impl PageServerHandler {
         timeline: &Timeline,
         effective_lsn: Lsn,
         requests: smallvec::SmallVec<[BatchedGetPageRequest; 1]>,
+        io_concurrency: IoConcurrency,
         ctx: &RequestContext,
     ) -> Vec<Result<(PagestreamBeMessage, SmgrOpTimer), BatchedPageStreamError>> {
         debug_assert_current_span_has_tenant_and_timeline_id();
@@ -1832,6 +1891,7 @@ impl PageServerHandler {
             .get_rel_page_at_lsn_batched(
                 requests.iter().map(|p| (&p.req.rel, &p.req.blkno)),
                 effective_lsn,
+                io_concurrency,
                 ctx,
             )
             .await;
@@ -1985,6 +2045,7 @@ impl PageServerHandler {
                 .wait_lsn(
                     lsn,
                     crate::tenant::timeline::WaitLsnWaiter::PageService,
+                    crate::tenant::timeline::WaitLsnTimeout::Default,
                     ctx,
                 )
                 .await?;

pageserver/src/pgdatadir_mapping.rs

@@ -17,6 +17,7 @@ use crate::span::{
     debug_assert_current_span_has_tenant_and_timeline_id,
     debug_assert_current_span_has_tenant_and_timeline_id_no_shard_id,
 };
+use crate::tenant::storage_layer::IoConcurrency;
 use crate::tenant::timeline::GetVectoredError;
 use anyhow::{ensure, Context};
 use bytes::{Buf, Bytes, BytesMut};
@@ -200,6 +201,7 @@ impl Timeline {
         blknum: BlockNumber,
         version: Version<'_>,
         ctx: &RequestContext,
+        io_concurrency: IoConcurrency,
     ) -> Result<Bytes, PageReconstructError> {
         match version {
             Version::Lsn(effective_lsn) => {
@@ -208,6 +210,7 @@ impl Timeline {
                     .get_rel_page_at_lsn_batched(
                         pages.iter().map(|(tag, blknum)| (tag, blknum)),
                         effective_lsn,
+                        io_concurrency.clone(),
                         ctx,
                     )
                     .await;
@@ -246,6 +249,7 @@ impl Timeline {
         &self,
         pages: impl ExactSizeIterator<Item = (&RelTag, &BlockNumber)>,
         effective_lsn: Lsn,
+        io_concurrency: IoConcurrency,
         ctx: &RequestContext,
     ) -> Vec<Result<Bytes, PageReconstructError>> {
         debug_assert_current_span_has_tenant_and_timeline_id();
@@ -309,7 +313,10 @@ impl Timeline {
             acc.to_keyspace()
         };
 
-        match self.get_vectored(keyspace, effective_lsn, ctx).await {
+        match self
+            .get_vectored(keyspace, effective_lsn, io_concurrency, ctx)
+            .await
+        {
             Ok(results) => {
                 for (key, res) in results {
                     let mut key_slots = keys_slots.remove(&key).unwrap().into_iter();
@@ -889,9 +896,15 @@ impl Timeline {
         &self,
         lsn: Lsn,
         ctx: &RequestContext,
+        io_concurrency: IoConcurrency,
     ) -> Result<HashMap<String, Bytes>, PageReconstructError> {
         let kv = self
-            .scan(KeySpace::single(Key::metadata_aux_key_range()), lsn, ctx)
+            .scan(
+                KeySpace::single(Key::metadata_aux_key_range()),
+                lsn,
+                ctx,
+                io_concurrency,
+            )
             .await?;
         let mut result = HashMap::new();
         let mut sz = 0;
@@ -914,8 +927,9 @@ impl Timeline {
         &self,
         lsn: Lsn,
         ctx: &RequestContext,
+        io_concurrency: IoConcurrency,
     ) -> Result<(), PageReconstructError> {
-        self.list_aux_files_v2(lsn, ctx).await?;
+        self.list_aux_files_v2(lsn, ctx, io_concurrency).await?;
         Ok(())
     }
 
@@ -923,17 +937,24 @@ impl Timeline {
         &self,
         lsn: Lsn,
         ctx: &RequestContext,
+        io_concurrency: IoConcurrency,
     ) -> Result<HashMap<String, Bytes>, PageReconstructError> {
-        self.list_aux_files_v2(lsn, ctx).await
+        self.list_aux_files_v2(lsn, ctx, io_concurrency).await
     }
 
     pub(crate) async fn get_replorigins(
         &self,
         lsn: Lsn,
         ctx: &RequestContext,
+        io_concurrency: IoConcurrency,
     ) -> Result<HashMap<RepOriginId, Lsn>, PageReconstructError> {
         let kv = self
-            .scan(KeySpace::single(repl_origin_key_range()), lsn, ctx)
+            .scan(
+                KeySpace::single(repl_origin_key_range()),
+                lsn,
+                ctx,
+                io_concurrency,
+            )
             .await?;
         let mut result = HashMap::new();
         for (k, v) in kv {
@@ -2432,7 +2453,11 @@ mod tests {
             ("foo/bar2".to_string(), Bytes::from_static(b"content2")),
         ]);
 
-        let readback = tline.list_aux_files(Lsn(0x1008), &ctx).await?;
+        let io_concurrency = IoConcurrency::spawn_for_test();
+
+        let readback = tline
+            .list_aux_files(Lsn(0x1008), &ctx, io_concurrency.clone())
+            .await?;
         assert_eq!(readback, expect_1008);
 
         // Second modification: update one key, remove the other
@@ -2444,11 +2469,15 @@ mod tests {
         let expect_2008 =
             HashMap::from([("foo/bar1".to_string(), Bytes::from_static(b"content3"))]);
 
-        let readback = tline.list_aux_files(Lsn(0x2008), &ctx).await?;
+        let readback = tline
+            .list_aux_files(Lsn(0x2008), &ctx, io_concurrency.clone())
+            .await?;
         assert_eq!(readback, expect_2008);
 
         // Reading back in time works
-        let readback = tline.list_aux_files(Lsn(0x1008), &ctx).await?;
+        let readback = tline
+            .list_aux_files(Lsn(0x1008), &ctx, io_concurrency.clone())
+            .await?;
         assert_eq!(readback, expect_1008);
 
         Ok(())

pageserver/src/tenant.rs

@@ -37,6 +37,8 @@ use remote_timeline_client::manifest::{
     OffloadedTimelineManifest, TenantManifest, LATEST_TENANT_MANIFEST_VERSION,
 };
 use remote_timeline_client::UploadQueueNotReadyError;
+use remote_timeline_client::FAILED_REMOTE_OP_RETRIES;
+use remote_timeline_client::FAILED_UPLOAD_WARN_THRESHOLD;
 use std::collections::BTreeMap;
 use std::fmt;
 use std::future::Future;
@@ -2558,7 +2560,12 @@ impl Tenant {
                     // sizes etc. and that would get confused if the previous page versions
                     // are not in the repository yet.
                     ancestor_timeline
-                        .wait_lsn(*lsn, timeline::WaitLsnWaiter::Tenant, ctx)
+                        .wait_lsn(
+                            *lsn,
+                            timeline::WaitLsnWaiter::Tenant,
+                            timeline::WaitLsnTimeout::Default,
+                            ctx,
+                        )
                         .await
                         .map_err(|e| match e {
                             e @ (WaitLsnError::Timeout(_) | WaitLsnError::BadState { .. }) => {
@@ -3809,6 +3816,13 @@ impl Tenant {
             .unwrap_or(self.conf.default_tenant_conf.compaction_threshold)
     }
 
+    pub fn get_compaction_upper_limit(&self) -> usize {
+        let tenant_conf = self.tenant_conf.load().tenant_conf.clone();
+        tenant_conf
+            .compaction_upper_limit
+            .unwrap_or(self.conf.default_tenant_conf.compaction_upper_limit)
+    }
+
     pub fn get_gc_horizon(&self) -> u64 {
         let tenant_conf = self.tenant_conf.load().tenant_conf.clone();
         tenant_conf
@@ -5308,27 +5322,37 @@ impl Tenant {
             return Ok(());
         }
 
-        upload_tenant_manifest(
-            &self.remote_storage,
-            &self.tenant_shard_id,
-            self.generation,
-            &manifest,
+        // Remote storage does no retries internally, so wrap it
+        match backoff::retry(
+            || async {
+                upload_tenant_manifest(
+                    &self.remote_storage,
+                    &self.tenant_shard_id,
+                    self.generation,
+                    &manifest,
+                    &self.cancel,
+                )
+                .await
+            },
+            |_e| self.cancel.is_cancelled(),
+            FAILED_UPLOAD_WARN_THRESHOLD,
+            FAILED_REMOTE_OP_RETRIES,
+            "uploading tenant manifest",
             &self.cancel,
         )
         .await
-        .map_err(|e| {
-            if self.cancel.is_cancelled() {
-                TenantManifestError::Cancelled
-            } else {
-                TenantManifestError::RemoteStorage(e)
+        {
+            None => Err(TenantManifestError::Cancelled),
+            Some(Err(_)) if self.cancel.is_cancelled() => Err(TenantManifestError::Cancelled),
+            Some(Err(e)) => Err(TenantManifestError::RemoteStorage(e)),
+            Some(Ok(_)) => {
+                // Store the successfully uploaded manifest, so that future callers can avoid
+                // re-uploading the same thing.
+                *guard = Some(manifest);
+
+                Ok(())
             }
-        })?;
-
-        // Store the successfully uploaded manifest, so that future callers can avoid
-        // re-uploading the same thing.
-        *guard = Some(manifest);
-
-        Ok(())
+        }
     }
 }
 
@@ -5452,7 +5476,11 @@ pub(crate) mod harness {
                 compaction_target_size: Some(tenant_conf.compaction_target_size),
                 compaction_period: Some(tenant_conf.compaction_period),
                 compaction_threshold: Some(tenant_conf.compaction_threshold),
+                compaction_upper_limit: Some(tenant_conf.compaction_upper_limit),
                 compaction_algorithm: Some(tenant_conf.compaction_algorithm),
+                l0_flush_delay_threshold: tenant_conf.l0_flush_delay_threshold,
+                l0_flush_stall_threshold: tenant_conf.l0_flush_stall_threshold,
+                l0_flush_wait_upload: Some(tenant_conf.l0_flush_wait_upload),
                 gc_horizon: Some(tenant_conf.gc_horizon),
                 gc_period: Some(tenant_conf.gc_period),
                 image_creation_threshold: Some(tenant_conf.image_creation_threshold),
@@ -5476,6 +5504,11 @@ pub(crate) mod harness {
                 timeline_offloading: Some(tenant_conf.timeline_offloading),
                 wal_receiver_protocol_override: tenant_conf.wal_receiver_protocol_override,
                 rel_size_v2_enabled: tenant_conf.rel_size_v2_enabled,
+                gc_compaction_enabled: Some(tenant_conf.gc_compaction_enabled),
+                gc_compaction_initial_threshold_kb: Some(
+                    tenant_conf.gc_compaction_initial_threshold_kb,
+                ),
+                gc_compaction_ratio_percent: Some(tenant_conf.gc_compaction_ratio_percent),
             }
         }
     }
@@ -5709,7 +5742,7 @@ mod tests {
     use pageserver_api::value::Value;
     use pageserver_compaction::helpers::overlaps_with;
     use rand::{thread_rng, Rng};
-    use storage_layer::PersistentLayerKey;
+    use storage_layer::{IoConcurrency, PersistentLayerKey};
     use tests::storage_layer::ValuesReconstructState;
     use tests::timeline::{GetVectoredError, ShutdownMode};
     use timeline::{CompactOptions, DeltaLayerTestDesc};
@@ -6490,6 +6523,7 @@ mod tests {
     async fn test_get_vectored() -> anyhow::Result<()> {
         let harness = TenantHarness::create("test_get_vectored").await?;
         let (tenant, ctx) = harness.load().await;
+        let io_concurrency = IoConcurrency::spawn_for_test();
         let tline = tenant
             .create_test_timeline(TIMELINE_ID, Lsn(0x08), DEFAULT_PG_VERSION, &ctx)
             .await?;
@@ -6554,7 +6588,7 @@ mod tests {
                 .get_vectored_impl(
                     read.clone(),
                     reads_lsn,
-                    &mut ValuesReconstructState::new(),
+                    &mut ValuesReconstructState::new(io_concurrency.clone()),
                     &ctx,
                 )
                 .await;
@@ -6601,6 +6635,7 @@ mod tests {
         let harness = TenantHarness::create("test_get_vectored_aux_files").await?;
 
         let (tenant, ctx) = harness.load().await;
+        let io_concurrency = IoConcurrency::spawn_for_test();
         let tline = tenant
             .create_empty_timeline(TIMELINE_ID, Lsn(0), DEFAULT_PG_VERSION, &ctx)
             .await?;
@@ -6635,7 +6670,7 @@ mod tests {
             .get_vectored_impl(
                 aux_keyspace.clone(),
                 read_lsn,
-                &mut ValuesReconstructState::new(),
+                &mut ValuesReconstructState::new(io_concurrency.clone()),
                 &ctx,
             )
             .await;
@@ -6683,6 +6718,7 @@ mod tests {
         )
         .await?;
         let (tenant, ctx) = harness.load().await;
+        let io_concurrency = IoConcurrency::spawn_for_test();
 
         let mut current_key = Key::from_hex("010000000033333333444444445500000000").unwrap();
         let gap_at_key = current_key.add(100);
@@ -6783,7 +6819,7 @@ mod tests {
             .get_vectored_impl(
                 read.clone(),
                 current_lsn,
-                &mut ValuesReconstructState::new(),
+                &mut ValuesReconstructState::new(io_concurrency.clone()),
                 &ctx,
             )
             .await?;
@@ -6826,6 +6862,7 @@ mod tests {
     async fn test_get_vectored_ancestor_descent() -> anyhow::Result<()> {
         let harness = TenantHarness::create("test_get_vectored_on_lsn_axis").await?;
         let (tenant, ctx) = harness.load().await;
+        let io_concurrency = IoConcurrency::spawn_for_test();
 
         let start_key = Key::from_hex("010000000033333333444444445500000000").unwrap();
         let end_key = start_key.add(1000);
@@ -6918,7 +6955,7 @@ mod tests {
                         ranges: vec![child_gap_at_key..child_gap_at_key.next()],
                     },
                     query_lsn,
-                    &mut ValuesReconstructState::new(),
+                    &mut ValuesReconstructState::new(io_concurrency.clone()),
                     &ctx,
                 )
                 .await;
@@ -7364,6 +7401,7 @@ mod tests {
     async fn test_metadata_scan() -> anyhow::Result<()> {
         let harness = TenantHarness::create("test_metadata_scan").await?;
         let (tenant, ctx) = harness.load().await;
+        let io_concurrency = IoConcurrency::spawn_for_test();
         let tline = tenant
             .create_test_timeline(TIMELINE_ID, Lsn(0x10), DEFAULT_PG_VERSION, &ctx)
             .await?;
@@ -7417,7 +7455,7 @@ mod tests {
                 .get_vectored_impl(
                     keyspace.clone(),
                     lsn,
-                    &mut ValuesReconstructState::default(),
+                    &mut ValuesReconstructState::new(io_concurrency.clone()),
                     &ctx,
                 )
                 .await?
@@ -7532,6 +7570,7 @@ mod tests {
         let harness = TenantHarness::create("test_aux_file_e2e").await.unwrap();
 
         let (tenant, ctx) = harness.load().await;
+        let io_concurrency = IoConcurrency::spawn_for_test();
 
         let mut lsn = Lsn(0x08);
 
@@ -7551,7 +7590,10 @@ mod tests {
         }
 
         // we can read everything from the storage
-        let files = tline.list_aux_files(lsn, &ctx).await.unwrap();
+        let files = tline
+            .list_aux_files(lsn, &ctx, io_concurrency.clone())
+            .await
+            .unwrap();
         assert_eq!(
             files.get("pg_logical/mappings/test1"),
             Some(&bytes::Bytes::from_static(b"first"))
@@ -7567,7 +7609,10 @@ mod tests {
             modification.commit(&ctx).await.unwrap();
         }
 
-        let files = tline.list_aux_files(lsn, &ctx).await.unwrap();
+        let files = tline
+            .list_aux_files(lsn, &ctx, io_concurrency.clone())
+            .await
+            .unwrap();
         assert_eq!(
             files.get("pg_logical/mappings/test2"),
             Some(&bytes::Bytes::from_static(b"second"))
@@ -7578,7 +7623,10 @@ mod tests {
             .await
             .unwrap();
 
-        let files = child.list_aux_files(lsn, &ctx).await.unwrap();
+        let files = child
+            .list_aux_files(lsn, &ctx, io_concurrency.clone())
+            .await
+            .unwrap();
         assert_eq!(files.get("pg_logical/mappings/test1"), None);
         assert_eq!(files.get("pg_logical/mappings/test2"), None);
     }
@@ -7587,6 +7635,7 @@ mod tests {
     async fn test_metadata_image_creation() -> anyhow::Result<()> {
         let harness = TenantHarness::create("test_metadata_image_creation").await?;
         let (tenant, ctx) = harness.load().await;
+        let io_concurrency = IoConcurrency::spawn_for_test();
         let tline = tenant
             .create_test_timeline(TIMELINE_ID, Lsn(0x10), DEFAULT_PG_VERSION, &ctx)
             .await?;
@@ -7606,8 +7655,9 @@ mod tests {
             keyspace: &KeySpace,
             lsn: Lsn,
             ctx: &RequestContext,
+            io_concurrency: IoConcurrency,
         ) -> anyhow::Result<(BTreeMap<Key, Result<Bytes, PageReconstructError>>, usize)> {
-            let mut reconstruct_state = ValuesReconstructState::default();
+            let mut reconstruct_state = ValuesReconstructState::new(io_concurrency);
             let res = tline
                 .get_vectored_impl(keyspace.clone(), lsn, &mut reconstruct_state, ctx)
                 .await?;
@@ -7655,7 +7705,8 @@ mod tests {
 
             if iter % 5 == 0 {
                 let (_, before_delta_file_accessed) =
-                    scan_with_statistics(&tline, &keyspace, lsn, &ctx).await?;
+                    scan_with_statistics(&tline, &keyspace, lsn, &ctx, io_concurrency.clone())
+                        .await?;
                 tline
                     .compact(
                         &cancel,
@@ -7669,7 +7720,8 @@ mod tests {
                     )
                     .await?;
                 let (_, after_delta_file_accessed) =
-                    scan_with_statistics(&tline, &keyspace, lsn, &ctx).await?;
+                    scan_with_statistics(&tline, &keyspace, lsn, &ctx, io_concurrency.clone())
+                        .await?;
                 assert!(after_delta_file_accessed < before_delta_file_accessed, "after_delta_file_accessed={after_delta_file_accessed}, before_delta_file_accessed={before_delta_file_accessed}");
                 // Given that we already produced an image layer, there should be no delta layer needed for the scan, but still setting a low threshold there for unforeseen circumstances.
                 assert!(
@@ -7758,6 +7810,7 @@ mod tests {
     async fn test_vectored_missing_metadata_key_reads() -> anyhow::Result<()> {
         let harness = TenantHarness::create("test_vectored_missing_metadata_key_reads").await?;
         let (tenant, ctx) = harness.load().await;
+        let io_concurrency = IoConcurrency::spawn_for_test();
 
         let base_key = Key::from_hex("620000000033333333444444445500000000").unwrap();
         let base_key_child = Key::from_hex("620000000033333333444444445500000001").unwrap();
@@ -7896,7 +7949,7 @@ mod tests {
         );
 
         // test vectored scan on parent timeline
-        let mut reconstruct_state = ValuesReconstructState::new();
+        let mut reconstruct_state = ValuesReconstructState::new(io_concurrency.clone());
         let res = tline
             .get_vectored_impl(
                 KeySpace::single(Key::metadata_key_range()),
@@ -7922,7 +7975,7 @@ mod tests {
         );
 
         // test vectored scan on child timeline
-        let mut reconstruct_state = ValuesReconstructState::new();
+        let mut reconstruct_state = ValuesReconstructState::new(io_concurrency.clone());
         let res = child
             .get_vectored_impl(
                 KeySpace::single(Key::metadata_key_range()),
@@ -7960,7 +8013,9 @@ mod tests {
         lsn: Lsn,
         ctx: &RequestContext,
     ) -> Result<Option<Bytes>, GetVectoredError> {
-        let mut reconstruct_state = ValuesReconstructState::new();
+        let io_concurrency =
+            IoConcurrency::spawn_from_conf(tline.conf, tline.gate.enter().unwrap());
+        let mut reconstruct_state = ValuesReconstructState::new(io_concurrency);
         let mut res = tline
             .get_vectored_impl(
                 KeySpace::single(key..key.next()),
@@ -8061,6 +8116,7 @@ mod tests {
             .await
             .unwrap();
         let (tenant, ctx) = harness.load().await;
+        let io_concurrency = IoConcurrency::spawn_for_test();
 
         let key0 = Key::from_hex("620000000033333333444444445500000000").unwrap();
         let key1 = Key::from_hex("620000000033333333444444445500000001").unwrap();
@@ -8120,7 +8176,7 @@ mod tests {
 
         // Image layers are created at last_record_lsn
         let images = tline
-            .inspect_image_layers(Lsn(0x40), &ctx)
+            .inspect_image_layers(Lsn(0x40), &ctx, io_concurrency.clone())
             .await
             .unwrap()
             .into_iter()
@@ -8135,6 +8191,7 @@ mod tests {
             .await
             .unwrap();
         let (tenant, ctx) = harness.load().await;
+        let io_concurrency = IoConcurrency::spawn_for_test();
 
         let key1 = Key::from_hex("620000000033333333444444445500000001").unwrap();
         let key2 = Key::from_hex("620000000033333333444444445500000002").unwrap();
@@ -8185,7 +8242,7 @@ mod tests {
 
         // Image layers are created at last_record_lsn
         let images = tline
-            .inspect_image_layers(Lsn(0x30), &ctx)
+            .inspect_image_layers(Lsn(0x30), &ctx, io_concurrency.clone())
             .await
             .unwrap()
             .into_iter()
@@ -8198,6 +8255,7 @@ mod tests {
     async fn test_simple_bottom_most_compaction_images() -> anyhow::Result<()> {
         let harness = TenantHarness::create("test_simple_bottom_most_compaction_images").await?;
         let (tenant, ctx) = harness.load().await;
+        let io_concurrency = IoConcurrency::spawn_for_test();
 
         fn get_key(id: u32) -> Key {
             // using aux key here b/c they are guaranteed to be inside `collect_keyspace`.
@@ -8339,7 +8397,7 @@ mod tests {
 
         // Check if the image layer at the GC horizon contains exactly what we want
         let image_at_gc_horizon = tline
-            .inspect_image_layers(Lsn(0x30), &ctx)
+            .inspect_image_layers(Lsn(0x30), &ctx, io_concurrency.clone())
             .await
             .unwrap()
             .into_iter()
@@ -10052,7 +10110,12 @@ mod tests {
 
         let keyspace = KeySpace::single(get_key(0)..get_key(10));
         let results = tline
-            .get_vectored(keyspace, delta_layer_end_lsn, &ctx)
+            .get_vectored(
+                keyspace,
+                delta_layer_end_lsn,
+                IoConcurrency::sequential(),
+                &ctx,
+            )
             .await
             .expect("No vectored errors");
         for (key, res) in results {

pageserver/src/tenant/config.rs

@@ -277,10 +277,26 @@ pub struct TenantConfOpt {
     #[serde(default)]
     pub compaction_threshold: Option<usize>,
 
+    #[serde(skip_serializing_if = "Option::is_none")]
+    #[serde(default)]
+    pub compaction_upper_limit: Option<usize>,
+
     #[serde(skip_serializing_if = "Option::is_none")]
     #[serde(default)]
     pub compaction_algorithm: Option<CompactionAlgorithmSettings>,
 
+    #[serde(skip_serializing_if = "Option::is_none")]
+    #[serde(default)]
+    pub l0_flush_delay_threshold: Option<usize>,
+
+    #[serde(skip_serializing_if = "Option::is_none")]
+    #[serde(default)]
+    pub l0_flush_stall_threshold: Option<usize>,
+
+    #[serde(skip_serializing_if = "Option::is_none")]
+    #[serde(default)]
+    pub l0_flush_wait_upload: Option<bool>,
+
     #[serde(skip_serializing_if = "Option::is_none")]
     #[serde(default)]
     pub gc_horizon: Option<u64>,
@@ -360,6 +376,15 @@ pub struct TenantConfOpt {
 
     #[serde(skip_serializing_if = "Option::is_none")]
     pub rel_size_v2_enabled: Option<bool>,
+
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub gc_compaction_enabled: Option<bool>,
+
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub gc_compaction_initial_threshold_kb: Option<u64>,
+
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub gc_compaction_ratio_percent: Option<u64>,
 }
 
 impl TenantConfOpt {
@@ -380,11 +405,23 @@ impl TenantConfOpt {
             compaction_threshold: self
                 .compaction_threshold
                 .unwrap_or(global_conf.compaction_threshold),
+            compaction_upper_limit: self
+                .compaction_upper_limit
+                .unwrap_or(global_conf.compaction_upper_limit),
             compaction_algorithm: self
                 .compaction_algorithm
                 .as_ref()
                 .unwrap_or(&global_conf.compaction_algorithm)
                 .clone(),
+            l0_flush_delay_threshold: self
+                .l0_flush_delay_threshold
+                .or(global_conf.l0_flush_delay_threshold),
+            l0_flush_stall_threshold: self
+                .l0_flush_stall_threshold
+                .or(global_conf.l0_flush_stall_threshold),
+            l0_flush_wait_upload: self
+                .l0_flush_wait_upload
+                .unwrap_or(global_conf.l0_flush_wait_upload),
             gc_horizon: self.gc_horizon.unwrap_or(global_conf.gc_horizon),
             gc_period: self.gc_period.unwrap_or(global_conf.gc_period),
             image_creation_threshold: self
@@ -429,6 +466,15 @@ impl TenantConfOpt {
                 .wal_receiver_protocol_override
                 .or(global_conf.wal_receiver_protocol_override),
             rel_size_v2_enabled: self.rel_size_v2_enabled.or(global_conf.rel_size_v2_enabled),
+            gc_compaction_enabled: self
+                .gc_compaction_enabled
+                .unwrap_or(global_conf.gc_compaction_enabled),
+            gc_compaction_initial_threshold_kb: self
+                .gc_compaction_initial_threshold_kb
+                .unwrap_or(global_conf.gc_compaction_initial_threshold_kb),
+            gc_compaction_ratio_percent: self
+                .gc_compaction_ratio_percent
+                .unwrap_or(global_conf.gc_compaction_ratio_percent),
         }
     }
 
@@ -439,7 +485,11 @@ impl TenantConfOpt {
             mut compaction_target_size,
             mut compaction_period,
             mut compaction_threshold,
+            mut compaction_upper_limit,
             mut compaction_algorithm,
+            mut l0_flush_delay_threshold,
+            mut l0_flush_stall_threshold,
+            mut l0_flush_wait_upload,
             mut gc_horizon,
             mut gc_period,
             mut image_creation_threshold,
@@ -459,6 +509,9 @@ impl TenantConfOpt {
             mut timeline_offloading,
             mut wal_receiver_protocol_override,
             mut rel_size_v2_enabled,
+            mut gc_compaction_enabled,
+            mut gc_compaction_initial_threshold_kb,
+            mut gc_compaction_ratio_percent,
         } = self;
 
         patch.checkpoint_distance.apply(&mut checkpoint_distance);
@@ -474,7 +527,17 @@ impl TenantConfOpt {
             .map(|v| humantime::parse_duration(&v))?
             .apply(&mut compaction_period);
         patch.compaction_threshold.apply(&mut compaction_threshold);
+        patch
+            .compaction_upper_limit
+            .apply(&mut compaction_upper_limit);
         patch.compaction_algorithm.apply(&mut compaction_algorithm);
+        patch
+            .l0_flush_delay_threshold
+            .apply(&mut l0_flush_delay_threshold);
+        patch
+            .l0_flush_stall_threshold
+            .apply(&mut l0_flush_stall_threshold);
+        patch.l0_flush_wait_upload.apply(&mut l0_flush_wait_upload);
         patch.gc_horizon.apply(&mut gc_horizon);
         patch
             .gc_period
@@ -528,6 +591,15 @@ impl TenantConfOpt {
             .wal_receiver_protocol_override
             .apply(&mut wal_receiver_protocol_override);
         patch.rel_size_v2_enabled.apply(&mut rel_size_v2_enabled);
+        patch
+            .gc_compaction_enabled
+            .apply(&mut gc_compaction_enabled);
+        patch
+            .gc_compaction_initial_threshold_kb
+            .apply(&mut gc_compaction_initial_threshold_kb);
+        patch
+            .gc_compaction_ratio_percent
+            .apply(&mut gc_compaction_ratio_percent);
 
         Ok(Self {
             checkpoint_distance,
@@ -535,7 +607,11 @@ impl TenantConfOpt {
             compaction_target_size,
             compaction_period,
             compaction_threshold,
+            compaction_upper_limit,
             compaction_algorithm,
+            l0_flush_delay_threshold,
+            l0_flush_stall_threshold,
+            l0_flush_wait_upload,
             gc_horizon,
             gc_period,
             image_creation_threshold,
@@ -555,6 +631,9 @@ impl TenantConfOpt {
             timeline_offloading,
             wal_receiver_protocol_override,
             rel_size_v2_enabled,
+            gc_compaction_enabled,
+            gc_compaction_initial_threshold_kb,
+            gc_compaction_ratio_percent,
         })
     }
 }
@@ -590,6 +669,10 @@ impl From<TenantConfOpt> for models::TenantConfig {
             compaction_target_size: value.compaction_target_size,
             compaction_period: value.compaction_period.map(humantime),
             compaction_threshold: value.compaction_threshold,
+            compaction_upper_limit: value.compaction_upper_limit,
+            l0_flush_delay_threshold: value.l0_flush_delay_threshold,
+            l0_flush_stall_threshold: value.l0_flush_stall_threshold,
+            l0_flush_wait_upload: value.l0_flush_wait_upload,
             gc_horizon: value.gc_horizon,
             gc_period: value.gc_period.map(humantime),
             image_creation_threshold: value.image_creation_threshold,
@@ -611,6 +694,9 @@ impl From<TenantConfOpt> for models::TenantConfig {
             timeline_offloading: value.timeline_offloading,
             wal_receiver_protocol_override: value.wal_receiver_protocol_override,
             rel_size_v2_enabled: value.rel_size_v2_enabled,
+            gc_compaction_enabled: value.gc_compaction_enabled,
+            gc_compaction_initial_threshold_kb: value.gc_compaction_initial_threshold_kb,
+            gc_compaction_ratio_percent: value.gc_compaction_ratio_percent,
         }
     }
 }

pageserver/src/tenant/layer_map.rs

@@ -57,6 +57,7 @@ use std::collections::{HashMap, VecDeque};
 use std::iter::Peekable;
 use std::ops::Range;
 use std::sync::Arc;
+use tokio::sync::watch;
 use utils::lsn::Lsn;
 
 use historic_layer_coverage::BufferedHistoricLayerCoverage;
@@ -67,7 +68,6 @@ use super::storage_layer::{LayerVisibilityHint, PersistentLayerDesc};
 ///
 /// LayerMap tracks what layers exist on a timeline.
 ///
-#[derive(Default)]
 pub struct LayerMap {
     //
     // 'open_layer' holds the current InMemoryLayer that is accepting new
@@ -93,7 +93,25 @@ pub struct LayerMap {
 
     /// L0 layers have key range Key::MIN..Key::MAX, and locating them using R-Tree search is very inefficient.
     /// So L0 layers are held in l0_delta_layers vector, in addition to the R-tree.
+    ///
+    /// NB: make sure to notify `watch_l0_deltas` on changes.
     l0_delta_layers: Vec<Arc<PersistentLayerDesc>>,
+
+    /// Notifies about L0 delta layer changes, sending the current number of L0 layers.
+    watch_l0_deltas: watch::Sender<usize>,
+}
+
+impl Default for LayerMap {
+    fn default() -> Self {
+        Self {
+            open_layer: Default::default(),
+            next_open_layer_at: Default::default(),
+            frozen_layers: Default::default(),
+            historic: Default::default(),
+            l0_delta_layers: Default::default(),
+            watch_l0_deltas: watch::channel(0).0,
+        }
+    }
 }
 
 /// The primary update API for the layer map.
@@ -466,6 +484,8 @@ impl LayerMap {
 
         if Self::is_l0(&layer_desc.key_range, layer_desc.is_delta) {
             self.l0_delta_layers.push(layer_desc.clone().into());
+            self.watch_l0_deltas
+                .send_replace(self.l0_delta_layers.len());
         }
 
         self.historic.insert(
@@ -488,6 +508,8 @@ impl LayerMap {
             let mut l0_delta_layers = std::mem::take(&mut self.l0_delta_layers);
             l0_delta_layers.retain(|other| other.key() != layer_key);
             self.l0_delta_layers = l0_delta_layers;
+            self.watch_l0_deltas
+                .send_replace(self.l0_delta_layers.len());
             // this assertion is related to use of Arc::ptr_eq in Self::compare_arced_layers,
             // there's a chance that the comparison fails at runtime due to it comparing (pointer,
             // vtable) pairs.
@@ -850,6 +872,11 @@ impl LayerMap {
         &self.l0_delta_layers
     }
 
+    /// Subscribes to L0 delta layer changes, sending the current number of L0 delta layers.
+    pub fn watch_level0_deltas(&self) -> watch::Receiver<usize> {
+        self.watch_l0_deltas.subscribe()
+    }
+
     /// debugging function to print out the contents of the layer map
     #[allow(unused)]
     pub async fn dump(&self, verbose: bool, ctx: &RequestContext) -> Result<()> {

pageserver/src/tenant/mgr.rs

@@ -1643,6 +1643,7 @@ impl TenantManager {
                         .wait_lsn(
                             *target_lsn,
                             crate::tenant::timeline::WaitLsnWaiter::Tenant,
+                            crate::tenant::timeline::WaitLsnTimeout::Default,
                             ctx,
                         )
                         .await

pageserver/src/tenant/remote_timeline_client/index.rs

@@ -77,11 +77,17 @@ pub struct IndexPart {
     ///
     /// None means no aux files have been written to the storage before the point
     /// when this flag is introduced.
+    ///
+    /// This flag is not used any more as all tenants have been transitioned to the new aux file policy.
     #[serde(skip_serializing_if = "Option::is_none", default)]
     pub(crate) last_aux_file_policy: Option<AuxFilePolicy>,
 
     #[serde(skip_serializing_if = "Option::is_none", default)]
     pub(crate) rel_size_migration: Option<RelSizeMigration>,
+
+    /// The LSN of gc-compaction horizon. Once gc-compaction is finished for all layer files below an LSN, this LSN will be updated.
+    #[serde(skip_serializing_if = "Option::is_none", default)]
+    pub(crate) l2_lsn: Option<Lsn>,
 }
 
 #[derive(Debug, Clone, PartialEq, Eq, Serialize, Deserialize)]
@@ -116,10 +122,11 @@ impl IndexPart {
     /// - 9: +gc_blocking
     /// - 10: +import_pgdata
     /// - 11: +rel_size_migration
-    const LATEST_VERSION: usize = 11;
+    /// - 12: +l2_lsn
+    const LATEST_VERSION: usize = 12;
 
     // Versions we may see when reading from a bucket.
-    pub const KNOWN_VERSIONS: &'static [usize] = &[1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11];
+    pub const KNOWN_VERSIONS: &'static [usize] = &[1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12];
 
     pub const FILE_NAME: &'static str = "index_part.json";
 
@@ -136,6 +143,7 @@ impl IndexPart {
             last_aux_file_policy: None,
             import_pgdata: None,
             rel_size_migration: None,
+            l2_lsn: None,
         }
     }
 
@@ -214,6 +222,10 @@ impl LayerFileMetadata {
             shard,
         }
     }
+    /// Helper to get both generation and file size in a tuple
+    pub fn generation_file_size(&self) -> (Generation, u64) {
+        (self.generation, self.file_size)
+    }
 }
 
 /// Limited history of earlier ancestors.
@@ -437,6 +449,7 @@ mod tests {
             last_aux_file_policy: None,
             import_pgdata: None,
             rel_size_migration: None,
+            l2_lsn: None,
         };
 
         let part = IndexPart::from_json_bytes(example.as_bytes()).unwrap();
@@ -483,6 +496,7 @@ mod tests {
             last_aux_file_policy: None,
             import_pgdata: None,
             rel_size_migration: None,
+            l2_lsn: None,
         };
 
         let part = IndexPart::from_json_bytes(example.as_bytes()).unwrap();
@@ -530,6 +544,7 @@ mod tests {
             last_aux_file_policy: None,
             import_pgdata: None,
             rel_size_migration: None,
+            l2_lsn: None,
         };
 
         let part = IndexPart::from_json_bytes(example.as_bytes()).unwrap();
@@ -580,6 +595,7 @@ mod tests {
             last_aux_file_policy: None,
             import_pgdata: None,
             rel_size_migration: None,
+            l2_lsn: None,
         };
 
         let empty_layers_parsed = IndexPart::from_json_bytes(empty_layers_json.as_bytes()).unwrap();
@@ -625,6 +641,7 @@ mod tests {
             last_aux_file_policy: None,
             import_pgdata: None,
             rel_size_migration: None,
+            l2_lsn: None,
         };
 
         let part = IndexPart::from_json_bytes(example.as_bytes()).unwrap();
@@ -673,6 +690,7 @@ mod tests {
             last_aux_file_policy: None,
             import_pgdata: None,
             rel_size_migration: None,
+            l2_lsn: None,
         };
 
         let part = IndexPart::from_json_bytes(example.as_bytes()).unwrap();
@@ -726,6 +744,7 @@ mod tests {
             last_aux_file_policy: Some(AuxFilePolicy::V2),
             import_pgdata: None,
             rel_size_migration: None,
+            l2_lsn: None,
         };
 
         let part = IndexPart::from_json_bytes(example.as_bytes()).unwrap();
@@ -784,6 +803,7 @@ mod tests {
             last_aux_file_policy: Default::default(),
             import_pgdata: None,
             rel_size_migration: None,
+            l2_lsn: None,
         };
 
         let part = IndexPart::from_json_bytes(example.as_bytes()).unwrap();
@@ -843,6 +863,7 @@ mod tests {
             last_aux_file_policy: Default::default(),
             import_pgdata: None,
             rel_size_migration: None,
+            l2_lsn: None,
         };
 
         let part = IndexPart::from_json_bytes(example.as_bytes()).unwrap();
@@ -907,6 +928,7 @@ mod tests {
             archived_at: None,
             import_pgdata: None,
             rel_size_migration: None,
+            l2_lsn: None,
         };
 
         let part = IndexPart::from_json_bytes(example.as_bytes()).unwrap();
@@ -984,6 +1006,7 @@ mod tests {
                 idempotency_key: import_pgdata::index_part_format::IdempotencyKey::new("specified-by-client-218a5213-5044-4562-a28d-d024c5f057f5".to_string()),
             }))),
             rel_size_migration: None,
+            l2_lsn: None,
         };
 
         let part = IndexPart::from_json_bytes(example.as_bytes()).unwrap();
@@ -1062,6 +1085,87 @@ mod tests {
                 idempotency_key: import_pgdata::index_part_format::IdempotencyKey::new("specified-by-client-218a5213-5044-4562-a28d-d024c5f057f5".to_string()),
             }))),
             rel_size_migration: Some(RelSizeMigration::Legacy),
+            l2_lsn: None,
+        };
+
+        let part = IndexPart::from_json_bytes(example.as_bytes()).unwrap();
+        assert_eq!(part, expected);
+    }
+
+    #[test]
+    fn v12_l2_lsn_is_parsed() {
+        let example = r#"{
+            "version": 12,
+            "layer_metadata":{
+                "000000000000000000000000000000000000-FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF__0000000001696070-00000000016960E9": { "file_size": 25600000 },
+                "000000000000000000000000000000000000-FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF__00000000016B59D8-00000000016B5A51": { "file_size": 9007199254741001 }
+            },
+            "disk_consistent_lsn":"0/16960E8",
+            "metadata": {
+                "disk_consistent_lsn": "0/16960E8",
+                "prev_record_lsn": "0/1696070",
+                "ancestor_timeline": "e45a7f37d3ee2ff17dc14bf4f4e3f52e",
+                "ancestor_lsn": "0/0",
+                "latest_gc_cutoff_lsn": "0/1696070",
+                "initdb_lsn": "0/1696070",
+                "pg_version": 14
+            },
+            "gc_blocking": {
+                "started_at": "2024-07-19T09:00:00.123",
+                "reasons": ["DetachAncestor"]
+            },
+            "import_pgdata": {
+                "V1": {
+                    "Done": {
+                        "idempotency_key": "specified-by-client-218a5213-5044-4562-a28d-d024c5f057f5",
+                        "started_at": "2024-11-13T09:23:42.123",
+                        "finished_at": "2024-11-13T09:42:23.123"
+                    }
+                }
+            },
+            "rel_size_migration": "legacy",
+            "l2_lsn": "0/16960E8"
+        }"#;
+
+        let expected = IndexPart {
+            version: 12,
+            layer_metadata: HashMap::from([
+                ("000000000000000000000000000000000000-FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF__0000000001696070-00000000016960E9".parse().unwrap(), LayerFileMetadata {
+                    file_size: 25600000,
+                    generation: Generation::none(),
+                    shard: ShardIndex::unsharded()
+                }),
+                ("000000000000000000000000000000000000-FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF__00000000016B59D8-00000000016B5A51".parse().unwrap(), LayerFileMetadata {
+                    file_size: 9007199254741001,
+                    generation: Generation::none(),
+                    shard: ShardIndex::unsharded()
+                })
+            ]),
+            disk_consistent_lsn: "0/16960E8".parse::<Lsn>().unwrap(),
+            metadata: TimelineMetadata::new(
+                Lsn::from_str("0/16960E8").unwrap(),
+                Some(Lsn::from_str("0/1696070").unwrap()),
+                Some(TimelineId::from_str("e45a7f37d3ee2ff17dc14bf4f4e3f52e").unwrap()),
+                Lsn::INVALID,
+                Lsn::from_str("0/1696070").unwrap(),
+                Lsn::from_str("0/1696070").unwrap(),
+                14,
+            ).with_recalculated_checksum().unwrap(),
+            deleted_at: None,
+            lineage: Default::default(),
+            gc_blocking: Some(GcBlocking {
+                started_at: parse_naive_datetime("2024-07-19T09:00:00.123000000"),
+                reasons: enumset::EnumSet::from_iter([GcBlockingReason::DetachAncestor]),
+            }),
+            last_aux_file_policy: Default::default(),
+            archived_at: None,
+            import_pgdata: Some(import_pgdata::index_part_format::Root::V1(import_pgdata::index_part_format::V1::Done(import_pgdata::index_part_format::Done{
+                started_at: parse_naive_datetime("2024-11-13T09:23:42.123000000"),
+                finished_at: parse_naive_datetime("2024-11-13T09:42:23.123000000"),
+                idempotency_key: import_pgdata::index_part_format::IdempotencyKey::new("specified-by-client-218a5213-5044-4562-a28d-d024c5f057f5".to_string()),
+            }))),
+            rel_size_migration: Some(RelSizeMigration::Legacy),
+            l2_lsn: Some("0/16960E8".parse::<Lsn>().unwrap()),
         };
 
         let part = IndexPart::from_json_bytes(example.as_bytes()).unwrap();

pageserver/src/tenant/secondary/downloader.rs

@@ -559,6 +559,13 @@ impl JobGenerator<PendingDownload, RunningDownload, CompleteDownload, DownloadCo
     }
 }
 
+enum LayerAction {
+    Download,
+    NoAction,
+    Skip,
+    Touch,
+}
+
 /// This type is a convenience to group together the various functions involved in
 /// freshening a secondary tenant.
 struct TenantDownloader<'a> {
@@ -1008,69 +1015,17 @@ impl<'a> TenantDownloader<'a> {
                 return (Err(UpdateError::Restart), touched);
             }
 
-            // Existing on-disk layers: just update their access time.
-            if let Some(on_disk) = timeline_state.on_disk_layers.get(&layer.name) {
-                tracing::debug!("Layer {} is already on disk", layer.name);
-
-                if cfg!(debug_assertions) {
-                    // Debug for https://github.com/neondatabase/neon/issues/6966: check that the files we think
-                    // are already present on disk are really there.
-                    match tokio::fs::metadata(&on_disk.local_path).await {
-                        Ok(meta) => {
-                            tracing::debug!(
-                                "Layer {} present at {}, size {}",
-                                layer.name,
-                                on_disk.local_path,
-                                meta.len(),
-                            );
-                        }
-                        Err(e) => {
-                            tracing::warn!(
-                                "Layer {} not found at {} ({})",
-                                layer.name,
-                                on_disk.local_path,
-                                e
-                            );
-                            debug_assert!(false);
-                        }
-                    }
-                }
-
-                if on_disk.metadata != layer.metadata || on_disk.access_time != layer.access_time {
-                    // We already have this layer on disk.  Update its access time.
-                    tracing::debug!(
-                        "Access time updated for layer {}: {} -> {}",
-                        layer.name,
-                        strftime(&on_disk.access_time),
-                        strftime(&layer.access_time)
-                    );
-                    touched.push(layer);
-                }
-                continue;
-            } else {
-                tracing::debug!("Layer {} not present on disk yet", layer.name);
-            }
-
-            // Eviction: if we evicted a layer, then do not re-download it unless it was accessed more
-            // recently than it was evicted.
-            if let Some(evicted_at) = timeline_state.evicted_at.get(&layer.name) {
-                if &layer.access_time > evicted_at {
-                    tracing::info!(
-                        "Re-downloading evicted layer {}, accessed at {}, evicted at {}",
-                        layer.name,
-                        strftime(&layer.access_time),
-                        strftime(evicted_at)
-                    );
-                } else {
-                    tracing::trace!(
-                        "Not re-downloading evicted layer {}, accessed at {}, evicted at {}",
-                        layer.name,
-                        strftime(&layer.access_time),
-                        strftime(evicted_at)
-                    );
+            match self.layer_action(&timeline_state, &layer).await {
+                LayerAction::Download => (),
+                LayerAction::NoAction => continue,
+                LayerAction::Skip => {
                     self.skip_layer(layer);
                     continue;
                 }
+                LayerAction::Touch => {
+                    touched.push(layer);
+                    continue;
+                }
             }
 
             match self
@@ -1091,6 +1046,86 @@ impl<'a> TenantDownloader<'a> {
         (Ok(()), touched)
     }
 
+    async fn layer_action(
+        &self,
+        timeline_state: &SecondaryDetailTimeline,
+        layer: &HeatMapLayer,
+    ) -> LayerAction {
+        // Existing on-disk layers: just update their access time.
+        if let Some(on_disk) = timeline_state.on_disk_layers.get(&layer.name) {
+            tracing::debug!("Layer {} is already on disk", layer.name);
+
+            if cfg!(debug_assertions) {
+                // Debug for https://github.com/neondatabase/neon/issues/6966: check that the files we think
+                // are already present on disk are really there.
+                match tokio::fs::metadata(&on_disk.local_path).await {
+                    Ok(meta) => {
+                        tracing::debug!(
+                            "Layer {} present at {}, size {}",
+                            layer.name,
+                            on_disk.local_path,
+                            meta.len(),
+                        );
+                    }
+                    Err(e) => {
+                        tracing::warn!(
+                            "Layer {} not found at {} ({})",
+                            layer.name,
+                            on_disk.local_path,
+                            e
+                        );
+                        debug_assert!(false);
+                    }
+                }
+            }
+
+            if on_disk.metadata.generation_file_size() != on_disk.metadata.generation_file_size() {
+                tracing::info!(
+                    "Re-downloading layer {} with changed size or generation: {:?}->{:?}",
+                    layer.name,
+                    on_disk.metadata.generation_file_size(),
+                    on_disk.metadata.generation_file_size()
+                );
+                return LayerAction::Download;
+            }
+            if on_disk.metadata != layer.metadata || on_disk.access_time != layer.access_time {
+                // We already have this layer on disk.  Update its access time.
+                tracing::debug!(
+                    "Access time updated for layer {}: {} -> {}",
+                    layer.name,
+                    strftime(&on_disk.access_time),
+                    strftime(&layer.access_time)
+                );
+                return LayerAction::Touch;
+            }
+            return LayerAction::NoAction;
+        } else {
+            tracing::debug!("Layer {} not present on disk yet", layer.name);
+        }
+
+        // Eviction: if we evicted a layer, then do not re-download it unless it was accessed more
+        // recently than it was evicted.
+        if let Some(evicted_at) = timeline_state.evicted_at.get(&layer.name) {
+            if &layer.access_time > evicted_at {
+                tracing::info!(
+                    "Re-downloading evicted layer {}, accessed at {}, evicted at {}",
+                    layer.name,
+                    strftime(&layer.access_time),
+                    strftime(evicted_at)
+                );
+            } else {
+                tracing::trace!(
+                    "Not re-downloading evicted layer {}, accessed at {}, evicted at {}",
+                    layer.name,
+                    strftime(&layer.access_time),
+                    strftime(evicted_at)
+                );
+                return LayerAction::Skip;
+            }
+        }
+        LayerAction::Download
+    }
+
     async fn download_timeline(
         &self,
         timeline: HeatMapTimeline,

pageserver/src/tenant/storage_layer.rs

@@ -10,21 +10,30 @@ mod layer_desc;
 mod layer_name;
 pub mod merge_iterator;
 
+use crate::config::PageServerConf;
 use crate::context::{AccessStatsBehavior, RequestContext};
 use bytes::Bytes;
+use futures::stream::FuturesUnordered;
+use futures::StreamExt;
 use pageserver_api::key::Key;
 use pageserver_api::keyspace::{KeySpace, KeySpaceRandomAccum};
 use pageserver_api::record::NeonWalRecord;
 use pageserver_api::value::Value;
-use std::cmp::{Ordering, Reverse};
+use std::cmp::Ordering;
 use std::collections::hash_map::Entry;
 use std::collections::{BinaryHeap, HashMap};
+use std::future::Future;
 use std::ops::Range;
+use std::pin::Pin;
+use std::sync::atomic::AtomicUsize;
 use std::sync::Arc;
 use std::time::{Duration, SystemTime, UNIX_EPOCH};
+use tracing::{trace, Instrument};
+use utils::sync::gate::GateGuard;
 
 use utils::lsn::Lsn;
 
+pub use batch_split_writer::{BatchLayerWriter, SplitDeltaLayerWriter, SplitImageLayerWriter};
 pub use delta_layer::{DeltaLayer, DeltaLayerWriter, ValueRef};
 pub use image_layer::{ImageLayer, ImageLayerWriter};
 pub use inmemory_layer::InMemoryLayer;
@@ -78,30 +87,151 @@ pub(crate) enum ValueReconstructSituation {
     Continue,
 }
 
-/// Reconstruct data accumulated for a single key during a vectored get
-#[derive(Debug, Default, Clone)]
-pub(crate) struct VectoredValueReconstructState {
-    pub(crate) records: Vec<(Lsn, NeonWalRecord)>,
-    pub(crate) img: Option<(Lsn, Bytes)>,
-
-    situation: ValueReconstructSituation,
+/// On disk representation of a value loaded in a buffer
+#[derive(Debug)]
+pub(crate) enum OnDiskValue {
+    /// Unencoded [`Value::Image`]
+    RawImage(Bytes),
+    /// Encoded [`Value`]. Can deserialize into an image or a WAL record
+    WalRecordOrImage(Bytes),
 }
 
-impl VectoredValueReconstructState {
-    fn get_cached_lsn(&self) -> Option<Lsn> {
-        self.img.as_ref().map(|img| img.0)
+/// Reconstruct data accumulated for a single key during a vectored get
+#[derive(Debug, Default)]
+pub(crate) struct VectoredValueReconstructState {
+    pub(crate) on_disk_values: Vec<(Lsn, OnDiskValueIoWaiter)>,
+
+    pub(crate) situation: ValueReconstructSituation,
+}
+
+#[derive(Debug)]
+pub(crate) struct OnDiskValueIoWaiter {
+    rx: tokio::sync::oneshot::Receiver<OnDiskValueIoResult>,
+}
+
+#[derive(Debug)]
+#[must_use]
+pub(crate) enum OnDiskValueIo {
+    /// Traversal identified this IO as required to complete the vectored get.
+    Required {
+        num_active_ios: Arc<AtomicUsize>,
+        tx: tokio::sync::oneshot::Sender<OnDiskValueIoResult>,
+    },
+    /// Sparse keyspace reads always read all the values for a given key,
+    /// even though only the first value is needed.
+    ///
+    /// This variant represents the unnecessary IOs for those values at lower LSNs
+    /// that aren't needed, but are currently still being done.
+    ///
+    /// The execution of unnecessary IOs was a pre-existing behavior before concurrent IO.
+    /// We added this explicit representation here so that we can drop
+    /// unnecessary IO results immediately, instead of buffering them in
+    /// `oneshot` channels inside [`VectoredValueReconstructState`] until
+    /// [`VectoredValueReconstructState::collect_pending_ios`] gets called.
+    Unnecessary,
+}
+
+type OnDiskValueIoResult = Result<OnDiskValue, std::io::Error>;
+
+impl OnDiskValueIo {
+    pub(crate) fn complete(self, res: OnDiskValueIoResult) {
+        match self {
+            OnDiskValueIo::Required { num_active_ios, tx } => {
+                num_active_ios.fetch_sub(1, std::sync::atomic::Ordering::Release);
+                let _ = tx.send(res);
+            }
+            OnDiskValueIo::Unnecessary => {
+                // Nobody cared, see variant doc comment.
+            }
+        }
     }
 }
 
-impl From<VectoredValueReconstructState> for ValueReconstructState {
-    fn from(mut state: VectoredValueReconstructState) -> Self {
-        // walredo expects the records to be descending in terms of Lsn
-        state.records.sort_by_key(|(lsn, _)| Reverse(*lsn));
+#[derive(Debug, thiserror::Error)]
+pub(crate) enum WaitCompletionError {
+    #[error("OnDiskValueIo was dropped without completing, likely the sidecar task panicked")]
+    IoDropped,
+}
 
-        ValueReconstructState {
-            records: state.records,
-            img: state.img,
+impl OnDiskValueIoWaiter {
+    pub(crate) async fn wait_completion(self) -> Result<OnDiskValueIoResult, WaitCompletionError> {
+        // NB: for Unnecessary IOs, this method never gets called because we don't add them to `on_disk_values`.
+        self.rx.await.map_err(|_| WaitCompletionError::IoDropped)
+    }
+}
+
+impl VectoredValueReconstructState {
+    /// # Cancel-Safety
+    ///
+    /// Technically fine to stop polling this future, but, the IOs will still
+    /// be executed to completion by the sidecar task and hold on to / consume resources.
+    /// Better not do it to make reasonsing about the system easier.
+    pub(crate) async fn collect_pending_ios(
+        self,
+    ) -> Result<ValueReconstructState, PageReconstructError> {
+        use utils::bin_ser::BeSer;
+
+        let mut res = Ok(ValueReconstructState::default());
+
+        // We should try hard not to bail early, so that by the time we return from this
+        // function, all IO for this value is done. It's not required -- we could totally
+        // stop polling the IO futures in the sidecar task, they need to support that,
+        // but just stopping to poll doesn't reduce the IO load on the disk. It's easier
+        // to reason about the system if we just wait for all IO to complete, even if
+        // we're no longer interested in the result.
+        //
+        // Revisit this when IO futures are replaced with a more sophisticated IO system
+        // and an IO scheduler, where we know which IOs were submitted and which ones
+        // just queued. Cf the comment on IoConcurrency::spawn_io.
+        for (lsn, waiter) in self.on_disk_values {
+            let value_recv_res = waiter
+                .wait_completion()
+                // we rely on the caller to poll us to completion, so this is not a bail point
+                .await;
+            // Force not bailing early by wrapping the code into a closure.
+            #[allow(clippy::redundant_closure_call)]
+            let _: () = (|| {
+                match (&mut res, value_recv_res) {
+                    (Err(_), _) => {
+                        // We've already failed, no need to process more.
+                    }
+                    (Ok(_), Err(wait_err)) => {
+                        // This shouldn't happen - likely the sidecar task panicked.
+                        res = Err(PageReconstructError::Other(wait_err.into()));
+                    }
+                    (Ok(_), Ok(Err(err))) => {
+                        let err: std::io::Error = err;
+                        // TODO: returning IO error here will fail a compute query.
+                        // Probably not what we want, we're not doing `maybe_fatal_err`
+                        // in the IO futures.
+                        // But it's been like that for a long time, not changing it
+                        // as part of concurrent IO.
+                        // => https://github.com/neondatabase/neon/issues/10454
+                        res = Err(PageReconstructError::Other(err.into()));
+                    }
+                    (Ok(ok), Ok(Ok(OnDiskValue::RawImage(img)))) => {
+                        assert!(ok.img.is_none());
+                        ok.img = Some((lsn, img));
+                    }
+                    (Ok(ok), Ok(Ok(OnDiskValue::WalRecordOrImage(buf)))) => {
+                        match Value::des(&buf) {
+                            Ok(Value::WalRecord(rec)) => {
+                                ok.records.push((lsn, rec));
+                            }
+                            Ok(Value::Image(img)) => {
+                                assert!(ok.img.is_none());
+                                ok.img = Some((lsn, img));
+                            }
+                            Err(err) => {
+                                res = Err(PageReconstructError::Other(err.into()));
+                            }
+                        }
+                    }
+                }
+            })();
         }
+
+        res
     }
 }
 
@@ -109,7 +239,7 @@ impl From<VectoredValueReconstructState> for ValueReconstructState {
 pub(crate) struct ValuesReconstructState {
     /// The keys will be removed after `get_vectored` completes. The caller outside `Timeline`
     /// should not expect to get anything from this hashmap.
-    pub(crate) keys: HashMap<Key, Result<VectoredValueReconstructState, PageReconstructError>>,
+    pub(crate) keys: HashMap<Key, VectoredValueReconstructState>,
     /// The keys which are already retrieved
     keys_done: KeySpaceRandomAccum,
 
@@ -119,27 +249,365 @@ pub(crate) struct ValuesReconstructState {
     // Statistics that are still accessible as a caller of `get_vectored_impl`.
     layers_visited: u32,
     delta_layers_visited: u32,
+
+    pub(crate) io_concurrency: IoConcurrency,
+    num_active_ios: Arc<AtomicUsize>,
+}
+
+/// The level of IO concurrency to be used on the read path
+///
+/// The desired end state is that we always do parallel IO.
+/// This struct and the dispatching in the impl will be removed once
+/// we've built enough confidence.
+pub(crate) enum IoConcurrency {
+    Sequential,
+    SidecarTask {
+        task_id: usize,
+        ios_tx: tokio::sync::mpsc::UnboundedSender<IoFuture>,
+    },
+}
+
+type IoFuture = Pin<Box<dyn Send + Future<Output = ()>>>;
+
+pub(crate) enum SelectedIoConcurrency {
+    Sequential,
+    SidecarTask(GateGuard),
+}
+
+impl std::fmt::Debug for IoConcurrency {
+    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
+        match self {
+            IoConcurrency::Sequential => write!(f, "Sequential"),
+            IoConcurrency::SidecarTask { .. } => write!(f, "SidecarTask"),
+        }
+    }
+}
+
+impl std::fmt::Debug for SelectedIoConcurrency {
+    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
+        match self {
+            SelectedIoConcurrency::Sequential => write!(f, "Sequential"),
+            SelectedIoConcurrency::SidecarTask(_) => write!(f, "SidecarTask"),
+        }
+    }
+}
+
+impl IoConcurrency {
+    /// Force sequential IO. This is a temporary workaround until we have
+    /// moved plumbing-through-the-call-stack
+    /// of IoConcurrency into `RequestContextq.
+    ///
+    /// DO NOT USE for new code.
+    ///
+    /// Tracking issue: <https://github.com/neondatabase/neon/issues/10460>.
+    pub(crate) fn sequential() -> Self {
+        Self::spawn(SelectedIoConcurrency::Sequential)
+    }
+
+    pub(crate) fn spawn_from_conf(
+        conf: &'static PageServerConf,
+        gate_guard: GateGuard,
+    ) -> IoConcurrency {
+        use pageserver_api::config::GetVectoredConcurrentIo;
+        let selected = match conf.get_vectored_concurrent_io {
+            GetVectoredConcurrentIo::Sequential => SelectedIoConcurrency::Sequential,
+            GetVectoredConcurrentIo::SidecarTask => SelectedIoConcurrency::SidecarTask(gate_guard),
+        };
+        Self::spawn(selected)
+    }
+
+    pub(crate) fn spawn(io_concurrency: SelectedIoConcurrency) -> Self {
+        match io_concurrency {
+            SelectedIoConcurrency::Sequential => IoConcurrency::Sequential,
+            SelectedIoConcurrency::SidecarTask(gate_guard) => {
+                let (ios_tx, ios_rx) = tokio::sync::mpsc::unbounded_channel();
+                static TASK_ID: AtomicUsize = AtomicUsize::new(0);
+                let task_id = TASK_ID.fetch_add(1, std::sync::atomic::Ordering::Relaxed);
+                // TODO: enrich the span with more context (tenant,shard,timeline) + (basebackup|pagestream|...)
+                let span =
+                    tracing::info_span!(parent: None, "IoConcurrency_sidecar", task_id = task_id);
+                trace!(task_id, "spawning sidecar task");
+                tokio::spawn(async move {
+                    trace!("start");
+                    scopeguard::defer!{ trace!("end") };
+                    type IosRx = tokio::sync::mpsc::UnboundedReceiver<IoFuture>;
+                    enum State {
+                        Waiting {
+                            // invariant: is_empty(), but we recycle the allocation
+                            empty_futures: FuturesUnordered<IoFuture>,
+                            ios_rx: IosRx,
+                        },
+                        Executing {
+                            futures: FuturesUnordered<IoFuture>,
+                            ios_rx: IosRx,
+                        },
+                        ShuttingDown {
+                            futures: FuturesUnordered<IoFuture>,
+                        },
+                    }
+                    let mut state = State::Waiting {
+                        empty_futures: FuturesUnordered::new(),
+                        ios_rx,
+                    };
+                    loop {
+                        match state {
+                            State::Waiting {
+                                empty_futures,
+                                mut ios_rx,
+                            } => {
+                                assert!(empty_futures.is_empty());
+                                tokio::select! {
+                                    fut = ios_rx.recv() => {
+                                        if let Some(fut) = fut {
+                                            trace!("received new io future");
+                                            empty_futures.push(fut);
+                                            state = State::Executing { futures: empty_futures, ios_rx };
+                                        } else {
+                                            state = State::ShuttingDown { futures: empty_futures }
+                                        }
+                                    }
+                                }
+                            }
+                            State::Executing {
+                                mut futures,
+                                mut ios_rx,
+                            } => {
+                                tokio::select! {
+                                    res = futures.next() => {
+                                        trace!("io future completed");
+                                        assert!(res.is_some());
+                                        if futures.is_empty() {
+                                            state = State::Waiting { empty_futures: futures, ios_rx};
+                                        } else {
+                                            state = State::Executing { futures, ios_rx };
+                                        }
+                                    }
+                                    fut = ios_rx.recv() => {
+                                        if let Some(fut) = fut {
+                                            trace!("received new io future");
+                                            futures.push(fut);
+                                            state =  State::Executing { futures, ios_rx};
+                                        } else {
+                                            state = State::ShuttingDown { futures };
+                                        }
+                                    }
+                                }
+                            }
+                            State::ShuttingDown {
+                                mut futures,
+                            } => {
+                                trace!("shutting down");
+                                while let Some(()) = futures.next().await {
+                                    trace!("io future completed (shutdown)");
+                                    // drain
+                                }
+                                trace!("shutdown complete");
+                                break;
+                            }
+                        }
+                    }
+                    drop(gate_guard); // drop it right before we exit
+                }.instrument(span));
+                IoConcurrency::SidecarTask { task_id, ios_tx }
+            }
+        }
+    }
+
+    pub(crate) fn clone(&self) -> Self {
+        match self {
+            IoConcurrency::Sequential => IoConcurrency::Sequential,
+            IoConcurrency::SidecarTask { task_id, ios_tx } => IoConcurrency::SidecarTask {
+                task_id: *task_id,
+                ios_tx: ios_tx.clone(),
+            },
+        }
+    }
+
+    /// Submit an IO to be executed in the background. DEADLOCK RISK, read the full doc string.
+    ///
+    /// The IO is represented as an opaque future.
+    /// IO completion must be handled inside the future, e.g., through a oneshot channel.
+    ///
+    /// The API seems simple but there are multiple **pitfalls** involving
+    /// DEADLOCK RISK.
+    ///
+    /// First, there are no guarantees about the exexecution of the IO.
+    /// It may be `await`ed in-place before this function returns.
+    /// It may be polled partially by this task and handed off to another task to be finished.
+    /// It may be polled and then dropped before returning ready.
+    ///
+    /// This means that submitted IOs must not be interedependent.
+    /// Interdependence may be through shared limited resources, e.g.,
+    /// - VirtualFile file descriptor cache slot acquisition
+    /// - tokio-epoll-uring slot
+    ///
+    /// # Why current usage is safe from deadlocks
+    ///
+    /// Textbook condition for a deadlock is that _all_ of the following be given
+    /// - Mutual exclusion
+    /// - Hold and wait
+    /// - No preemption
+    /// - Circular wait
+    ///
+    /// The current usage is safe because:
+    /// - Mutual exclusion: IO futures definitely use mutexes, no way around that for now
+    /// - Hold and wait: IO futures currently hold two kinds of locks/resources while waiting
+    ///   for acquisition of other resources:
+    ///    - VirtualFile file descriptor cache slot tokio mutex
+    ///    - tokio-epoll-uring slot (uses tokio notify => wait queue, much like mutex)
+    /// - No preemption: there's no taking-away of acquired locks/resources => given
+    /// - Circular wait: this is the part of the condition that isn't met: all IO futures
+    ///   first acquire VirtualFile mutex, then tokio-epoll-uring slot.
+    ///   There is no IO future that acquires slot before VirtualFile.
+    ///   Hence there can be no circular waiting.
+    ///   Hence there cannot be a deadlock.
+    ///
+    /// This is a very fragile situation and must be revisited whenver any code called from
+    /// inside the IO futures is changed.
+    ///
+    /// We will move away from opaque IO futures towards well-defined IOs at some point in
+    /// the future when we have shipped this first version of concurrent IO to production
+    /// and are ready to retire the Sequential mode which runs the futures in place.
+    /// Right now, while brittle, the opaque IO approach allows us to ship the feature
+    /// with minimal changes to the code and minimal changes to existing behavior in Sequential mode.
+    ///
+    /// Also read the comment in `collect_pending_ios`.
+    pub(crate) async fn spawn_io<F>(&mut self, fut: F)
+    where
+        F: std::future::Future<Output = ()> + Send + 'static,
+    {
+        match self {
+            IoConcurrency::Sequential => fut.await,
+            IoConcurrency::SidecarTask { ios_tx, .. } => {
+                let fut = Box::pin(fut);
+                // NB: experiments showed that doing an opportunistic poll of `fut` here was bad for throughput
+                // while insignificant for latency.
+                // It would make sense to revisit the tokio-epoll-uring API in the future such that we can try
+                // a submission here, but never poll the future. That way, io_uring can make proccess while
+                // the future sits in the ios_tx queue.
+                match ios_tx.send(fut) {
+                    Ok(()) => {}
+                    Err(_) => {
+                        unreachable!("the io task must have exited, likely it panicked")
+                    }
+                }
+            }
+        }
+    }
+
+    #[cfg(test)]
+    pub(crate) fn spawn_for_test() -> impl std::ops::DerefMut<Target = Self> {
+        use std::ops::{Deref, DerefMut};
+        use tracing::info;
+        use utils::sync::gate::Gate;
+
+        // Spawn needs a Gate, give it one.
+        struct Wrapper {
+            inner: IoConcurrency,
+            #[allow(dead_code)]
+            gate: Box<Gate>,
+        }
+        impl Deref for Wrapper {
+            type Target = IoConcurrency;
+
+            fn deref(&self) -> &Self::Target {
+                &self.inner
+            }
+        }
+        impl DerefMut for Wrapper {
+            fn deref_mut(&mut self) -> &mut Self::Target {
+                &mut self.inner
+            }
+        }
+        let gate = Box::new(Gate::default());
+
+        // The default behavior when running Rust unit tests without any further
+        // flags is to use the new behavior.
+        // The CI uses the following environment variable to unit test both old
+        // and new behavior.
+        // NB: the Python regression & perf tests take the `else` branch
+        // below and have their own defaults management.
+        let selected = {
+            // The pageserver_api::config type is unsuitable because it's internally tagged.
+            #[derive(serde::Deserialize)]
+            #[serde(rename_all = "kebab-case")]
+            enum TestOverride {
+                Sequential,
+                SidecarTask,
+            }
+            use once_cell::sync::Lazy;
+            static TEST_OVERRIDE: Lazy<TestOverride> = Lazy::new(|| {
+                utils::env::var_serde_json_string(
+                    "NEON_PAGESERVER_UNIT_TEST_GET_VECTORED_CONCURRENT_IO",
+                )
+                .unwrap_or(TestOverride::SidecarTask)
+            });
+
+            match *TEST_OVERRIDE {
+                TestOverride::Sequential => SelectedIoConcurrency::Sequential,
+                TestOverride::SidecarTask => {
+                    SelectedIoConcurrency::SidecarTask(gate.enter().expect("just created it"))
+                }
+            }
+        };
+
+        info!(?selected, "get_vectored_concurrent_io test");
+
+        Wrapper {
+            inner: Self::spawn(selected),
+            gate,
+        }
+    }
+}
+
+/// Make noise in case the [`ValuesReconstructState`] gets dropped while
+/// there are still IOs in flight.
+/// Refer to `collect_pending_ios` for why we prefer not to do that.
+//
+/// We log from here instead of from the sidecar task because the [`ValuesReconstructState`]
+/// gets dropped in a tracing span with more context.
+/// We repeat the sidecar tasks's `task_id` so we can correlate what we emit here with
+/// the logs / panic handler logs from the sidecar task, which also logs the `task_id`.
+impl Drop for ValuesReconstructState {
+    fn drop(&mut self) {
+        let num_active_ios = self
+            .num_active_ios
+            .load(std::sync::atomic::Ordering::Acquire);
+        if num_active_ios == 0 {
+            return;
+        }
+        let sidecar_task_id = match &self.io_concurrency {
+            IoConcurrency::Sequential => None,
+            IoConcurrency::SidecarTask { task_id, .. } => Some(*task_id),
+        };
+        tracing::warn!(
+            num_active_ios,
+            ?sidecar_task_id,
+            backtrace=%std::backtrace::Backtrace::force_capture(),
+            "dropping ValuesReconstructState while some IOs have not been completed",
+        );
+    }
 }
 
 impl ValuesReconstructState {
-    pub(crate) fn new() -> Self {
+    pub(crate) fn new(io_concurrency: IoConcurrency) -> Self {
         Self {
             keys: HashMap::new(),
             keys_done: KeySpaceRandomAccum::new(),
             keys_with_image_coverage: None,
             layers_visited: 0,
             delta_layers_visited: 0,
+            io_concurrency,
+            num_active_ios: Arc::new(AtomicUsize::new(0)),
         }
     }
 
-    /// Associate a key with the error which it encountered and mark it as done
-    pub(crate) fn on_key_error(&mut self, key: Key, err: PageReconstructError) {
-        let previous = self.keys.insert(key, Err(err));
-        if let Some(Ok(state)) = previous {
-            if state.situation == ValueReconstructSituation::Continue {
-                self.keys_done.add_key(key);
-            }
-        }
+    /// Absolutely read [`IoConcurrency::spawn_io`] to learn about assumptions & pitfalls.
+    pub(crate) async fn spawn_io<F>(&mut self, fut: F)
+    where
+        F: std::future::Future<Output = ()> + Send + 'static,
+    {
+        self.io_concurrency.spawn_io(fut).await;
     }
 
     pub(crate) fn on_layer_visited(&mut self, layer: &ReadableLayer) {
@@ -159,29 +627,6 @@ impl ValuesReconstructState {
         self.layers_visited
     }
 
-    /// This function is called after reading a keyspace from a layer.
-    /// It checks if the read path has now moved past the cached Lsn for any keys.
-    ///
-    /// Implementation note: We intentionally iterate over the keys for which we've
-    /// already collected some reconstruct data. This avoids scaling complexity with
-    /// the size of the search space.
-    pub(crate) fn on_lsn_advanced(&mut self, keyspace: &KeySpace, advanced_to: Lsn) {
-        for (key, value) in self.keys.iter_mut() {
-            if !keyspace.contains(key) {
-                continue;
-            }
-
-            if let Ok(state) = value {
-                if state.situation != ValueReconstructSituation::Complete
-                    && state.get_cached_lsn() >= Some(advanced_to)
-                {
-                    state.situation = ValueReconstructSituation::Complete;
-                    self.keys_done.add_key(*key);
-                }
-            }
-        }
-    }
-
     /// On hitting image layer, we can mark all keys in this range as done, because
     /// if the image layer does not contain a key, it is deleted/never added.
     pub(crate) fn on_image_layer_visited(&mut self, key_range: &Range<Key>) {
@@ -199,70 +644,42 @@ impl ValuesReconstructState {
     ///
     /// If the key is in the sparse keyspace (i.e., aux files), we do not track them in
     /// `key_done`.
-    pub(crate) fn update_key(
-        &mut self,
-        key: &Key,
-        lsn: Lsn,
-        value: Value,
-    ) -> ValueReconstructSituation {
-        let state = self
-            .keys
-            .entry(*key)
-            .or_insert(Ok(VectoredValueReconstructState::default()));
+    // TODO: rename this method & update description.
+    pub(crate) fn update_key(&mut self, key: &Key, lsn: Lsn, completes: bool) -> OnDiskValueIo {
+        let state = self.keys.entry(*key).or_default();
+
         let is_sparse_key = key.is_sparse();
-        if let Ok(state) = state {
-            let key_done = match state.situation {
-                ValueReconstructSituation::Complete => {
-                    if is_sparse_key {
-                        // Sparse keyspace might be visited multiple times because
-                        // we don't track unmapped keyspaces.
-                        return ValueReconstructSituation::Complete;
-                    } else {
-                        unreachable!()
-                    }
-                }
-                ValueReconstructSituation::Continue => match value {
-                    Value::Image(img) => {
-                        state.img = Some((lsn, img));
-                        true
-                    }
-                    Value::WalRecord(rec) => {
-                        debug_assert!(
-                            Some(lsn) > state.get_cached_lsn(),
-                            "Attempt to collect a record below cached LSN for walredo: {} < {}",
-                            lsn,
-                            state
-                                .get_cached_lsn()
-                                .expect("Assertion can only fire if a cached lsn is present")
-                        );
 
-                        let will_init = rec.will_init();
-                        state.records.push((lsn, rec));
-                        will_init
-                    }
-                },
-            };
-
-            if key_done && state.situation == ValueReconstructSituation::Continue {
-                state.situation = ValueReconstructSituation::Complete;
-                if !is_sparse_key {
-                    self.keys_done.add_key(*key);
+        let required_io = match state.situation {
+            ValueReconstructSituation::Complete => {
+                if is_sparse_key {
+                    // Sparse keyspace might be visited multiple times because
+                    // we don't track unmapped keyspaces.
+                    return OnDiskValueIo::Unnecessary;
+                } else {
+                    unreachable!()
                 }
             }
+            ValueReconstructSituation::Continue => {
+                self.num_active_ios
+                    .fetch_add(1, std::sync::atomic::Ordering::Release);
+                let (tx, rx) = tokio::sync::oneshot::channel();
+                state.on_disk_values.push((lsn, OnDiskValueIoWaiter { rx }));
+                OnDiskValueIo::Required {
+                    tx,
+                    num_active_ios: Arc::clone(&self.num_active_ios),
+                }
+            }
+        };
 
-            state.situation
-        } else {
-            ValueReconstructSituation::Complete
+        if completes && state.situation == ValueReconstructSituation::Continue {
+            state.situation = ValueReconstructSituation::Complete;
+            if !is_sparse_key {
+                self.keys_done.add_key(*key);
+            }
         }
-    }
 
-    /// Returns the Lsn at which this key is cached if one exists.
-    /// The read path should go no further than this Lsn for the given key.
-    pub(crate) fn get_cached_lsn(&self, key: &Key) -> Option<Lsn> {
-        self.keys
-            .get(key)
-            .and_then(|k| k.as_ref().ok())
-            .and_then(|state| state.get_cached_lsn())
+        required_io
     }
 
     /// Returns the key space describing the keys that have
@@ -276,12 +693,6 @@ impl ValuesReconstructState {
     }
 }
 
-impl Default for ValuesReconstructState {
-    fn default() -> Self {
-        Self::new()
-    }
-}
-
 /// A key that uniquely identifies a layer in a timeline
 #[derive(Debug, PartialEq, Eq, Clone, Hash)]
 pub(crate) enum LayerId {
@@ -720,3 +1131,78 @@ impl<T: std::fmt::Display> std::fmt::Debug for RangeDisplayDebug<'_, T> {
         write!(f, "{}..{}", self.0.start, self.0.end)
     }
 }
+
+#[cfg(test)]
+mod tests2 {
+    use pageserver_api::key::DBDIR_KEY;
+    use tracing::info;
+
+    use super::*;
+    use crate::tenant::storage_layer::IoConcurrency;
+
+    /// TODO: currently this test relies on manual visual inspection of the --no-capture output.
+    /// Should look like so:
+    /// ```text
+    /// RUST_LOG=trace cargo nextest run  --features testing  --no-capture test_io_concurrency_noise
+    /// running 1 test
+    /// 2025-01-21T17:42:01.335679Z  INFO get_vectored_concurrent_io test selected=SidecarTask
+    /// 2025-01-21T17:42:01.335680Z TRACE spawning sidecar task task_id=0
+    /// 2025-01-21T17:42:01.335937Z TRACE IoConcurrency_sidecar{task_id=0}: start
+    /// 2025-01-21T17:42:01.335972Z TRACE IoConcurrency_sidecar{task_id=0}: received new io future
+    /// 2025-01-21T17:42:01.335999Z  INFO IoConcurrency_sidecar{task_id=0}: waiting for signal to complete IO
+    /// 2025-01-21T17:42:01.336229Z  WARN dropping ValuesReconstructState while some IOs have not been completed num_active_ios=1 sidecar_task_id=Some(0) backtrace=   0: <pageserver::tenant::storage_layer::ValuesReconstructState as core::ops::drop::Drop>::drop
+    ///              at ./src/tenant/storage_layer.rs:553:24
+    ///    1: core::ptr::drop_in_place<pageserver::tenant::storage_layer::ValuesReconstructState>
+    ///              at /home/christian/.rustup/toolchains/1.84.0-x86_64-unknown-linux-gnu/lib/rustlib/src/rust/library/core/src/ptr/mod.rs:521:1
+    ///    2: core::mem::drop
+    ///              at /home/christian/.rustup/toolchains/1.84.0-x86_64-unknown-linux-gnu/lib/rustlib/src/rust/library/core/src/mem/mod.rs:942:24
+    ///    3: pageserver::tenant::storage_layer::tests2::test_io_concurrency_noise::{{closure}}
+    ///              at ./src/tenant/storage_layer.rs:1159:9
+    ///   ...
+    ///   49: <unknown>
+    /// 2025-01-21T17:42:01.452293Z  INFO IoConcurrency_sidecar{task_id=0}: completing IO
+    /// 2025-01-21T17:42:01.452357Z TRACE IoConcurrency_sidecar{task_id=0}: io future completed
+    /// 2025-01-21T17:42:01.452473Z TRACE IoConcurrency_sidecar{task_id=0}: end
+    /// test tenant::storage_layer::tests2::test_io_concurrency_noise ... ok
+    ///
+    /// ```
+    #[tokio::test]
+    async fn test_io_concurrency_noise() {
+        crate::tenant::harness::setup_logging();
+
+        let io_concurrency = IoConcurrency::spawn_for_test();
+        match *io_concurrency {
+            IoConcurrency::Sequential => {
+                // This test asserts behavior in sidecar mode, doesn't make sense in sequential mode.
+                return;
+            }
+            IoConcurrency::SidecarTask { .. } => {}
+        }
+        let mut reconstruct_state = ValuesReconstructState::new(io_concurrency.clone());
+
+        let (io_fut_is_waiting_tx, io_fut_is_waiting) = tokio::sync::oneshot::channel();
+        let (do_complete_io, should_complete_io) = tokio::sync::oneshot::channel();
+        let (io_fut_exiting_tx, io_fut_exiting) = tokio::sync::oneshot::channel();
+
+        let io = reconstruct_state.update_key(&DBDIR_KEY, Lsn(8), true);
+        reconstruct_state
+            .spawn_io(async move {
+                info!("waiting for signal to complete IO");
+                io_fut_is_waiting_tx.send(()).unwrap();
+                should_complete_io.await.unwrap();
+                info!("completing IO");
+                io.complete(Ok(OnDiskValue::RawImage(Bytes::new())));
+                io_fut_exiting_tx.send(()).unwrap();
+            })
+            .await;
+
+        io_fut_is_waiting.await.unwrap();
+
+        // this is what makes the noise
+        drop(reconstruct_state);
+
+        do_complete_io.send(()).unwrap();
+
+        io_fut_exiting.await.unwrap();
+    }
+}

pageserver/src/tenant/storage_layer/batch_split_writer.rs

@@ -87,6 +87,23 @@ impl BatchLayerWriter {
         ));
     }
 
+    pub(crate) async fn finish(
+        self,
+        tline: &Arc<Timeline>,
+        ctx: &RequestContext,
+    ) -> anyhow::Result<Vec<ResidentLayer>> {
+        let res = self
+            .finish_with_discard_fn(tline, ctx, |_| async { false })
+            .await?;
+        let mut output = Vec::new();
+        for r in res {
+            if let BatchWriterResult::Produced(layer) = r {
+                output.push(layer);
+            }
+        }
+        Ok(output)
+    }
+
     pub(crate) async fn finish_with_discard_fn<D, F>(
         self,
         tline: &Arc<Timeline>,

pageserver/src/tenant/storage_layer/delta_layer.rs

@@ -41,13 +41,12 @@ use crate::tenant::vectored_blob_io::{
     BlobFlag, BufView, StreamingVectoredReadPlanner, VectoredBlobReader, VectoredRead,
     VectoredReadPlanner,
 };
-use crate::tenant::PageReconstructError;
 use crate::virtual_file::owned_buffers_io::io_buf_ext::{FullSlice, IoBufExt};
 use crate::virtual_file::IoBufferMut;
 use crate::virtual_file::{self, MaybeFatalIo, VirtualFile};
 use crate::TEMP_FILE_SUFFIX;
 use crate::{DELTA_FILE_MAGIC, STORAGE_FORMAT_VERSION};
-use anyhow::{anyhow, bail, ensure, Context, Result};
+use anyhow::{bail, ensure, Context, Result};
 use camino::{Utf8Path, Utf8PathBuf};
 use futures::StreamExt;
 use itertools::Itertools;
@@ -60,7 +59,7 @@ use pageserver_api::shard::TenantShardId;
 use pageserver_api::value::Value;
 use rand::{distributions::Alphanumeric, Rng};
 use serde::{Deserialize, Serialize};
-use std::collections::VecDeque;
+use std::collections::{HashMap, VecDeque};
 use std::fs::File;
 use std::io::SeekFrom;
 use std::ops::Range;
@@ -77,7 +76,10 @@ use utils::{
     lsn::Lsn,
 };
 
-use super::{AsLayerDesc, LayerName, PersistentLayerDesc, ValuesReconstructState};
+use super::{
+    AsLayerDesc, LayerName, OnDiskValue, OnDiskValueIo, PersistentLayerDesc, ResidentLayer,
+    ValuesReconstructState,
+};
 
 ///
 /// Header stored in the beginning of the file
@@ -226,7 +228,7 @@ pub struct DeltaLayerInner {
     index_start_blk: u32,
     index_root_blk: u32,
 
-    file: VirtualFile,
+    file: Arc<VirtualFile>,
     file_id: FileId,
 
     layer_key_range: Range<Key>,
@@ -795,9 +797,11 @@ impl DeltaLayerInner {
         max_vectored_read_bytes: Option<MaxVectoredReadBytes>,
         ctx: &RequestContext,
     ) -> anyhow::Result<Self> {
-        let file = VirtualFile::open_v2(path, ctx)
-            .await
-            .context("open layer file")?;
+        let file = Arc::new(
+            VirtualFile::open_v2(path, ctx)
+                .await
+                .context("open layer file")?,
+        );
 
         let file_id = page_cache::next_file_id();
 
@@ -842,12 +846,11 @@ impl DeltaLayerInner {
     // Look up the keys in the provided keyspace and update
     // the reconstruct state with whatever is found.
     //
-    // If the key is cached, go no further than the cached Lsn.
-    //
     // Currently, the index is visited for each range, but this
     // can be further optimised to visit the index only once.
     pub(super) async fn get_values_reconstruct_data(
         &self,
+        this: ResidentLayer,
         keyspace: KeySpace,
         lsn_range: Range<Lsn>,
         reconstruct_state: &mut ValuesReconstructState,
@@ -875,17 +878,14 @@ impl DeltaLayerInner {
             data_end_offset,
             index_reader,
             planner,
-            reconstruct_state,
             ctx,
         )
         .await
         .map_err(GetVectoredError::Other)?;
 
-        self.do_reads_and_update_state(reads, reconstruct_state, ctx)
+        self.do_reads_and_update_state(this, reads, reconstruct_state, ctx)
             .await;
 
-        reconstruct_state.on_lsn_advanced(&keyspace, lsn_range.start);
-
         Ok(())
     }
 
@@ -895,7 +895,6 @@ impl DeltaLayerInner {
         data_end_offset: u64,
         index_reader: DiskBtreeReader<Reader, DELTA_KEY_SIZE>,
         mut planner: VectoredReadPlanner,
-        reconstruct_state: &mut ValuesReconstructState,
         ctx: &RequestContext,
     ) -> anyhow::Result<Vec<VectoredRead>>
     where
@@ -922,10 +921,9 @@ impl DeltaLayerInner {
                 assert!(key >= range.start);
 
                 let outside_lsn_range = !lsn_range.contains(&lsn);
-                let below_cached_lsn = reconstruct_state.get_cached_lsn(&key) >= Some(lsn);
 
                 let flag = {
-                    if outside_lsn_range || below_cached_lsn {
+                    if outside_lsn_range {
                         BlobFlag::Ignore
                     } else if blob_ref.will_init() {
                         BlobFlag::ReplaceAll
@@ -994,98 +992,78 @@ impl DeltaLayerInner {
 
     async fn do_reads_and_update_state(
         &self,
+        this: ResidentLayer,
         reads: Vec<VectoredRead>,
         reconstruct_state: &mut ValuesReconstructState,
         ctx: &RequestContext,
     ) {
-        let vectored_blob_reader = VectoredBlobReader::new(&self.file);
-        let mut ignore_key_with_err = None;
-
         let max_vectored_read_bytes = self
             .max_vectored_read_bytes
             .expect("Layer is loaded with max vectored bytes config")
             .0
             .into();
         let buf_size = Self::get_min_read_buffer_size(&reads, max_vectored_read_bytes);
-        let mut buf = Some(IoBufferMut::with_capacity(buf_size));
 
         // Note that reads are processed in reverse order (from highest key+lsn).
         // This is the order that `ReconstructState` requires such that it can
         // track when a key is done.
         for read in reads.into_iter().rev() {
-            let res = vectored_blob_reader
-                .read_blobs(&read, buf.take().expect("Should have a buffer"), ctx)
-                .await;
-
-            let blobs_buf = match res {
-                Ok(blobs_buf) => blobs_buf,
-                Err(err) => {
-                    let kind = err.kind();
-                    for (_, blob_meta) in read.blobs_at.as_slice() {
-                        reconstruct_state.on_key_error(
-                            blob_meta.key,
-                            PageReconstructError::Other(anyhow!(
-                                "Failed to read blobs from virtual file {}: {}",
-                                self.file.path(),
-                                kind
-                            )),
-                        );
-                    }
-
-                    // We have "lost" the buffer since the lower level IO api
-                    // doesn't return the buffer on error. Allocate a new one.
-                    buf = Some(IoBufferMut::with_capacity(buf_size));
-
-                    continue;
-                }
-            };
-            let view = BufView::new_slice(&blobs_buf.buf);
-            for meta in blobs_buf.blobs.iter().rev() {
-                if Some(meta.meta.key) == ignore_key_with_err {
-                    continue;
-                }
-                let blob_read = meta.read(&view).await;
-                let blob_read = match blob_read {
-                    Ok(buf) => buf,
-                    Err(e) => {
-                        reconstruct_state.on_key_error(
-                            meta.meta.key,
-                            PageReconstructError::Other(anyhow!(e).context(format!(
-                                "Failed to decompress blob from virtual file {}",
-                                self.file.path(),
-                            ))),
-                        );
-
-                        ignore_key_with_err = Some(meta.meta.key);
-                        continue;
-                    }
-                };
-
-                let value = Value::des(&blob_read);
-
-                let value = match value {
-                    Ok(v) => v,
-                    Err(e) => {
-                        reconstruct_state.on_key_error(
-                            meta.meta.key,
-                            PageReconstructError::Other(anyhow!(e).context(format!(
-                                "Failed to deserialize blob from virtual file {}",
-                                self.file.path(),
-                            ))),
-                        );
-
-                        ignore_key_with_err = Some(meta.meta.key);
-                        continue;
-                    }
-                };
-
-                // Invariant: once a key reaches [`ValueReconstructSituation::Complete`]
-                // state, no further updates shall be made to it. The call below will
-                // panic if the invariant is violated.
-                reconstruct_state.update_key(&meta.meta.key, meta.meta.lsn, value);
+            let mut ios: HashMap<(Key, Lsn), OnDiskValueIo> = Default::default();
+            for (_, blob_meta) in read.blobs_at.as_slice().iter().rev() {
+                let io = reconstruct_state.update_key(
+                    &blob_meta.key,
+                    blob_meta.lsn,
+                    blob_meta.will_init,
+                );
+                ios.insert((blob_meta.key, blob_meta.lsn), io);
             }
 
-            buf = Some(blobs_buf.buf);
+            let read_extend_residency = this.clone();
+            let read_from = self.file.clone();
+            let read_ctx = ctx.attached_child();
+            reconstruct_state
+                .spawn_io(async move {
+                    let vectored_blob_reader = VectoredBlobReader::new(&read_from);
+                    let buf = IoBufferMut::with_capacity(buf_size);
+
+                    let res = vectored_blob_reader.read_blobs(&read, buf, &read_ctx).await;
+                    match res {
+                        Ok(blobs_buf) => {
+                            let view = BufView::new_slice(&blobs_buf.buf);
+                            for meta in blobs_buf.blobs.iter().rev() {
+                                let io = ios.remove(&(meta.meta.key, meta.meta.lsn)).unwrap();
+
+                                let blob_read = meta.read(&view).await;
+                                let blob_read = match blob_read {
+                                    Ok(buf) => buf,
+                                    Err(e) => {
+                                        io.complete(Err(e));
+                                        continue;
+                                    }
+                                };
+
+                                io.complete(Ok(OnDiskValue::WalRecordOrImage(
+                                    blob_read.into_bytes(),
+                                )));
+                            }
+
+                            assert!(ios.is_empty());
+                        }
+                        Err(err) => {
+                            for (_, sender) in ios {
+                                sender.complete(Err(std::io::Error::new(
+                                    err.kind(),
+                                    "vec read failed",
+                                )));
+                            }
+                        }
+                    }
+
+                    // keep layer resident until this IO is done; this spawned IO future generally outlives the
+                    // call to `self` / the `Arc<DownloadedLayer>` / the `ResidentLayer` that guarantees residency
+                    drop(read_extend_residency);
+                })
+                .await;
         }
     }
 
@@ -1224,7 +1202,14 @@ impl DeltaLayerInner {
             let actionable = if let Some((key, lsn, start_offset)) = prev.take() {
                 let end_offset = offset;
 
-                Some((BlobMeta { key, lsn }, start_offset..end_offset))
+                Some((
+                    BlobMeta {
+                        key,
+                        lsn,
+                        will_init: false,
+                    },
+                    start_offset..end_offset,
+                ))
             } else {
                 None
             };
@@ -1560,7 +1545,9 @@ impl DeltaLayerIterator<'_> {
                 let lsn = DeltaKey::extract_lsn_from_buf(&raw_key);
                 let blob_ref = BlobRef(value);
                 let offset = blob_ref.pos();
-                if let Some(batch_plan) = self.planner.handle(key, lsn, offset) {
+                if let Some(batch_plan) =
+                    self.planner.handle(key, lsn, offset, blob_ref.will_init())
+                {
                     break batch_plan;
                 }
             } else {
@@ -1673,7 +1660,6 @@ pub(crate) mod test {
             .expect("In memory disk finish should never fail");
         let reader = DiskBtreeReader::<_, DELTA_KEY_SIZE>::new(0, root_offset, disk);
         let planner = VectoredReadPlanner::new(100);
-        let mut reconstruct_state = ValuesReconstructState::new();
         let ctx = RequestContext::new(TaskKind::UnitTest, DownloadBehavior::Error);
 
         let keyspace = KeySpace {
@@ -1691,7 +1677,6 @@ pub(crate) mod test {
             disk_offset,
             reader,
             planner,
-            &mut reconstruct_state,
             &ctx,
         )
         .await
@@ -1935,7 +1920,6 @@ pub(crate) mod test {
             );
 
             let planner = VectoredReadPlanner::new(constants::MAX_VECTORED_READ_BYTES);
-            let mut reconstruct_state = ValuesReconstructState::new();
             let keyspace = pick_random_keyspace(rng, &entries_meta.key_range);
             let data_end_offset = inner.index_start_blk as u64 * PAGE_SZ as u64;
 
@@ -1945,7 +1929,6 @@ pub(crate) mod test {
                 data_end_offset,
                 index_reader,
                 planner,
-                &mut reconstruct_state,
                 &ctx,
             )
             .await?;

pageserver/src/tenant/storage_layer/image_layer.rs

@@ -38,12 +38,11 @@ use crate::tenant::vectored_blob_io::{
     BlobFlag, BufView, StreamingVectoredReadPlanner, VectoredBlobReader, VectoredRead,
     VectoredReadPlanner,
 };
-use crate::tenant::PageReconstructError;
 use crate::virtual_file::owned_buffers_io::io_buf_ext::IoBufExt;
 use crate::virtual_file::IoBufferMut;
 use crate::virtual_file::{self, MaybeFatalIo, VirtualFile};
 use crate::{IMAGE_FILE_MAGIC, STORAGE_FORMAT_VERSION, TEMP_FILE_SUFFIX};
-use anyhow::{anyhow, bail, ensure, Context, Result};
+use anyhow::{bail, ensure, Context, Result};
 use bytes::Bytes;
 use camino::{Utf8Path, Utf8PathBuf};
 use hex;
@@ -56,12 +55,13 @@ use pageserver_api::shard::{ShardIdentity, TenantShardId};
 use pageserver_api::value::Value;
 use rand::{distributions::Alphanumeric, Rng};
 use serde::{Deserialize, Serialize};
-use std::collections::VecDeque;
+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 tokio::sync::OnceCell;
 use tokio_stream::StreamExt;
 use tracing::*;
@@ -73,7 +73,10 @@ use utils::{
 };
 
 use super::layer_name::ImageLayerName;
-use super::{AsLayerDesc, LayerName, PersistentLayerDesc, ValuesReconstructState};
+use super::{
+    AsLayerDesc, LayerName, OnDiskValue, OnDiskValueIo, PersistentLayerDesc, ResidentLayer,
+    ValuesReconstructState,
+};
 
 ///
 /// Header stored in the beginning of the file
@@ -164,7 +167,7 @@ pub struct ImageLayerInner {
     key_range: Range<Key>,
     lsn: Lsn,
 
-    file: VirtualFile,
+    file: Arc<VirtualFile>,
     file_id: FileId,
 
     max_vectored_read_bytes: Option<MaxVectoredReadBytes>,
@@ -391,9 +394,11 @@ impl ImageLayerInner {
         max_vectored_read_bytes: Option<MaxVectoredReadBytes>,
         ctx: &RequestContext,
     ) -> anyhow::Result<Self> {
-        let file = VirtualFile::open_v2(path, ctx)
-            .await
-            .context("open layer file")?;
+        let file = Arc::new(
+            VirtualFile::open_v2(path, ctx)
+                .await
+                .context("open layer file")?,
+        );
         let file_id = page_cache::next_file_id();
         let block_reader = FileBlockReader::new(&file, file_id);
         let summary_blk = block_reader
@@ -439,6 +444,7 @@ impl ImageLayerInner {
     // the reconstruct state with whatever is found.
     pub(super) async fn get_values_reconstruct_data(
         &self,
+        this: ResidentLayer,
         keyspace: KeySpace,
         reconstruct_state: &mut ValuesReconstructState,
         ctx: &RequestContext,
@@ -448,7 +454,7 @@ impl ImageLayerInner {
             .await
             .map_err(GetVectoredError::Other)?;
 
-        self.do_reads_and_update_state(reads, reconstruct_state, ctx)
+        self.do_reads_and_update_state(this, reads, reconstruct_state, ctx)
             .await;
 
         reconstruct_state.on_image_layer_visited(&self.key_range);
@@ -570,6 +576,7 @@ impl ImageLayerInner {
 
     async fn do_reads_and_update_state(
         &self,
+        this: ResidentLayer,
         reads: Vec<VectoredRead>,
         reconstruct_state: &mut ValuesReconstructState,
         ctx: &RequestContext,
@@ -580,8 +587,13 @@ impl ImageLayerInner {
             .0
             .into();
 
-        let vectored_blob_reader = VectoredBlobReader::new(&self.file);
         for read in reads.into_iter() {
+            let mut ios: HashMap<(Key, Lsn), OnDiskValueIo> = Default::default();
+            for (_, blob_meta) in read.blobs_at.as_slice() {
+                let io = reconstruct_state.update_key(&blob_meta.key, blob_meta.lsn, true);
+                ios.insert((blob_meta.key, blob_meta.lsn), io);
+            }
+
             let buf_size = read.size();
 
             if buf_size > max_vectored_read_bytes {
@@ -611,50 +623,51 @@ impl ImageLayerInner {
                 }
             }
 
-            let buf = IoBufferMut::with_capacity(buf_size);
-            let res = vectored_blob_reader.read_blobs(&read, buf, ctx).await;
+            let read_extend_residency = this.clone();
+            let read_from = self.file.clone();
+            let read_ctx = ctx.attached_child();
+            reconstruct_state
+                .spawn_io(async move {
+                    let buf = IoBufferMut::with_capacity(buf_size);
+                    let vectored_blob_reader = VectoredBlobReader::new(&read_from);
+                    let res = vectored_blob_reader.read_blobs(&read, buf, &read_ctx).await;
 
-            match res {
-                Ok(blobs_buf) => {
-                    let view = BufView::new_slice(&blobs_buf.buf);
-                    for meta in blobs_buf.blobs.iter() {
-                        let img_buf = meta.read(&view).await;
+                    match res {
+                        Ok(blobs_buf) => {
+                            let view = BufView::new_slice(&blobs_buf.buf);
+                            for meta in blobs_buf.blobs.iter() {
+                                let io: OnDiskValueIo =
+                                    ios.remove(&(meta.meta.key, meta.meta.lsn)).unwrap();
+                                let img_buf = meta.read(&view).await;
 
-                        let img_buf = match img_buf {
-                            Ok(img_buf) => img_buf,
-                            Err(e) => {
-                                reconstruct_state.on_key_error(
-                                    meta.meta.key,
-                                    PageReconstructError::Other(anyhow!(e).context(format!(
-                                        "Failed to decompress blob from virtual file {}",
-                                        self.file.path(),
-                                    ))),
-                                );
+                                let img_buf = match img_buf {
+                                    Ok(img_buf) => img_buf,
+                                    Err(e) => {
+                                        io.complete(Err(e));
+                                        continue;
+                                    }
+                                };
 
-                                continue;
+                                io.complete(Ok(OnDiskValue::RawImage(img_buf.into_bytes())));
                             }
-                        };
-                        reconstruct_state.update_key(
-                            &meta.meta.key,
-                            self.lsn,
-                            Value::Image(img_buf.into_bytes()),
-                        );
+
+                            assert!(ios.is_empty());
+                        }
+                        Err(err) => {
+                            for (_, io) in ios {
+                                io.complete(Err(std::io::Error::new(
+                                    err.kind(),
+                                    "vec read failed",
+                                )));
+                            }
+                        }
                     }
-                }
-                Err(err) => {
-                    let kind = err.kind();
-                    for (_, blob_meta) in read.blobs_at.as_slice() {
-                        reconstruct_state.on_key_error(
-                            blob_meta.key,
-                            PageReconstructError::from(anyhow!(
-                                "Failed to read blobs from virtual file {}: {}",
-                                self.file.path(),
-                                kind
-                            )),
-                        );
-                    }
-                }
-            };
+
+                    // keep layer resident until this IO is done; this spawned IO future generally outlives the
+                    // call to `self` / the `Arc<DownloadedLayer>` / the `ResidentLayer` that guarantees residency
+                    drop(read_extend_residency);
+                })
+                .await;
         }
     }
 
@@ -1069,6 +1082,7 @@ impl ImageLayerIterator<'_> {
                     Key::from_slice(&raw_key[..KEY_SIZE]),
                     self.image_layer.lsn,
                     offset,
+                    true,
                 ) {
                     break batch_plan;
                 }

pageserver/src/tenant/storage_layer/inmemory_layer.rs

@@ -8,23 +8,22 @@ use crate::assert_u64_eq_usize::{u64_to_usize, U64IsUsize, UsizeIsU64};
 use crate::config::PageServerConf;
 use crate::context::{PageContentKind, RequestContext, RequestContextBuilder};
 use crate::tenant::ephemeral_file::EphemeralFile;
+use crate::tenant::storage_layer::{OnDiskValue, OnDiskValueIo};
 use crate::tenant::timeline::GetVectoredError;
-use crate::tenant::PageReconstructError;
 use crate::virtual_file::owned_buffers_io::io_buf_ext::IoBufExt;
 use crate::{l0_flush, page_cache};
-use anyhow::{anyhow, Result};
+use anyhow::Result;
 use camino::Utf8PathBuf;
 use pageserver_api::key::CompactKey;
 use pageserver_api::key::Key;
 use pageserver_api::keyspace::KeySpace;
 use pageserver_api::models::InMemoryLayerInfo;
 use pageserver_api::shard::TenantShardId;
-use pageserver_api::value::Value;
 use std::collections::{BTreeMap, HashMap};
 use std::sync::{Arc, OnceLock};
 use std::time::Instant;
 use tracing::*;
-use utils::{bin_ser::BeSer, id::TimelineId, lsn::Lsn, vec_map::VecMap};
+use utils::{id::TimelineId, lsn::Lsn, vec_map::VecMap};
 use wal_decoder::serialized_batch::{SerializedValueBatch, SerializedValueMeta, ValueMeta};
 // avoid binding to Write (conflicts with std::io::Write)
 // while being able to use std::fmt::Write's methods
@@ -36,9 +35,7 @@ use std::sync::atomic::Ordering as AtomicOrdering;
 use std::sync::atomic::{AtomicU64, AtomicUsize};
 use tokio::sync::RwLock;
 
-use super::{
-    DeltaLayerWriter, PersistentLayerDesc, ValueReconstructSituation, ValuesReconstructState,
-};
+use super::{DeltaLayerWriter, PersistentLayerDesc, ValuesReconstructState};
 
 pub(crate) mod vectored_dio_read;
 
@@ -415,10 +412,8 @@ impl InMemoryLayer {
 
     // Look up the keys in the provided keyspace and update
     // the reconstruct state with whatever is found.
-    //
-    // If the key is cached, go no further than the cached Lsn.
     pub(crate) async fn get_values_reconstruct_data(
-        &self,
+        self: &Arc<InMemoryLayer>,
         keyspace: KeySpace,
         end_lsn: Lsn,
         reconstruct_state: &mut ValuesReconstructState,
@@ -435,6 +430,9 @@ impl InMemoryLayer {
             read: vectored_dio_read::LogicalRead<Vec<u8>>,
         }
         let mut reads: HashMap<Key, Vec<ValueRead>> = HashMap::new();
+        let mut ios: HashMap<(Key, Lsn), OnDiskValueIo> = Default::default();
+
+        let lsn_range = self.start_lsn..end_lsn;
 
         for range in keyspace.ranges.iter() {
             for (key, vec_map) in inner
@@ -442,12 +440,7 @@ impl InMemoryLayer {
                 .range(range.start.to_compact()..range.end.to_compact())
             {
                 let key = Key::from_compact(*key);
-                let lsn_range = match reconstruct_state.get_cached_lsn(&key) {
-                    Some(cached_lsn) => (cached_lsn + 1)..end_lsn,
-                    None => self.start_lsn..end_lsn,
-                };
-
-                let slice = vec_map.slice_range(lsn_range);
+                let slice = vec_map.slice_range(lsn_range.clone());
 
                 for (entry_lsn, index_entry) in slice.iter().rev() {
                     let IndexEntryUnpacked {
@@ -463,55 +456,59 @@ impl InMemoryLayer {
                             Vec::with_capacity(len as usize),
                         ),
                     });
+
+                    let io = reconstruct_state.update_key(&key, *entry_lsn, will_init);
+                    ios.insert((key, *entry_lsn), io);
+
                     if will_init {
                         break;
                     }
                 }
             }
         }
+        drop(inner); // release the lock before we spawn the IO; if it's serial-mode IO we will deadlock on the read().await below
+        let read_from = Arc::clone(self);
+        let read_ctx = ctx.attached_child();
+        reconstruct_state
+            .spawn_io(async move {
+                let inner = read_from.inner.read().await;
+                let f = vectored_dio_read::execute(
+                    &inner.file,
+                    reads
+                        .iter()
+                        .flat_map(|(_, value_reads)| value_reads.iter().map(|v| &v.read)),
+                    &read_ctx,
+                );
+                send_future::SendFuture::send(f) // https://github.com/rust-lang/rust/issues/96865
+                    .await;
 
-        // Execute the reads.
-
-        let f = vectored_dio_read::execute(
-            &inner.file,
-            reads
-                .iter()
-                .flat_map(|(_, value_reads)| value_reads.iter().map(|v| &v.read)),
-            &ctx,
-        );
-        send_future::SendFuture::send(f) // https://github.com/rust-lang/rust/issues/96865
-            .await;
-
-        // Process results into the reconstruct state
-        'next_key: for (key, value_reads) in reads {
-            for ValueRead { entry_lsn, read } in value_reads {
-                match read.into_result().expect("we run execute() above") {
-                    Err(e) => {
-                        reconstruct_state.on_key_error(key, PageReconstructError::from(anyhow!(e)));
-                        continue 'next_key;
-                    }
-                    Ok(value_buf) => {
-                        let value = Value::des(&value_buf);
-                        if let Err(e) = value {
-                            reconstruct_state
-                                .on_key_error(key, PageReconstructError::from(anyhow!(e)));
-                            continue 'next_key;
+                for (key, value_reads) in reads {
+                    for ValueRead { entry_lsn, read } in value_reads {
+                        let io = ios.remove(&(key, entry_lsn)).expect("sender must exist");
+                        match read.into_result().expect("we run execute() above") {
+                            Err(e) => {
+                                io.complete(Err(std::io::Error::new(
+                                    e.kind(),
+                                    "dio vec read failed",
+                                )));
+                            }
+                            Ok(value_buf) => {
+                                io.complete(Ok(OnDiskValue::WalRecordOrImage(value_buf.into())));
+                            }
                         }
-
-                        let key_situation =
-                            reconstruct_state.update_key(&key, entry_lsn, value.unwrap());
-                        if key_situation == ValueReconstructSituation::Complete {
-                            // TODO: metric to see if we fetched more values than necessary
-                            continue 'next_key;
-                        }
-
-                        // process the next value in the next iteration of the loop
                     }
                 }
-            }
-        }
 
-        reconstruct_state.on_lsn_advanced(&keyspace, self.start_lsn);
+                assert!(ios.is_empty());
+
+                // Keep layer existent until this IO is done;
+                // This is kinda forced for InMemoryLayer because we need to inner.read() anyway,
+                // but it's less obvious for DeltaLayer and ImageLayer. So, keep this explicit
+                // drop for consistency among all three layer types.
+                drop(inner);
+                drop(read_from);
+            })
+            .await;
 
         Ok(())
     }
@@ -606,6 +603,7 @@ impl InMemoryLayer {
         // Write the batch to the file
         inner.file.write_raw(&raw, ctx).await?;
         let new_size = inner.file.len();
+
         let expected_new_len = base_offset
             .checked_add(raw.len().into_u64())
             // write_raw would error if we were to overflow u64.

pageserver/src/tenant/storage_layer/layer.rs

@@ -308,7 +308,7 @@ impl Layer {
         reconstruct_data: &mut ValuesReconstructState,
         ctx: &RequestContext,
     ) -> Result<(), GetVectoredError> {
-        let layer = self
+        let downloaded = self
             .0
             .get_or_maybe_download(true, Some(ctx))
             .await
@@ -318,11 +318,15 @@ impl Layer {
                 }
                 other => GetVectoredError::Other(anyhow::anyhow!(other)),
             })?;
+        let this = ResidentLayer {
+            downloaded: downloaded.clone(),
+            owner: self.clone(),
+        };
 
         self.record_access(ctx);
 
-        layer
-            .get_values_reconstruct_data(keyspace, lsn_range, reconstruct_data, &self.0, ctx)
+        downloaded
+            .get_values_reconstruct_data(this, keyspace, lsn_range, reconstruct_data, ctx)
             .instrument(tracing::debug_span!("get_values_reconstruct_data", layer=%self))
             .await
             .map_err(|err| match err {
@@ -1768,25 +1772,25 @@ impl DownloadedLayer {
 
     async fn get_values_reconstruct_data(
         &self,
+        this: ResidentLayer,
         keyspace: KeySpace,
         lsn_range: Range<Lsn>,
         reconstruct_data: &mut ValuesReconstructState,
-        owner: &Arc<LayerInner>,
         ctx: &RequestContext,
     ) -> Result<(), GetVectoredError> {
         use LayerKind::*;
 
         match self
-            .get(owner, ctx)
+            .get(&this.owner.0, ctx)
             .await
             .map_err(GetVectoredError::Other)?
         {
             Delta(d) => {
-                d.get_values_reconstruct_data(keyspace, lsn_range, reconstruct_data, ctx)
+                d.get_values_reconstruct_data(this, keyspace, lsn_range, reconstruct_data, ctx)
                     .await
             }
             Image(i) => {
-                i.get_values_reconstruct_data(keyspace, reconstruct_data, ctx)
+                i.get_values_reconstruct_data(this, keyspace, reconstruct_data, ctx)
                     .await
             }
         }

pageserver/src/tenant/storage_layer/layer/tests.rs

@@ -11,7 +11,10 @@ use super::failpoints::{Failpoint, FailpointKind};
 use super::*;
 use crate::{
     context::DownloadBehavior,
-    tenant::{harness::test_img, storage_layer::LayerVisibilityHint},
+    tenant::{
+        harness::test_img,
+        storage_layer::{IoConcurrency, LayerVisibilityHint},
+    },
 };
 use crate::{task_mgr::TaskKind, tenant::harness::TenantHarness};
 
@@ -31,6 +34,7 @@ async fn smoke_test() {
     let span = h.span();
     let download_span = span.in_scope(|| tracing::info_span!("downloading", timeline_id = 1));
     let (tenant, _) = h.load().await;
+    let io_concurrency = IoConcurrency::spawn_for_test();
 
     let ctx = RequestContext::new(TaskKind::UnitTest, DownloadBehavior::Download);
 
@@ -89,7 +93,7 @@ async fn smoke_test() {
     };
 
     let img_before = {
-        let mut data = ValuesReconstructState::default();
+        let mut data = ValuesReconstructState::new(io_concurrency.clone());
         layer
             .get_values_reconstruct_data(
                 controlfile_keyspace.clone(),
@@ -99,10 +103,13 @@ async fn smoke_test() {
             )
             .await
             .unwrap();
+
         data.keys
             .remove(&CONTROLFILE_KEY)
             .expect("must be present")
-            .expect("should not error")
+            .collect_pending_ios()
+            .await
+            .expect("must not error")
             .img
             .take()
             .expect("tenant harness writes the control file")
@@ -121,7 +128,7 @@ async fn smoke_test() {
 
     // on accesses when the layer is evicted, it will automatically be downloaded.
     let img_after = {
-        let mut data = ValuesReconstructState::default();
+        let mut data = ValuesReconstructState::new(io_concurrency.clone());
         layer
             .get_values_reconstruct_data(
                 controlfile_keyspace.clone(),
@@ -135,7 +142,9 @@ async fn smoke_test() {
         data.keys
             .remove(&CONTROLFILE_KEY)
             .expect("must be present")
-            .expect("should not error")
+            .collect_pending_ios()
+            .await
+            .expect("must not error")
             .img
             .take()
             .expect("tenant harness writes the control file")