Enable blob_batch deletions

.github/actionlint.yml

@@ -23,5 +23,3 @@ config-variables:
   - BENCHMARK_INGEST_TARGET_PROJECTID
   - PGREGRESS_PG16_PROJECT_ID
   - PGREGRESS_PG17_PROJECT_ID
-  - SLACK_ON_CALL_QA_STAGING_STREAM
-  - DEV_AWS_OIDC_ROLE_MANAGE_BENCHMARK_EC2_VMS_ARN

.github/actions/allure-report-generate/action.yml

@@ -7,9 +7,10 @@ inputs:
     type: boolean
     required: false
     default: false
-  aws-oicd-role-arn:
-    description: 'OIDC role arn to interract with S3'
-    required: true
+  aws_oicd_role_arn:
+    description: 'the OIDC role arn to (re-)acquire for allure report upload - if not set call must acquire OIDC role'
+    required: false
+    default: ''
 
 outputs:
   base-url:
@@ -83,11 +84,12 @@ runs:
         ALLURE_VERSION: 2.27.0
         ALLURE_ZIP_SHA256: b071858fb2fa542c65d8f152c5c40d26267b2dfb74df1f1608a589ecca38e777
 
-    - uses: aws-actions/configure-aws-credentials@v4
-      if: ${{ !cancelled() }}
+    - name: (Re-)configure AWS credentials # necessary to upload reports to S3 after a long-running test
+      if: ${{ !cancelled() && (inputs.aws_oicd_role_arn != '') }}
+      uses: aws-actions/configure-aws-credentials@v4
       with:
         aws-region: eu-central-1
-        role-to-assume: ${{ inputs.aws-oicd-role-arn }}
+        role-to-assume: ${{ inputs.aws_oicd_role_arn }}
         role-duration-seconds: 3600 # 1 hour should be more than enough to upload report
 
     # Potentially we could have several running build for the same key (for example, for the main branch), so we use improvised lock for this

.github/actions/allure-report-store/action.yml

@@ -8,9 +8,10 @@ inputs:
   unique-key:
     description: 'string to distinguish different results in the same run'
     required: true
-  aws-oicd-role-arn:
-    description: 'OIDC role arn to interract with S3'
-    required: true
+  aws_oicd_role_arn:
+    description: 'the OIDC role arn to (re-)acquire for allure report upload - if not set call must acquire OIDC role'
+    required: false
+    default: ''
 
 runs:
   using: "composite"
@@ -35,11 +36,12 @@ runs:
       env:
         REPORT_DIR: ${{ inputs.report-dir }}
 
-    - uses: aws-actions/configure-aws-credentials@v4
-      if: ${{ !cancelled() }}
+    - name: (Re-)configure AWS credentials # necessary to upload reports to S3 after a long-running test
+      if: ${{ !cancelled() && (inputs.aws_oicd_role_arn != '') }}
+      uses: aws-actions/configure-aws-credentials@v4
       with:
         aws-region: eu-central-1
-        role-to-assume: ${{ inputs.aws-oicd-role-arn }}
+        role-to-assume: ${{ inputs.aws_oicd_role_arn }}
         role-duration-seconds: 3600 # 1 hour should be more than enough to upload report
 
     - name: Upload test results

.github/actions/download/action.yml

@@ -15,19 +15,10 @@ inputs:
   prefix:
     description: "S3 prefix. Default is '${GITHUB_RUN_ID}/${GITHUB_RUN_ATTEMPT}'"
     required: false
-  aws-oicd-role-arn:
-    description: 'OIDC role arn to interract with S3'
-    required: true
 
 runs:
   using: "composite"
   steps:
-    - uses: aws-actions/configure-aws-credentials@v4
-      with:
-        aws-region: eu-central-1
-        role-to-assume: ${{ inputs.aws-oicd-role-arn }}
-        role-duration-seconds: 3600
-
     - name: Download artifact
       id: download-artifact
       shell: bash -euxo pipefail {0}

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

@@ -48,9 +48,10 @@ inputs:
     description: 'benchmark durations JSON'
     required: false
     default: '{}'
-  aws-oicd-role-arn:
-    description: 'OIDC role arn to interract with S3'
-    required: true
+  aws_oicd_role_arn:
+    description: 'the OIDC role arn to (re-)acquire for allure report upload - if not set call must acquire OIDC role'
+    required: false
+    default: ''
 
 runs:
   using: "composite"
@@ -61,7 +62,6 @@ runs:
       with:
         name: neon-${{ runner.os }}-${{ runner.arch }}-${{ inputs.build_type }}-artifact
         path: /tmp/neon
-        aws-oicd-role-arn: ${{ inputs.aws-oicd-role-arn }}
 
     - name: Download Neon binaries for the previous release
       if: inputs.build_type != 'remote'
@@ -70,7 +70,6 @@ runs:
         name: neon-${{ runner.os }}-${{ runner.arch }}-${{ inputs.build_type }}-artifact
         path: /tmp/neon-previous
         prefix: latest
-        aws-oicd-role-arn: ${{ inputs.aws-oicd-role-arn }}
 
     - name: Download compatibility snapshot
       if: inputs.build_type != 'remote'
@@ -82,7 +81,6 @@ runs:
         # The lack of compatibility snapshot (for example, for the new Postgres version)
         # shouldn't fail the whole job. Only relevant test should fail.
         skip-if-does-not-exist: true
-        aws-oicd-role-arn: ${{ inputs.aws-oicd-role-arn }}
 
     - name: Checkout
       if: inputs.needs_postgres_source == 'true'
@@ -169,7 +167,7 @@ runs:
         fi
 
         if [[ $BUILD_TYPE == "debug" && $RUNNER_ARCH == 'X64' ]]; then
-          cov_prefix=()
+          cov_prefix=(scripts/coverage "--profraw-prefix=$GITHUB_JOB" --dir=/tmp/coverage run)
         else
           cov_prefix=()
         fi
@@ -220,19 +218,17 @@ runs:
         # The lack of compatibility snapshot shouldn't fail the job
         # (for example if we didn't run the test for non build-and-test workflow)
         skip-if-does-not-exist: true
-        aws-oicd-role-arn: ${{ inputs.aws-oicd-role-arn }}
 
-    - uses: aws-actions/configure-aws-credentials@v4
-      if: ${{ !cancelled() }}
+    - name: (Re-)configure AWS credentials # necessary to upload reports to S3 after a long-running test
+      if: ${{ !cancelled() && (inputs.aws_oicd_role_arn != '') }}
+      uses: aws-actions/configure-aws-credentials@v4
       with:
         aws-region: eu-central-1
-        role-to-assume: ${{ inputs.aws-oicd-role-arn }}
+        role-to-assume: ${{ inputs.aws_oicd_role_arn }}
         role-duration-seconds: 3600 # 1 hour should be more than enough to upload report
-
     - name: Upload test results
       if: ${{ !cancelled() }}
       uses: ./.github/actions/allure-report-store
       with:
         report-dir: /tmp/test_output/allure/results
         unique-key: ${{ inputs.build_type }}-${{ inputs.pg_version }}
-        aws-oicd-role-arn: ${{ inputs.aws-oicd-role-arn }}

.github/actions/save-coverage-data/action.yml

@@ -14,11 +14,9 @@ runs:
         name: coverage-data-artifact
         path: /tmp/coverage
         skip-if-does-not-exist: true # skip if there's no previous coverage to download
-        aws-oicd-role-arn: ${{ inputs.aws-oicd-role-arn }}
 
     - name: Upload coverage data
       uses: ./.github/actions/upload
       with:
         name: coverage-data-artifact
         path: /tmp/coverage
-        aws-oicd-role-arn: ${{ inputs.aws-oicd-role-arn }}

.github/actions/upload/action.yml

@@ -14,10 +14,6 @@ inputs:
   prefix:
     description: "S3 prefix. Default is '${GITHUB_SHA}/${GITHUB_RUN_ID}/${GITHUB_RUN_ATTEMPT}'"
     required: false
-  aws-oicd-role-arn:
-    description: "the OIDC role arn for aws auth"
-    required: false
-    default: ""
 
 runs:
   using: "composite"
@@ -57,13 +53,6 @@ runs:
 
         echo 'SKIPPED=false' >> $GITHUB_OUTPUT
 
-    - name: Configure AWS credentials
-      uses: aws-actions/configure-aws-credentials@v4
-      with:
-        aws-region: eu-central-1
-        role-to-assume: ${{ inputs.aws-oicd-role-arn }}
-        role-duration-seconds: 3600
-
     - name: Upload artifact
       if: ${{ steps.prepare-artifact.outputs.SKIPPED == 'false' }}
       shell: bash -euxo pipefail {0}

.github/workflows/_benchmarking_preparation.yml

@@ -70,7 +70,6 @@ jobs:
         name: neon-${{ runner.os }}-${{ runner.arch }}-release-artifact
         path: /tmp/neon/
         prefix: latest
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
 
     # we create a table that has one row for each database that we want to restore with the status whether the restore is done
     - name: Create benchmark_restore_status table if it does not exist

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

@@ -31,13 +31,12 @@ defaults:
 env:
   RUST_BACKTRACE: 1
   COPT: '-Werror'
+  AWS_ACCESS_KEY_ID: ${{ secrets.AWS_ACCESS_KEY_DEV }}
+  AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_KEY_DEV }}
 
 jobs:
   build-neon:
     runs-on: ${{ fromJson(format('["self-hosted", "{0}"]', inputs.arch == 'arm64' && 'large-arm64' || 'large')) }}
-    permissions:
-      id-token: write # aws-actions/configure-aws-credentials
-      contents: read
     container:
       image: ${{ inputs.build-tools-image }}
       credentials:
@@ -90,7 +89,7 @@ jobs:
         run: |
           CARGO_FEATURES="--features testing"
           if [[ $BUILD_TYPE == "debug" && $ARCH == 'x64' ]]; then
-            cov_prefix=""
+            cov_prefix="scripts/coverage --profraw-prefix=$GITHUB_JOB --dir=/tmp/coverage run"
             CARGO_FLAGS="--locked"
           elif [[ $BUILD_TYPE == "debug" ]]; then
             cov_prefix=""
@@ -206,13 +205,6 @@ jobs:
             done
           fi
 
-      - name: Configure AWS credentials
-        uses: aws-actions/configure-aws-credentials@v4
-        with:
-          aws-region: eu-central-1
-          role-to-assume: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
-          role-duration-seconds: 18000 # 5 hours
-
       - name: Run rust tests
         env:
           NEXTEST_RETRIES: 3
@@ -264,7 +256,6 @@ jobs:
         with:
           name: neon-${{ runner.os }}-${{ runner.arch }}-${{ inputs.build-type }}-artifact
           path: /tmp/neon
-          aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
 
       # XXX: keep this after the binaries.list is formed, so the coverage can properly work later
       - name: Merge and upload coverage data
@@ -274,10 +265,6 @@ jobs:
   regress-tests:
     # Don't run regression tests on debug arm64 builds
     if: inputs.build-type != 'debug' || inputs.arch != 'arm64'
-    permissions:
-      id-token: write # aws-actions/configure-aws-credentials
-      contents: read
-      statuses: write
     needs: [ build-neon ]
     runs-on: ${{ fromJson(format('["self-hosted", "{0}"]', inputs.arch == 'arm64' && 'large-arm64' || 'large')) }}
     container:
@@ -296,7 +283,7 @@ jobs:
           submodules: true
 
       - name: Pytest regression tests
-        continue-on-error: ${{ matrix.lfc_state == 'with-lfc' && inputs.build-type == 'debug' }}
+        continue-on-error: ${{ matrix.lfc_state == 'with-lfc' }}
         uses: ./.github/actions/run-python-test-set
         timeout-minutes: 60
         with:
@@ -308,7 +295,6 @@ jobs:
           real_s3_region: eu-central-1
           rerun_failed: true
           pg_version: ${{ matrix.pg_version }}
-          aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
         env:
           TEST_RESULT_CONNSTR: ${{ secrets.REGRESS_TEST_RESULT_CONNSTR_NEW }}
           CHECK_ONDISK_DATA_COMPATIBILITY: nonempty

.github/workflows/benchmarking.yml

@@ -105,7 +105,6 @@ jobs:
         name: neon-${{ runner.os }}-${{ runner.arch }}-release-artifact
         path: /tmp/neon/
         prefix: latest
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
 
     - name: Create Neon Project
       id: create-neon-project
@@ -123,7 +122,7 @@ jobs:
         run_in_parallel: false
         save_perf_report: ${{ env.SAVE_PERF_REPORT }}
         pg_version: ${{ env.DEFAULT_PG_VERSION }}
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
+        aws_oicd_role_arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
         # Set --sparse-ordering option of pytest-order plugin
         # to ensure tests are running in order of appears in the file.
         # It's important for test_perf_pgbench.py::test_pgbench_remote_* tests
@@ -153,7 +152,7 @@ jobs:
       if: ${{ !cancelled() }}
       uses: ./.github/actions/allure-report-generate
       with:
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
+        aws_oicd_role_arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
 
     - name: Post to a Slack channel
       if: ${{ github.event.schedule && failure() }}
@@ -205,7 +204,6 @@ jobs:
         name: neon-${{ runner.os }}-${{ runner.arch }}-release-artifact
         path: /tmp/neon/
         prefix: latest
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
 
     - name: Run Logical Replication benchmarks
       uses: ./.github/actions/run-python-test-set
@@ -216,7 +214,7 @@ jobs:
         save_perf_report: ${{ env.SAVE_PERF_REPORT }}
         extra_params: -m remote_cluster --timeout 5400
         pg_version: ${{ env.DEFAULT_PG_VERSION }}
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
+        aws_oicd_role_arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
       env:
         VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
         PERF_TEST_RESULT_CONNSTR: "${{ secrets.PERF_TEST_RESULT_CONNSTR }}"
@@ -233,7 +231,7 @@ jobs:
         save_perf_report: ${{ env.SAVE_PERF_REPORT }}
         extra_params: -m remote_cluster --timeout 5400
         pg_version: ${{ env.DEFAULT_PG_VERSION }}
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
+        aws_oicd_role_arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
       env:
         VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
         PERF_TEST_RESULT_CONNSTR: "${{ secrets.PERF_TEST_RESULT_CONNSTR }}"
@@ -245,7 +243,7 @@ jobs:
       uses: ./.github/actions/allure-report-generate
       with:
         store-test-results-into-db: true
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
+        aws_oicd_role_arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
       env:
         REGRESS_TEST_RESULT_CONNSTR_NEW: ${{ secrets.REGRESS_TEST_RESULT_CONNSTR_NEW }}
 
@@ -407,7 +405,6 @@ jobs:
         name: neon-${{ runner.os }}-${{ runner.arch }}-release-artifact
         path: /tmp/neon/
         prefix: latest
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
 
     - name: Create Neon Project
       if: contains(fromJson('["neonvm-captest-new", "neonvm-captest-freetier", "neonvm-azure-captest-freetier", "neonvm-azure-captest-new"]'), matrix.platform)
@@ -455,7 +452,7 @@ jobs:
         save_perf_report: ${{ env.SAVE_PERF_REPORT }}
         extra_params: -m remote_cluster --timeout 21600 -k test_pgbench_remote_init
         pg_version: ${{ env.DEFAULT_PG_VERSION }}
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
+        aws_oicd_role_arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
       env:
         BENCHMARK_CONNSTR: ${{ steps.set-up-connstr.outputs.connstr }}
         VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
@@ -470,7 +467,7 @@ jobs:
         save_perf_report: ${{ env.SAVE_PERF_REPORT }}
         extra_params: -m remote_cluster --timeout 21600 -k test_pgbench_remote_simple_update
         pg_version: ${{ env.DEFAULT_PG_VERSION }}
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
+        aws_oicd_role_arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
       env:
         BENCHMARK_CONNSTR: ${{ steps.set-up-connstr.outputs.connstr }}
         VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
@@ -485,7 +482,7 @@ jobs:
         save_perf_report: ${{ env.SAVE_PERF_REPORT }}
         extra_params: -m remote_cluster --timeout 21600 -k test_pgbench_remote_select_only
         pg_version: ${{ env.DEFAULT_PG_VERSION }}
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
+        aws_oicd_role_arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
       env:
         BENCHMARK_CONNSTR: ${{ steps.set-up-connstr.outputs.connstr }}
         VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
@@ -503,7 +500,7 @@ jobs:
       if: ${{ !cancelled() }}
       uses: ./.github/actions/allure-report-generate
       with:
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
+        aws_oicd_role_arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
 
     - name: Post to a Slack channel
       if: ${{ github.event.schedule && failure() }}
@@ -614,7 +611,7 @@ jobs:
         save_perf_report: ${{ env.SAVE_PERF_REPORT }}
         extra_params: -m remote_cluster --timeout 21600 -k test_pgvector_indexing
         pg_version: ${{ env.DEFAULT_PG_VERSION }}
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
+        aws_oicd_role_arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
       env:
         VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
         PERF_TEST_RESULT_CONNSTR: "${{ secrets.PERF_TEST_RESULT_CONNSTR }}"
@@ -629,7 +626,7 @@ jobs:
         save_perf_report: ${{ env.SAVE_PERF_REPORT }}
         extra_params: -m remote_cluster --timeout 21600
         pg_version: ${{ env.DEFAULT_PG_VERSION }}
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
+        aws_oicd_role_arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
       env:
         BENCHMARK_CONNSTR: ${{ steps.set-up-connstr.outputs.connstr }}
         VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
@@ -640,7 +637,7 @@ jobs:
       if: ${{ !cancelled() }}
       uses: ./.github/actions/allure-report-generate
       with:
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
+        aws_oicd_role_arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
 
     - name: Post to a Slack channel
       if: ${{ github.event.schedule && failure() }}
@@ -711,7 +708,6 @@ jobs:
         name: neon-${{ runner.os }}-${{ runner.arch }}-release-artifact
         path: /tmp/neon/
         prefix: latest
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
 
     - name: Set up Connection String
       id: set-up-connstr
@@ -743,7 +739,7 @@ jobs:
         save_perf_report: ${{ env.SAVE_PERF_REPORT }}
         extra_params: -m remote_cluster --timeout 43200 -k test_clickbench
         pg_version: ${{ env.DEFAULT_PG_VERSION }}
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
+        aws_oicd_role_arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
       env:
         VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
         PERF_TEST_RESULT_CONNSTR: "${{ secrets.PERF_TEST_RESULT_CONNSTR }}"
@@ -757,7 +753,7 @@ jobs:
       if: ${{ !cancelled() }}
       uses: ./.github/actions/allure-report-generate
       with:
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
+        aws_oicd_role_arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
 
     - name: Post to a Slack channel
       if: ${{ github.event.schedule && failure() }}
@@ -822,7 +818,6 @@ jobs:
         name: neon-${{ runner.os }}-${{ runner.arch }}-release-artifact
         path: /tmp/neon/
         prefix: latest
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
 
     - name: Get Connstring Secret Name
       run: |
@@ -861,7 +856,7 @@ jobs:
         save_perf_report: ${{ env.SAVE_PERF_REPORT }}
         extra_params: -m remote_cluster --timeout 21600 -k test_tpch
         pg_version: ${{ env.DEFAULT_PG_VERSION }}
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
+        aws_oicd_role_arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
       env:
         VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
         PERF_TEST_RESULT_CONNSTR: "${{ secrets.PERF_TEST_RESULT_CONNSTR }}"
@@ -873,7 +868,7 @@ jobs:
       if: ${{ !cancelled() }}
       uses: ./.github/actions/allure-report-generate
       with:
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
+        aws_oicd_role_arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
 
     - name: Post to a Slack channel
       if: ${{ github.event.schedule && failure() }}
@@ -931,7 +926,6 @@ jobs:
         name: neon-${{ runner.os }}-${{ runner.arch }}-release-artifact
         path: /tmp/neon/
         prefix: latest
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
 
     - name: Set up Connection String
       id: set-up-connstr
@@ -963,7 +957,7 @@ jobs:
         save_perf_report: ${{ env.SAVE_PERF_REPORT }}
         extra_params: -m remote_cluster --timeout 21600 -k test_user_examples
         pg_version: ${{ env.DEFAULT_PG_VERSION }}
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
+        aws_oicd_role_arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
       env:
         VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
         PERF_TEST_RESULT_CONNSTR: "${{ secrets.PERF_TEST_RESULT_CONNSTR }}"
@@ -974,7 +968,7 @@ jobs:
       if: ${{ !cancelled() }}
       uses: ./.github/actions/allure-report-generate
       with:
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
+        aws_oicd_role_arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
 
     - name: Post to a Slack channel
       if: ${{ github.event.schedule && failure() }}

.github/workflows/build_and_test.yml

@@ -21,6 +21,8 @@ concurrency:
 env:
   RUST_BACKTRACE: 1
   COPT: '-Werror'
+  AWS_ACCESS_KEY_ID: ${{ secrets.AWS_ACCESS_KEY_DEV }}
+  AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_KEY_DEV }}
   # A concurrency group that we use for e2e-tests runs, matches `concurrency.group` above with `github.repository` as a prefix
   E2E_CONCURRENCY_GROUP: ${{ github.repository }}-e2e-tests-${{ github.ref_name }}-${{ github.ref_name == 'main' && github.sha || 'anysha' }}
 
@@ -253,15 +255,15 @@ jobs:
       build-tools-image: ${{ needs.build-build-tools-image.outputs.image }}-bookworm
       build-tag: ${{ needs.tag.outputs.build-tag }}
       build-type: ${{ matrix.build-type }}
-      # Run tests on all Postgres versions in release builds and only on the latest version in debug builds.
-      # Run without LFC on v17 release and debug builds only. For all the other cases LFC is enabled.
+      # Run tests on all Postgres versions in release builds and only on the latest version in debug builds
+      # run without LFC on v17 release only
       test-cfg: |
-        ${{ matrix.build-type == 'release' && '[{"pg_version":"v14", "lfc_state": "with-lfc"},
-                                                {"pg_version":"v15", "lfc_state": "with-lfc"},
-                                                {"pg_version":"v16", "lfc_state": "with-lfc"},
-                                                {"pg_version":"v17", "lfc_state": "with-lfc"},
-                                                {"pg_version":"v17", "lfc_state": "without-lfc"}]'
-                                           || '[{"pg_version":"v17", "lfc_state": "without-lfc" }]' }}
+        ${{ matrix.build-type == 'release' && '[{"pg_version":"v14", "lfc_state": "without-lfc"},
+                                                {"pg_version":"v15", "lfc_state": "without-lfc"},
+                                                {"pg_version":"v16", "lfc_state": "without-lfc"},
+                                                {"pg_version":"v17", "lfc_state": "without-lfc"},
+                                                {"pg_version":"v17", "lfc_state": "with-lfc"}]'
+                                           || '[{"pg_version":"v17", "lfc_state": "without-lfc"}]' }}
     secrets: inherit
 
   # Keep `benchmarks` job outside of `build-and-test-locally` workflow to make job failures non-blocking
@@ -303,11 +305,6 @@ jobs:
   benchmarks:
     if: github.ref_name == 'main' || contains(github.event.pull_request.labels.*.name, 'run-benchmarks')
     needs: [ check-permissions, build-and-test-locally, build-build-tools-image, get-benchmarks-durations ]
-    permissions:
-      id-token: write # aws-actions/configure-aws-credentials
-      statuses: write
-      contents: write
-      pull-requests: write
     runs-on: [ self-hosted, small ]
     container:
       image: ${{ needs.build-build-tools-image.outputs.image }}-bookworm
@@ -336,7 +333,6 @@ jobs:
           extra_params: --splits 5 --group ${{ matrix.pytest_split_group }}
           benchmark_durations: ${{ needs.get-benchmarks-durations.outputs.json }}
           pg_version: v16
-          aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
         env:
           VIP_VAP_ACCESS_TOKEN: "${{ secrets.VIP_VAP_ACCESS_TOKEN }}"
           PERF_TEST_RESULT_CONNSTR: "${{ secrets.PERF_TEST_RESULT_CONNSTR }}"
@@ -349,11 +345,6 @@ jobs:
   report-benchmarks-failures:
     needs: [ benchmarks, create-test-report ]
     if: github.ref_name == 'main' && failure() && needs.benchmarks.result == 'failure'
-    permissions:
-      id-token: write # aws-actions/configure-aws-credentials
-      statuses: write
-      contents: write
-      pull-requests: write
     runs-on: ubuntu-22.04
 
     steps:
@@ -369,11 +360,6 @@ jobs:
   create-test-report:
     needs: [ check-permissions, build-and-test-locally, coverage-report, build-build-tools-image, benchmarks ]
     if: ${{ !cancelled() && contains(fromJSON('["skipped", "success"]'), needs.check-permissions.result) }}
-    permissions:
-      id-token: write # aws-actions/configure-aws-credentials
-      statuses: write
-      contents: write
-      pull-requests: write
     outputs:
       report-url: ${{ steps.create-allure-report.outputs.report-url }}
 
@@ -394,7 +380,6 @@ jobs:
         uses: ./.github/actions/allure-report-generate
         with:
           store-test-results-into-db: true
-          aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
         env:
           REGRESS_TEST_RESULT_CONNSTR_NEW: ${{ secrets.REGRESS_TEST_RESULT_CONNSTR_NEW }}
 
@@ -426,10 +411,6 @@ jobs:
   coverage-report:
     if: ${{ !startsWith(github.ref_name, 'release') }}
     needs: [ check-permissions, build-build-tools-image, build-and-test-locally ]
-    permissions:
-      id-token: write # aws-actions/configure-aws-credentials
-      statuses: write
-      contents: write
     runs-on: [ self-hosted, small ]
     container:
       image: ${{ needs.build-build-tools-image.outputs.image }}-bookworm
@@ -456,14 +437,12 @@ jobs:
         with:
           name: neon-${{ runner.os }}-${{ runner.arch }}-${{ matrix.build_type }}-artifact
           path: /tmp/neon
-          aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
 
       - name: Get coverage artifact
         uses: ./.github/actions/download
         with:
           name: coverage-data-artifact
           path: /tmp/coverage
-          aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
 
       - name: Merge coverage data
         run: scripts/coverage "--profraw-prefix=$GITHUB_JOB" --dir=/tmp/coverage merge
@@ -594,10 +573,6 @@ jobs:
   neon-image:
     needs: [ neon-image-arch, tag ]
     runs-on: ubuntu-22.04
-    permissions:
-      id-token: write # aws-actions/configure-aws-credentials
-      statuses: write
-      contents: read
 
     steps:
       - uses: docker/login-action@v3
@@ -612,15 +587,11 @@ jobs:
                                              neondatabase/neon:${{ needs.tag.outputs.build-tag }}-bookworm-x64 \
                                              neondatabase/neon:${{ needs.tag.outputs.build-tag }}-bookworm-arm64
 
-      - name: Configure AWS credentials
-        uses: aws-actions/configure-aws-credentials@v4
+      - uses: docker/login-action@v3
         with:
-          aws-region: eu-central-1
-          role-to-assume: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
-          role-duration-seconds: 3600
-
-      - name: Login to Amazon Dev ECR
-        uses: aws-actions/amazon-ecr-login@v2
+          registry: 369495373322.dkr.ecr.eu-central-1.amazonaws.com
+          username: ${{ secrets.AWS_ACCESS_KEY_DEV }}
+          password: ${{ secrets.AWS_SECRET_KEY_DEV }}
 
       - name: Push multi-arch image to ECR
         run: |
@@ -629,10 +600,6 @@ jobs:
 
   compute-node-image-arch:
     needs: [ check-permissions, build-build-tools-image, tag ]
-    permissions:
-      id-token: write # aws-actions/configure-aws-credentials
-      statuses: write
-      contents: read
     strategy:
       fail-fast: false
       matrix:
@@ -673,15 +640,11 @@ jobs:
           username: ${{ secrets.NEON_DOCKERHUB_USERNAME }}
           password: ${{ secrets.NEON_DOCKERHUB_PASSWORD }}
 
-      - name: Configure AWS credentials
-        uses: aws-actions/configure-aws-credentials@v4
+      - uses: docker/login-action@v3
         with:
-          aws-region: eu-central-1
-          role-to-assume: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
-          role-duration-seconds: 3600
-
-      - name: Login to Amazon Dev ECR
-        uses: aws-actions/amazon-ecr-login@v2
+          registry: 369495373322.dkr.ecr.eu-central-1.amazonaws.com
+          username: ${{ secrets.AWS_ACCESS_KEY_DEV }}
+          password: ${{ secrets.AWS_SECRET_KEY_DEV }}
 
       - uses: docker/login-action@v3
         with:
@@ -754,10 +717,6 @@ jobs:
 
   compute-node-image:
     needs: [ compute-node-image-arch, tag ]
-    permissions:
-      id-token: write # aws-actions/configure-aws-credentials
-      statuses: write
-      contents: read
     runs-on: ubuntu-22.04
 
     strategy:
@@ -802,15 +761,11 @@ jobs:
                                              neondatabase/compute-tools:${{ needs.tag.outputs.build-tag }}-${{ matrix.version.debian }}-x64 \
                                              neondatabase/compute-tools:${{ needs.tag.outputs.build-tag }}-${{ matrix.version.debian }}-arm64
 
-      - name: Configure AWS credentials
-        uses: aws-actions/configure-aws-credentials@v4
+      - uses: docker/login-action@v3
         with:
-          aws-region: eu-central-1
-          role-to-assume: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
-          role-duration-seconds: 3600
-
-      - name: Login to Amazon Dev ECR
-        uses: aws-actions/amazon-ecr-login@v2
+          registry: 369495373322.dkr.ecr.eu-central-1.amazonaws.com
+          username: ${{ secrets.AWS_ACCESS_KEY_DEV }}
+          password: ${{ secrets.AWS_SECRET_KEY_DEV }}
 
       - name: Push multi-arch compute-node-${{ matrix.version.pg }} image to ECR
         run: |
@@ -840,7 +795,7 @@ jobs:
           - pg: v17
             debian: bookworm
     env:
-      VM_BUILDER_VERSION: v0.37.1
+      VM_BUILDER_VERSION: v0.35.0
 
     steps:
       - uses: actions/checkout@v4
@@ -935,9 +890,7 @@ jobs:
     runs-on: ubuntu-22.04
 
     permissions:
-      id-token: write # aws-actions/configure-aws-credentials
-      statuses: write
-      contents: read
+      id-token: write # for `aws-actions/configure-aws-credentials`
 
     env:
       VERSIONS: v14 v15 v16 v17
@@ -948,15 +901,12 @@ jobs:
           username: ${{ secrets.NEON_DOCKERHUB_USERNAME }}
           password: ${{ secrets.NEON_DOCKERHUB_PASSWORD }}
 
-      - name: Configure AWS credentials
-        uses: aws-actions/configure-aws-credentials@v4
+      - name: Login to dev ECR
+        uses: docker/login-action@v3
         with:
-          aws-region: eu-central-1
-          role-to-assume: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
-          role-duration-seconds: 3600
-
-      - name: Login to Amazon Dev ECR
-        uses: aws-actions/amazon-ecr-login@v2
+          registry: 369495373322.dkr.ecr.eu-central-1.amazonaws.com
+          username: ${{ secrets.AWS_ACCESS_KEY_DEV }}
+          password: ${{ secrets.AWS_SECRET_KEY_DEV }}
 
       - name: Copy vm-compute-node images to ECR
         run: |
@@ -1035,11 +985,6 @@ jobs:
   trigger-custom-extensions-build-and-wait:
     needs: [ check-permissions, tag ]
     runs-on: ubuntu-22.04
-    permissions:
-      id-token: write # aws-actions/configure-aws-credentials
-      statuses: write
-      contents: write
-      pull-requests: write
     steps:
       - name: Set PR's status to pending and request a remote CI test
         run: |
@@ -1115,79 +1060,12 @@ jobs:
     needs: [ check-permissions, promote-images, tag, build-and-test-locally, trigger-custom-extensions-build-and-wait, push-to-acr-dev, push-to-acr-prod ]
     # `!failure() && !cancelled()` is required because the workflow depends on the job that can be skipped: `push-to-acr-dev` and `push-to-acr-prod`
     if: (github.ref_name == 'main' || github.ref_name == 'release' || github.ref_name == 'release-proxy' || github.ref_name == 'release-compute') && !failure() && !cancelled()
-    permissions:
-      id-token: write # aws-actions/configure-aws-credentials
-      statuses: write
-      contents: write
+
     runs-on: [ self-hosted, small ]
     container: 369495373322.dkr.ecr.eu-central-1.amazonaws.com/ansible:latest
     steps:
       - uses: actions/checkout@v4
 
-      - name: Create git tag and GitHub release
-        if: github.ref_name == 'release' || github.ref_name == 'release-proxy' || github.ref_name == 'release-compute'
-        uses: actions/github-script@v7
-        with:
-          retries: 5
-          script: |
-            const tag = "${{ needs.tag.outputs.build-tag }}";
-
-            try {
-              const existingRef = await github.rest.git.getRef({
-                owner: context.repo.owner,
-                repo: context.repo.repo,
-                ref: `tags/${tag}`,
-              });
-
-              if (existingRef.data.object.sha !== context.sha) {
-                throw new Error(`Tag ${tag} already exists but points to a different commit (expected: ${context.sha}, actual: ${existingRef.data.object.sha}).`);
-              }
-
-              console.log(`Tag ${tag} already exists and points to ${context.sha} as expected.`);
-            } catch (error) {
-              if (error.status !== 404) {
-                throw error;
-              }
-
-              console.log(`Tag ${tag} does not exist. Creating it...`);
-              await github.rest.git.createRef({
-                owner: context.repo.owner,
-                repo: context.repo.repo,
-                ref: `refs/tags/${tag}`,
-                sha: context.sha,
-              });
-              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,
-                repo: context.repo.repo,
-                tag: tag,
-              });
-
-              console.log(`Release for tag ${tag} already exists (ID: ${existingRelease.data.id}).`);
-            } catch (error) {
-              if (error.status !== 404) {
-                throw error;
-              }
-
-              console.log(`Release for tag ${tag} does not exist. Creating it...`);
-              await github.rest.repos.createRelease({
-                owner: context.repo.owner,
-                repo: context.repo.repo,
-                tag_name: tag,
-                generate_release_notes: true,
-              });
-              console.log(`Release for tag ${tag} created successfully.`);
-            }
-
       - name: Trigger deploy workflow
         env:
           GH_TOKEN: ${{ secrets.CI_ACCESS_TOKEN }}
@@ -1237,13 +1115,38 @@ jobs:
             exit 1
           fi
 
+      - name: Create git tag
+        if: github.ref_name == 'release' || github.ref_name == 'release-proxy' || github.ref_name == 'release-compute'
+        uses: actions/github-script@v7
+        with:
+          # Retry script for 5XX server errors: https://github.com/actions/github-script#retries
+          retries: 5
+          script: |
+            await github.rest.git.createRef({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              ref: "refs/tags/${{ needs.tag.outputs.build-tag }}",
+              sha: context.sha,
+            })
+
+      # TODO: check how GitHub releases looks for proxy releases and enable it if it's ok
+      - name: Create GitHub release
+        if: github.ref_name == 'release'
+        uses: actions/github-script@v7
+        with:
+          # Retry script for 5XX server errors: https://github.com/actions/github-script#retries
+          retries: 5
+          script: |
+            await github.rest.repos.createRelease({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              tag_name: "${{ needs.tag.outputs.build-tag }}",
+              generate_release_notes: true,
+            })
+
   # The job runs on `release` branch and copies compatibility data and Neon artifact from the last *release PR* to the latest directory
   promote-compatibility-data:
     needs: [ deploy ]
-    permissions:
-      id-token: write # aws-actions/configure-aws-credentials
-      statuses: write
-      contents: read
     # `!failure() && !cancelled()` is required because the workflow transitively depends on the job that can be skipped: `push-to-acr-dev` and `push-to-acr-prod`
     if: github.ref_name == 'release' && !failure() && !cancelled()
 
@@ -1280,12 +1183,6 @@ jobs:
           echo "run-id=${run_id}" | tee -a ${GITHUB_OUTPUT}
           echo "commit-sha=${last_commit_sha}" | tee -a ${GITHUB_OUTPUT}
 
-      - uses: aws-actions/configure-aws-credentials@v4
-        with:
-          aws-region: eu-central-1
-          role-to-assume: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
-          role-duration-seconds: 3600
-
       - name: Promote compatibility snapshot and Neon artifact
         env:
           BUCKET: neon-github-public-dev

.github/workflows/cloud-regress.yml

@@ -19,15 +19,14 @@ concurrency:
   group: ${{ github.workflow }}
   cancel-in-progress: true
 
-permissions:
-  id-token: write # aws-actions/configure-aws-credentials
-
 jobs:
   regress:
     env:
       POSTGRES_DISTRIB_DIR: /tmp/neon/pg_install
       TEST_OUTPUT: /tmp/test_output
       BUILD_TYPE: remote
+      AWS_ACCESS_KEY_ID: ${{ secrets.AWS_ACCESS_KEY_DEV }}
+      AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_KEY_DEV }}
     strategy:
       fail-fast: false
       matrix:
@@ -79,7 +78,6 @@ jobs:
           name: neon-${{ runner.os }}-${{ runner.arch }}-release-artifact
           path: /tmp/neon/
           prefix: latest
-          aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
 
       - name: Create a new branch
         id: create-branch
@@ -95,12 +93,10 @@ jobs:
           test_selection: cloud_regress
           pg_version: ${{matrix.pg-version}}
           extra_params: -m remote_cluster
-          aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
         env:
           BENCHMARK_CONNSTR: ${{steps.create-branch.outputs.dsn}}
 
       - name: Delete branch
-        if: always()
         uses: ./.github/actions/neon-branch-delete
         with:
           api_key: ${{ secrets.NEON_STAGING_API_KEY }}
@@ -111,14 +107,12 @@ jobs:
         id: create-allure-report
         if: ${{ !cancelled() }}
         uses: ./.github/actions/allure-report-generate
-        with:
-          aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
 
       - name: Post to a Slack channel
         if: ${{ github.event.schedule && failure() }}
         uses: slackapi/slack-github-action@v1
         with:
-          channel-id: ${{ vars.SLACK_ON_CALL_QA_STAGING_STREAM }}
+          channel-id: "C033QLM5P7D" # on-call-staging-stream
           slack-message: |
             Periodic pg_regress on staging: ${{ job.status }}
             <${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}|GitHub Run>

.github/workflows/ingest_benchmark.yml

@@ -13,7 +13,7 @@ on:
     #          │ │ │ │ ┌───────────── day of the week (0 - 6 or SUN-SAT)
     - cron:   '0 9 * * *' # run once a day, timezone is utc
   workflow_dispatch: # adds ability to run this manually
-
+    
 defaults:
   run:
     shell: bash -euxo pipefail {0}
@@ -28,7 +28,7 @@ jobs:
     strategy:
       fail-fast: false # allow other variants to continue even if one fails
       matrix:
-        target_project: [new_empty_project, large_existing_project]
+        target_project: [new_empty_project, large_existing_project]  
     permissions:
       contents: write
       statuses: write
@@ -56,7 +56,7 @@ jobs:
       with:
         aws-region: eu-central-1
         role-to-assume: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
-        role-duration-seconds: 18000 # 5 hours is currently max associated with IAM role
+        role-duration-seconds: 18000 # 5 hours is currently max associated with IAM role 
 
     - name: Download Neon artifact
       uses: ./.github/actions/download
@@ -64,7 +64,6 @@ jobs:
         name: neon-${{ runner.os }}-${{ runner.arch }}-release-artifact
         path: /tmp/neon/
         prefix: latest
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
 
     - name: Create Neon Project
       if: ${{ matrix.target_project == 'new_empty_project' }}
@@ -95,7 +94,7 @@ jobs:
         project_id: ${{ vars.BENCHMARK_INGEST_TARGET_PROJECTID }}
         api_key: ${{ secrets.NEON_STAGING_API_KEY }}
 
-    - name: Initialize Neon project
+    - name: Initialize Neon project 
       if: ${{ matrix.target_project == 'large_existing_project' }}
       env:
           BENCHMARK_INGEST_TARGET_CONNSTR: ${{ steps.create-neon-branch-ingest-target.outputs.dsn }}
@@ -123,7 +122,7 @@ jobs:
         ${PSQL} "${BENCHMARK_INGEST_TARGET_CONNSTR}" -c "CREATE EXTENSION IF NOT EXISTS neon; CREATE EXTENSION IF NOT EXISTS neon_utils;"
         echo "BENCHMARK_INGEST_TARGET_CONNSTR=${BENCHMARK_INGEST_TARGET_CONNSTR}" >> $GITHUB_ENV
 
-    - name: Invoke pgcopydb
+    - name: Invoke pgcopydb  
       uses: ./.github/actions/run-python-test-set
       with:
         build_type: remote
@@ -132,7 +131,7 @@ jobs:
         extra_params: -s -m remote_cluster --timeout 86400 -k test_ingest_performance_using_pgcopydb
         pg_version: v16
         save_perf_report: true
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
+        aws_oicd_role_arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
       env:
         BENCHMARK_INGEST_SOURCE_CONNSTR: ${{ secrets.BENCHMARK_INGEST_SOURCE_CONNSTR }}
         TARGET_PROJECT_TYPE: ${{ matrix.target_project }}
@@ -144,7 +143,7 @@ jobs:
       run: |
         export LD_LIBRARY_PATH=${PG_16_LIB_PATH}
         ${PSQL} "${BENCHMARK_INGEST_TARGET_CONNSTR}" -c "\dt+"
-
+      
     - name: Delete Neon Project
       if: ${{ always() && matrix.target_project == 'new_empty_project' }}
       uses: ./.github/actions/neon-project-delete

.github/workflows/neon_extra_builds.yml

@@ -143,10 +143,6 @@ jobs:
 
   gather-rust-build-stats:
     needs: [ check-permissions, build-build-tools-image ]
-    permissions:
-      id-token: write # aws-actions/configure-aws-credentials
-      statuses: write
-      contents: write
     if: |
       contains(github.event.pull_request.labels.*.name, 'run-extra-build-stats') ||
       contains(github.event.pull_request.labels.*.name, 'run-extra-build-*') ||
@@ -181,18 +177,13 @@ jobs:
       - name: Produce the build stats
         run: PQ_LIB_DIR=$(pwd)/pg_install/v17/lib cargo build --all --release --timings -j$(nproc)
 
-      - name: Configure AWS credentials
-        uses: aws-actions/configure-aws-credentials@v4
-        with:
-          aws-region: eu-central-1
-          role-to-assume: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
-          role-duration-seconds: 3600
-
       - name: Upload the build stats
         id: upload-stats
         env:
           BUCKET: neon-github-public-dev
           SHA: ${{ github.event.pull_request.head.sha || github.sha }}
+          AWS_ACCESS_KEY_ID: ${{ secrets.AWS_ACCESS_KEY_DEV }}
+          AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_KEY_DEV }}
         run: |
           REPORT_URL=https://${BUCKET}.s3.amazonaws.com/build-stats/${SHA}/${GITHUB_RUN_ID}/cargo-timing.html
           aws s3 cp --only-show-errors ./target/cargo-timings/cargo-timing.html "s3://${BUCKET}/build-stats/${SHA}/${GITHUB_RUN_ID}/"

.github/workflows/periodic_pagebench.yml

@@ -27,11 +27,6 @@ concurrency:
 
 jobs:
   trigger_bench_on_ec2_machine_in_eu_central_1:
-    permissions:
-      id-token: write # aws-actions/configure-aws-credentials
-      statuses: write
-      contents: write
-      pull-requests: write
     runs-on: [ self-hosted, small ]
     container:
       image: neondatabase/build-tools:pinned-bookworm
@@ -43,6 +38,8 @@ jobs:
     env:
       API_KEY: ${{ secrets.PERIODIC_PAGEBENCH_EC2_RUNNER_API_KEY }}
       RUN_ID: ${{ github.run_id }}
+      AWS_ACCESS_KEY_ID: ${{ secrets.AWS_EC2_US_TEST_RUNNER_ACCESS_KEY_ID }}
+      AWS_SECRET_ACCESS_KEY : ${{ secrets.AWS_EC2_US_TEST_RUNNER_ACCESS_KEY_SECRET }}
       AWS_DEFAULT_REGION : "eu-central-1"
       AWS_INSTANCE_ID : "i-02a59a3bf86bc7e74"
     steps:
@@ -53,13 +50,6 @@ jobs:
     - name: Show my own (github runner) external IP address - usefull for IP allowlisting
       run: curl https://ifconfig.me
 
-    - name: Assume AWS OIDC role that allows to manage (start/stop/describe... EC machine)
-      uses: aws-actions/configure-aws-credentials@v4
-      with:
-        aws-region: eu-central-1
-        role-to-assume: ${{ vars.DEV_AWS_OIDC_ROLE_MANAGE_BENCHMARK_EC2_VMS_ARN }}
-        role-duration-seconds: 3600
-
     - name: Start EC2 instance and wait for the instance to boot up
       run: |
         aws ec2 start-instances --instance-ids $AWS_INSTANCE_ID
@@ -134,10 +124,11 @@ jobs:
         cat "test_log_${GITHUB_RUN_ID}"
 
     - name: Create Allure report
+      env:
+        AWS_ACCESS_KEY_ID: ${{ secrets.AWS_ACCESS_KEY_DEV }}
+        AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_KEY_DEV }}
       if: ${{ !cancelled() }}
       uses: ./.github/actions/allure-report-generate
-      with:
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
 
     - name: Post to a Slack channel
       if: ${{ github.event.schedule && failure() }}
@@ -157,14 +148,6 @@ jobs:
         -H "Authorization: Bearer $API_KEY" \
         -d ''
 
-    - name: Assume AWS OIDC role that allows to manage (start/stop/describe... EC machine)
-      if: always() && steps.poll_step.outputs.too_many_runs != 'true'
-      uses: aws-actions/configure-aws-credentials@v4
-      with:
-        aws-region: eu-central-1
-        role-to-assume: ${{ vars.DEV_AWS_OIDC_ROLE_MANAGE_BENCHMARK_EC2_VMS_ARN }}
-        role-duration-seconds: 3600
-
     - name: Stop EC2 instance and wait for the instance to be stopped
       if: always() && steps.poll_step.outputs.too_many_runs != 'true'
       run: |

.github/workflows/pg-clients.yml

@@ -25,13 +25,11 @@ defaults:
   run:
     shell: bash -euxo pipefail {0}
 
-permissions:
-  id-token: write # aws-actions/configure-aws-credentials
-  statuses: write # require for posting a status update
-
 env:
   DEFAULT_PG_VERSION: 16
   PLATFORM: neon-captest-new
+  AWS_ACCESS_KEY_ID: ${{ secrets.AWS_ACCESS_KEY_DEV }}
+  AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_KEY_DEV }}
   AWS_DEFAULT_REGION: eu-central-1
 
 jobs:
@@ -96,7 +94,6 @@ jobs:
           name: neon-${{ runner.os }}-${{ runner.arch }}-release-artifact
           path: /tmp/neon/
           prefix: latest
-          aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
 
       - name: Create Neon Project
         id: create-neon-project
@@ -113,7 +110,6 @@ jobs:
           run_in_parallel: false
           extra_params: -m remote_cluster
           pg_version: ${{ env.DEFAULT_PG_VERSION }}
-          aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
         env:
           BENCHMARK_CONNSTR: ${{ steps.create-neon-project.outputs.dsn }}
 
@@ -130,7 +126,6 @@ jobs:
         uses: ./.github/actions/allure-report-generate
         with:
           store-test-results-into-db: true
-          aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
         env:
           REGRESS_TEST_RESULT_CONNSTR_NEW: ${{ secrets.REGRESS_TEST_RESULT_CONNSTR_NEW }}
 
@@ -164,7 +159,6 @@ jobs:
         name: neon-${{ runner.os }}-${{ runner.arch }}-release-artifact
         path: /tmp/neon/
         prefix: latest
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
 
     - name: Create Neon Project
       id: create-neon-project
@@ -181,7 +175,6 @@ jobs:
         run_in_parallel: false
         extra_params: -m remote_cluster
         pg_version: ${{ env.DEFAULT_PG_VERSION }}
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
       env:
         BENCHMARK_CONNSTR: ${{ steps.create-neon-project.outputs.dsn }}
 
@@ -198,7 +191,6 @@ jobs:
       uses: ./.github/actions/allure-report-generate
       with:
         store-test-results-into-db: true
-        aws-oicd-role-arn: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
       env:
         REGRESS_TEST_RESULT_CONNSTR_NEW: ${{ secrets.REGRESS_TEST_RESULT_CONNSTR_NEW }}
 

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

@@ -67,7 +67,7 @@ jobs:
     runs-on: ubuntu-22.04
 
     permissions:
-      id-token: write # for `azure/login` and aws auth
+      id-token: write # for `azure/login`
 
     steps:
       - uses: docker/login-action@v3
@@ -75,15 +75,11 @@ jobs:
           username: ${{ secrets.NEON_DOCKERHUB_USERNAME }}
           password: ${{ secrets.NEON_DOCKERHUB_PASSWORD }}
 
-      - name: Configure AWS credentials
-        uses: aws-actions/configure-aws-credentials@v4
+      - uses: docker/login-action@v3
         with:
-          aws-region: eu-central-1
-          role-to-assume: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
-          role-duration-seconds: 3600
-
-      - name: Login to Amazon Dev ECR
-        uses: aws-actions/amazon-ecr-login@v2
+          registry: 369495373322.dkr.ecr.eu-central-1.amazonaws.com
+          username: ${{ secrets.AWS_ACCESS_KEY_DEV }}
+          password: ${{ secrets.AWS_SECRET_KEY_DEV }}
 
       - name: Azure login
         uses: azure/login@6c251865b4e6290e7b78be643ea2d005bc51f69a  # @v2.1.1

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

@@ -63,7 +63,6 @@ jobs:
     if: always()
     permissions:
       statuses: write # for `github.repos.createCommitStatus(...)`
-      contents: write
     needs:
       - get-changed-files
       - check-codestyle-python

.github/workflows/release.yml

@@ -3,7 +3,7 @@ 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 * * MON' # Storage release
     - cron: '0 6 * * THU' # Proxy release
   workflow_dispatch:
     inputs:
@@ -29,7 +29,7 @@ defaults:
 
 jobs:
   create-storage-release-branch:
-    if: ${{ github.event.schedule == '0 6 * * FRI' || inputs.create-storage-release-branch }}
+    if: ${{ github.event.schedule == '0 6 * * MON' || inputs.create-storage-release-branch }}
 
     permissions:
       contents: write

Cargo.lock

@@ -10,9 +10,9 @@ checksum = "8b5ace29ee3216de37c0546865ad08edef58b0f9e76838ed8959a84a990e58c5"
 
 [[package]]
 name = "addr2line"
-version = "0.24.2"
+version = "0.21.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "dfbe277e56a376000877090da837660b4427aad530e3028d44e0bffe4f89a1c1"
+checksum = "8a30b2e23b9e17a9f90641c7ab1549cd9b44f296d3ccbf309d2863cfe398a0cb"
 dependencies = [
  "gimli",
 ]
@@ -23,12 +23,6 @@ version = "1.0.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "f26201604c87b1e01bd3d98f8d5d9a8fcbb815e8cedb41ffccbeb4bf593a35fe"
 
-[[package]]
-name = "adler2"
-version = "2.0.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "512761e0bb2578dd7380c6baaa0f4ce03e84f95e960231d1dec8bf4d7d6e2627"
-
 [[package]]
 name = "ahash"
 version = "0.8.11"
@@ -777,7 +771,7 @@ dependencies = [
 [[package]]
 name = "azure_core"
 version = "0.21.0"
-source = "git+https://github.com/neondatabase/azure-sdk-for-rust.git?branch=neon#66e77bdd87bf87e773acf3b0c84b532c1124367d"
+source = "git+https://github.com/neondatabase/azure-sdk-for-rust.git?branch=arpad%2Fblob_batch#e34e9d277391a394dce6bd016db8d50991462f72"
 dependencies = [
  "async-trait",
  "base64 0.22.1",
@@ -806,7 +800,7 @@ dependencies = [
 [[package]]
 name = "azure_identity"
 version = "0.21.0"
-source = "git+https://github.com/neondatabase/azure-sdk-for-rust.git?branch=neon#66e77bdd87bf87e773acf3b0c84b532c1124367d"
+source = "git+https://github.com/neondatabase/azure-sdk-for-rust.git?branch=arpad%2Fblob_batch#e34e9d277391a394dce6bd016db8d50991462f72"
 dependencies = [
  "async-lock",
  "async-trait",
@@ -825,7 +819,7 @@ dependencies = [
 [[package]]
 name = "azure_storage"
 version = "0.21.0"
-source = "git+https://github.com/neondatabase/azure-sdk-for-rust.git?branch=neon#66e77bdd87bf87e773acf3b0c84b532c1124367d"
+source = "git+https://github.com/neondatabase/azure-sdk-for-rust.git?branch=arpad%2Fblob_batch#e34e9d277391a394dce6bd016db8d50991462f72"
 dependencies = [
  "RustyXML",
  "async-lock",
@@ -843,7 +837,7 @@ dependencies = [
 [[package]]
 name = "azure_storage_blobs"
 version = "0.21.0"
-source = "git+https://github.com/neondatabase/azure-sdk-for-rust.git?branch=neon#66e77bdd87bf87e773acf3b0c84b532c1124367d"
+source = "git+https://github.com/neondatabase/azure-sdk-for-rust.git?branch=arpad%2Fblob_batch#e34e9d277391a394dce6bd016db8d50991462f72"
 dependencies = [
  "RustyXML",
  "azure_core",
@@ -863,7 +857,7 @@ dependencies = [
 [[package]]
 name = "azure_svc_blobstorage"
 version = "0.21.0"
-source = "git+https://github.com/neondatabase/azure-sdk-for-rust.git?branch=neon#66e77bdd87bf87e773acf3b0c84b532c1124367d"
+source = "git+https://github.com/neondatabase/azure-sdk-for-rust.git?branch=arpad%2Fblob_batch#e34e9d277391a394dce6bd016db8d50991462f72"
 dependencies = [
  "azure_core",
  "bytes",
@@ -877,17 +871,17 @@ dependencies = [
 
 [[package]]
 name = "backtrace"
-version = "0.3.74"
+version = "0.3.69"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "8d82cb332cdfaed17ae235a638438ac4d4839913cc2af585c3c6746e8f8bee1a"
+checksum = "2089b7e3f35b9dd2d0ed921ead4f6d318c27680d4a5bd167b3ee120edb105837"
 dependencies = [
  "addr2line",
+ "cc",
  "cfg-if",
  "libc",
- "miniz_oxide 0.8.0",
+ "miniz_oxide",
  "object",
  "rustc-demangle",
- "windows-targets 0.52.6",
 ]
 
 [[package]]
@@ -1133,7 +1127,7 @@ dependencies = [
  "num-traits",
  "serde",
  "wasm-bindgen",
- "windows-targets 0.52.6",
+ "windows-targets 0.52.4",
 ]
 
 [[package]]
@@ -2113,7 +2107,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "3b9429470923de8e8cbd4d2dc513535400b4b3fef0319fb5c4e1f520a7bef743"
 dependencies = [
  "crc32fast",
- "miniz_oxide 0.7.1",
+ "miniz_oxide",
 ]
 
 [[package]]
@@ -2314,9 +2308,9 @@ dependencies = [
 
 [[package]]
 name = "gimli"
-version = "0.31.1"
+version = "0.28.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "07e28edb80900c19c28f1072f2e8aeca7fa06b23cd4169cefe1af5aa3260783f"
+checksum = "4271d37baee1b8c7e4b708028c57d816cf9d2434acb33a549475f78c181f6253"
 
 [[package]]
 name = "git-version"
@@ -3410,15 +3404,6 @@ dependencies = [
  "adler",
 ]
 
-[[package]]
-name = "miniz_oxide"
-version = "0.8.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "e2d80299ef12ff69b16a84bb182e3b9df68b5a91574d3d4fa6e41b65deec4df1"
-dependencies = [
- "adler2",
-]
-
 [[package]]
 name = "mio"
 version = "0.8.11"
@@ -3653,9 +3638,9 @@ dependencies = [
 
 [[package]]
 name = "object"
-version = "0.36.5"
+version = "0.32.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "aedf0a2d09c573ed1d8d85b30c119153926a2b36dce0ab28322c09a117a4683e"
+checksum = "a6a622008b6e321afc04970976f62ee297fdbaa6f95318ca343e3eebb9648441"
 dependencies = [
  "memchr",
 ]
@@ -4416,13 +4401,11 @@ dependencies = [
  "bindgen",
  "bytes",
  "crc32c",
- "criterion",
  "env_logger",
  "log",
  "memoffset 0.9.0",
  "once_cell",
  "postgres",
- "pprof",
  "regex",
  "serde",
  "thiserror",
@@ -5079,7 +5062,6 @@ dependencies = [
  "once_cell",
  "pin-project-lite",
  "rand 0.8.5",
- "reqwest",
  "scopeguard",
  "serde",
  "serde_json",
@@ -5338,9 +5320,9 @@ dependencies = [
 
 [[package]]
 name = "rustc-demangle"
-version = "0.1.24"
+version = "0.1.23"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "719b953e2095829ee67db738b3bfa9fa368c94900df327b3f07fe6e794d2fe1f"
+checksum = "d626bb9dae77e28219937af045c257c28bfd3f69333c512553507f5f9798cb76"
 
 [[package]]
 name = "rustc-hash"
@@ -5553,7 +5535,6 @@ dependencies = [
  "remote_storage",
  "reqwest",
  "safekeeper_api",
- "safekeeper_client",
  "scopeguard",
  "sd-notify",
  "serde",
@@ -5584,25 +5565,10 @@ name = "safekeeper_api"
 version = "0.1.0"
 dependencies = [
  "const_format",
- "postgres_ffi",
- "pq_proto",
  "serde",
- "tokio",
  "utils",
 ]
 
-[[package]]
-name = "safekeeper_client"
-version = "0.1.0"
-dependencies = [
- "reqwest",
- "safekeeper_api",
- "serde",
- "thiserror",
- "utils",
- "workspace_hack",
-]
-
 [[package]]
 name = "same-file"
 version = "1.0.6"
@@ -7234,7 +7200,6 @@ dependencies = [
  "anyhow",
  "arc-swap",
  "async-compression",
- "backtrace",
  "bincode",
  "byteorder",
  "bytes",
@@ -7245,14 +7210,12 @@ dependencies = [
  "criterion",
  "diatomic-waker",
  "fail",
- "flate2",
  "futures",
  "git-version",
  "hex",
  "hex-literal",
  "humantime",
  "hyper 0.14.30",
- "itertools 0.10.5",
  "jemalloc_pprof",
  "jsonwebtoken",
  "metrics",
@@ -7609,7 +7572,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "e48a53791691ab099e5e2ad123536d0fff50652600abaf43bbf952894110d0be"
 dependencies = [
  "windows-core",
- "windows-targets 0.52.6",
+ "windows-targets 0.52.4",
 ]
 
 [[package]]
@@ -7618,7 +7581,7 @@ version = "0.52.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "33ab640c8d7e35bf8ba19b884ba838ceb4fba93a4e8c65a9059d08afcfc683d9"
 dependencies = [
- "windows-targets 0.52.6",
+ "windows-targets 0.52.4",
 ]
 
 [[package]]
@@ -7636,7 +7599,7 @@ version = "0.52.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "282be5f36a8ce781fad8c8ae18fa3f9beff57ec1b52cb3de0789201425d9a33d"
 dependencies = [
- "windows-targets 0.52.6",
+ "windows-targets 0.52.4",
 ]
 
 [[package]]
@@ -7656,18 +7619,17 @@ dependencies = [
 
 [[package]]
 name = "windows-targets"
-version = "0.52.6"
+version = "0.52.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "9b724f72796e036ab90c1021d4780d4d3d648aca59e491e6b98e725b84e99973"
+checksum = "7dd37b7e5ab9018759f893a1952c9420d060016fc19a472b4bb20d1bdd694d1b"
 dependencies = [
- "windows_aarch64_gnullvm 0.52.6",
- "windows_aarch64_msvc 0.52.6",
- "windows_i686_gnu 0.52.6",
- "windows_i686_gnullvm",
- "windows_i686_msvc 0.52.6",
- "windows_x86_64_gnu 0.52.6",
- "windows_x86_64_gnullvm 0.52.6",
- "windows_x86_64_msvc 0.52.6",
+ "windows_aarch64_gnullvm 0.52.4",
+ "windows_aarch64_msvc 0.52.4",
+ "windows_i686_gnu 0.52.4",
+ "windows_i686_msvc 0.52.4",
+ "windows_x86_64_gnu 0.52.4",
+ "windows_x86_64_gnullvm 0.52.4",
+ "windows_x86_64_msvc 0.52.4",
 ]
 
 [[package]]
@@ -7678,9 +7640,9 @@ checksum = "91ae572e1b79dba883e0d315474df7305d12f569b400fcf90581b06062f7e1bc"
 
 [[package]]
 name = "windows_aarch64_gnullvm"
-version = "0.52.6"
+version = "0.52.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "32a4622180e7a0ec044bb555404c800bc9fd9ec262ec147edd5989ccd0c02cd3"
+checksum = "bcf46cf4c365c6f2d1cc93ce535f2c8b244591df96ceee75d8e83deb70a9cac9"
 
 [[package]]
 name = "windows_aarch64_msvc"
@@ -7690,9 +7652,9 @@ checksum = "b2ef27e0d7bdfcfc7b868b317c1d32c641a6fe4629c171b8928c7b08d98d7cf3"
 
 [[package]]
 name = "windows_aarch64_msvc"
-version = "0.52.6"
+version = "0.52.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "09ec2a7bb152e2252b53fa7803150007879548bc709c039df7627cabbd05d469"
+checksum = "da9f259dd3bcf6990b55bffd094c4f7235817ba4ceebde8e6d11cd0c5633b675"
 
 [[package]]
 name = "windows_i686_gnu"
@@ -7702,15 +7664,9 @@ checksum = "622a1962a7db830d6fd0a69683c80a18fda201879f0f447f065a3b7467daa241"
 
 [[package]]
 name = "windows_i686_gnu"
-version = "0.52.6"
+version = "0.52.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "8e9b5ad5ab802e97eb8e295ac6720e509ee4c243f69d781394014ebfe8bbfa0b"
-
-[[package]]
-name = "windows_i686_gnullvm"
-version = "0.52.6"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "0eee52d38c090b3caa76c563b86c3a4bd71ef1a819287c19d586d7334ae8ed66"
+checksum = "b474d8268f99e0995f25b9f095bc7434632601028cf86590aea5c8a5cb7801d3"
 
 [[package]]
 name = "windows_i686_msvc"
@@ -7720,9 +7676,9 @@ checksum = "4542c6e364ce21bf45d69fdd2a8e455fa38d316158cfd43b3ac1c5b1b19f8e00"
 
 [[package]]
 name = "windows_i686_msvc"
-version = "0.52.6"
+version = "0.52.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "240948bc05c5e7c6dabba28bf89d89ffce3e303022809e73deaefe4f6ec56c66"
+checksum = "1515e9a29e5bed743cb4415a9ecf5dfca648ce85ee42e15873c3cd8610ff8e02"
 
 [[package]]
 name = "windows_x86_64_gnu"
@@ -7732,9 +7688,9 @@ checksum = "ca2b8a661f7628cbd23440e50b05d705db3686f894fc9580820623656af974b1"
 
 [[package]]
 name = "windows_x86_64_gnu"
-version = "0.52.6"
+version = "0.52.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "147a5c80aabfbf0c7d901cb5895d1de30ef2907eb21fbbab29ca94c5b08b1a78"
+checksum = "5eee091590e89cc02ad514ffe3ead9eb6b660aedca2183455434b93546371a03"
 
 [[package]]
 name = "windows_x86_64_gnullvm"
@@ -7744,9 +7700,9 @@ checksum = "7896dbc1f41e08872e9d5e8f8baa8fdd2677f29468c4e156210174edc7f7b953"
 
 [[package]]
 name = "windows_x86_64_gnullvm"
-version = "0.52.6"
+version = "0.52.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "24d5b23dc417412679681396f2b49f3de8c1473deb516bd34410872eff51ed0d"
+checksum = "77ca79f2451b49fa9e2af39f0747fe999fcda4f5e241b2898624dca97a1f2177"
 
 [[package]]
 name = "windows_x86_64_msvc"
@@ -7756,9 +7712,9 @@ checksum = "1a515f5799fe4961cb532f983ce2b23082366b898e52ffbce459c86f67c8378a"
 
 [[package]]
 name = "windows_x86_64_msvc"
-version = "0.52.6"
+version = "0.52.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "589f6da84c646204747d1270a2a5661ea66ed1cced2631d546fdfb155959f9ec"
+checksum = "32b752e52a2da0ddfbdbcc6fceadfeede4c939ed16d13e648833a61dfb611ed8"
 
 [[package]]
 name = "winnow"

Cargo.toml

@@ -11,7 +11,6 @@ members = [
     "pageserver/pagebench",
     "proxy",
     "safekeeper",
-    "safekeeper/client",
     "storage_broker",
     "storage_controller",
     "storage_controller/client",
@@ -52,7 +51,6 @@ anyhow = { version = "1.0", features = ["backtrace"] }
 arc-swap = "1.6"
 async-compression = { version = "0.4.0", features = ["tokio", "gzip", "zstd"] }
 atomic-take = "1.1.0"
-backtrace = "0.3.74"
 flate2 = "1.0.26"
 async-stream = "0.3"
 async-trait = "0.1"
@@ -215,10 +213,10 @@ postgres-types = { git = "https://github.com/neondatabase/rust-postgres.git", br
 tokio-postgres = { git = "https://github.com/neondatabase/rust-postgres.git", branch = "neon" }
 
 ## Azure SDK crates
-azure_core = { git = "https://github.com/neondatabase/azure-sdk-for-rust.git", branch = "neon", default-features = false, features = ["enable_reqwest_rustls", "hmac_rust"] }
-azure_identity = { git = "https://github.com/neondatabase/azure-sdk-for-rust.git", branch = "neon", default-features = false, features = ["enable_reqwest_rustls"] }
-azure_storage = { git = "https://github.com/neondatabase/azure-sdk-for-rust.git", branch = "neon", default-features = false, features = ["enable_reqwest_rustls"] }
-azure_storage_blobs = { git = "https://github.com/neondatabase/azure-sdk-for-rust.git", branch = "neon", default-features = false, features = ["enable_reqwest_rustls"] }
+azure_core = { git = "https://github.com/neondatabase/azure-sdk-for-rust.git", branch = "arpad/blob_batch", default-features = false, features = ["enable_reqwest_rustls", "hmac_rust"] }
+azure_identity = { git = "https://github.com/neondatabase/azure-sdk-for-rust.git", branch = "arpad/blob_batch", default-features = false, features = ["enable_reqwest_rustls"] }
+azure_storage = { git = "https://github.com/neondatabase/azure-sdk-for-rust.git", branch = "arpad/blob_batch", default-features = false, features = ["enable_reqwest_rustls"] }
+azure_storage_blobs = { git = "https://github.com/neondatabase/azure-sdk-for-rust.git", branch = "arpad/blob_batch", default-features = false, features = ["enable_reqwest_rustls"] }
 
 ## Local libraries
 compute_api = { version = "0.1", path = "./libs/compute_api/" }
@@ -235,7 +233,6 @@ postgres_initdb = { path = "./libs/postgres_initdb" }
 pq_proto = { version = "0.1", path = "./libs/pq_proto/" }
 remote_storage = { version = "0.1", path = "./libs/remote_storage/" }
 safekeeper_api = { version = "0.1", path = "./libs/safekeeper_api" }
-safekeeper_client = { path = "./safekeeper/client" }
 desim = { version = "0.1", path = "./libs/desim" }
 storage_broker = { version = "0.1", path = "./storage_broker/" } # Note: main broker code is inside the binary crate, so linking with the library shouldn't be heavy.
 storage_controller_client = { path = "./storage_controller/client" }

compute/compute-node.Dockerfile

@@ -35,12 +35,10 @@ RUN case $DEBIAN_VERSION in \
       ;; \
     esac && \
     apt update &&  \
-    apt install --no-install-recommends --no-install-suggests -y \
-    ninja-build git autoconf automake libtool build-essential bison flex libreadline-dev \
+    apt install --no-install-recommends -y git autoconf automake libtool build-essential bison flex libreadline-dev \
     zlib1g-dev libxml2-dev libcurl4-openssl-dev libossp-uuid-dev wget ca-certificates pkg-config libssl-dev \
     libicu-dev libxslt1-dev liblz4-dev libzstd-dev zstd \
-    $VERSION_INSTALLS \
-    && apt clean && rm -rf /var/lib/apt/lists/*
+    $VERSION_INSTALLS
 
 #########################################################################################
 #
@@ -115,12 +113,10 @@ ARG DEBIAN_VERSION
 ARG PG_VERSION
 COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
 RUN apt update && \
-    apt install --no-install-recommends --no-install-suggests -y \
-    gdal-bin libboost-dev libboost-thread-dev libboost-filesystem-dev \
+    apt install --no-install-recommends -y gdal-bin libboost-dev libboost-thread-dev libboost-filesystem-dev \
     libboost-system-dev libboost-iostreams-dev libboost-program-options-dev libboost-timer-dev \
     libcgal-dev libgdal-dev libgmp-dev libmpfr-dev libopenscenegraph-dev libprotobuf-c-dev \
-    protobuf-c-compiler xsltproc \
-    && apt clean && rm -rf /var/lib/apt/lists/*
+    protobuf-c-compiler xsltproc
 
 
 # Postgis 3.5.0 requires SFCGAL 1.4+
@@ -147,9 +143,9 @@ RUN case "${DEBIAN_VERSION}" in \
     wget https://gitlab.com/sfcgal/SFCGAL/-/archive/v${SFCGAL_VERSION}/SFCGAL-v${SFCGAL_VERSION}.tar.gz -O SFCGAL.tar.gz && \
     echo "${SFCGAL_CHECKSUM} SFCGAL.tar.gz" | sha256sum --check && \
     mkdir sfcgal-src && cd sfcgal-src && tar xzf ../SFCGAL.tar.gz --strip-components=1 -C . && \
-    cmake -DCMAKE_BUILD_TYPE=Release -GNinja . && ninja -j $(getconf _NPROCESSORS_ONLN) && \
-    DESTDIR=/sfcgal ninja install -j $(getconf _NPROCESSORS_ONLN) && \
-    ninja clean && cp -R /sfcgal/* /
+    cmake -DCMAKE_BUILD_TYPE=Release . && make -j $(getconf _NPROCESSORS_ONLN) && \
+    DESTDIR=/sfcgal make install -j $(getconf _NPROCESSORS_ONLN) && \
+    make clean && cp -R /sfcgal/* /
 
 ENV PATH="/usr/local/pgsql/bin:$PATH"
 
@@ -217,9 +213,9 @@ RUN case "${PG_VERSION}" in \
     echo "${PGROUTING_CHECKSUM} pgrouting.tar.gz" | sha256sum --check && \
     mkdir pgrouting-src && cd pgrouting-src && tar xzf ../pgrouting.tar.gz --strip-components=1 -C . && \
     mkdir build && cd build && \
-    cmake -GNinja -DCMAKE_BUILD_TYPE=Release .. && \
-    ninja -j $(getconf _NPROCESSORS_ONLN) && \
-    ninja -j $(getconf _NPROCESSORS_ONLN) install && \
+    cmake -DCMAKE_BUILD_TYPE=Release .. && \
+    make -j $(getconf _NPROCESSORS_ONLN) && \
+    make -j $(getconf _NPROCESSORS_ONLN) install && \
     echo 'trusted = true' >> /usr/local/pgsql/share/extension/pgrouting.control && \
     find /usr/local/pgsql -type f | sed 's|^/usr/local/pgsql/||' > /after.txt &&\
     cp /usr/local/pgsql/share/extension/pgrouting.control /extensions/postgis && \
@@ -239,9 +235,7 @@ COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
 COPY compute/patches/plv8-3.1.10.patch /plv8-3.1.10.patch
 
 RUN apt update && \
-    apt install --no-install-recommends --no-install-suggests -y \
-    ninja-build python3-dev libncurses5 binutils clang \
-    && apt clean && rm -rf /var/lib/apt/lists/*
+    apt install --no-install-recommends -y ninja-build python3-dev libncurses5 binutils clang
 
 # plv8 3.2.3 supports v17
 # last release v3.2.3 - Sep 7, 2024
@@ -307,10 +301,9 @@ RUN mkdir -p /h3/usr/ && \
     echo "ec99f1f5974846bde64f4513cf8d2ea1b8d172d2218ab41803bf6a63532272bc h3.tar.gz" | sha256sum --check && \
     mkdir h3-src && cd h3-src && tar xzf ../h3.tar.gz --strip-components=1 -C . && \
     mkdir build && cd build && \
-    cmake .. -GNinja -DBUILD_BENCHMARKS=0 -DCMAKE_BUILD_TYPE=Release \
-        -DBUILD_FUZZERS=0 -DBUILD_FILTERS=0 -DBUILD_GENERATORS=0 -DBUILD_TESTING=0 \
-    && ninja -j $(getconf _NPROCESSORS_ONLN) && \
-    DESTDIR=/h3 ninja install && \
+    cmake .. -DCMAKE_BUILD_TYPE=Release && \
+    make -j $(getconf _NPROCESSORS_ONLN) && \
+    DESTDIR=/h3 make install && \
     cp -R /h3/usr / && \
     rm -rf build
 
@@ -657,15 +650,14 @@ FROM build-deps AS rdkit-pg-build
 ARG PG_VERSION
 COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
 
-RUN apt update && \
-    apt install --no-install-recommends --no-install-suggests -y \
+RUN apt-get update && \
+    apt-get install --no-install-recommends -y \
         libboost-iostreams1.74-dev \
         libboost-regex1.74-dev \
         libboost-serialization1.74-dev \
         libboost-system1.74-dev \
         libeigen3-dev \
-        libboost-all-dev \
-    && apt clean && rm -rf /var/lib/apt/lists/*
+        libboost-all-dev
 
 # rdkit Release_2024_09_1 supports v17
 # last release Release_2024_09_1 - Sep 27, 2024
@@ -701,8 +693,6 @@ RUN case "${PG_VERSION}" in \
         -D RDK_BUILD_MOLINTERCHANGE_SUPPORT=OFF \
         -D RDK_BUILD_YAEHMOP_SUPPORT=OFF \
         -D RDK_BUILD_STRUCTCHECKER_SUPPORT=OFF \
-        -D RDK_TEST_MULTITHREADED=OFF \
-        -D RDK_BUILD_CPP_TESTS=OFF \
         -D RDK_USE_URF=OFF \
         -D RDK_BUILD_PGSQL=ON \
         -D RDK_PGSQL_STATIC=ON \
@@ -714,10 +704,9 @@ RUN case "${PG_VERSION}" in \
         -D RDK_INSTALL_COMIC_FONTS=OFF \
         -D RDK_BUILD_FREETYPE_SUPPORT=OFF \
         -D CMAKE_BUILD_TYPE=Release \
-        -GNinja \
         . && \
-    ninja -j $(getconf _NPROCESSORS_ONLN) && \
-    ninja -j $(getconf _NPROCESSORS_ONLN) install && \
+    make -j $(getconf _NPROCESSORS_ONLN) && \
+    make -j $(getconf _NPROCESSORS_ONLN) install && \
     echo 'trusted = true' >> /usr/local/pgsql/share/extension/rdkit.control
 
 #########################################################################################
@@ -860,9 +849,8 @@ FROM build-deps AS rust-extensions-build
 ARG PG_VERSION
 COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
 
-RUN apt update && \
-    apt install --no-install-recommends --no-install-suggests -y curl libclang-dev && \
-    apt clean && rm -rf /var/lib/apt/lists/* && \
+RUN apt-get update && \
+    apt-get install --no-install-recommends -y curl libclang-dev && \
     useradd -ms /bin/bash nonroot -b /home
 
 ENV HOME=/home/nonroot
@@ -897,9 +885,8 @@ FROM build-deps AS rust-extensions-build-pgrx12
 ARG PG_VERSION
 COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
 
-RUN apt update && \
-    apt install --no-install-recommends --no-install-suggests -y curl libclang-dev && \
-    apt clean && rm -rf /var/lib/apt/lists/* && \
+RUN apt-get update && \
+    apt-get install --no-install-recommends -y curl libclang-dev && \
     useradd -ms /bin/bash nonroot -b /home
 
 ENV HOME=/home/nonroot
@@ -927,22 +914,18 @@ FROM rust-extensions-build-pgrx12 AS pg-onnx-build
 
 # cmake 3.26 or higher is required, so installing it using pip (bullseye-backports has cmake 3.25).
 # Install it using virtual environment, because Python 3.11 (the default version on Debian 12 (Bookworm)) complains otherwise
-RUN apt update && apt install --no-install-recommends --no-install-suggests -y \
-    python3 python3-pip python3-venv && \
-    apt clean && rm -rf /var/lib/apt/lists/* && \
+RUN apt-get update && apt-get install -y python3 python3-pip python3-venv && \
     python3 -m venv venv && \
     . venv/bin/activate && \
     python3 -m pip install cmake==3.30.5 && \
     wget https://github.com/microsoft/onnxruntime/archive/refs/tags/v1.18.1.tar.gz -O onnxruntime.tar.gz && \
     mkdir onnxruntime-src && cd onnxruntime-src && tar xzf ../onnxruntime.tar.gz --strip-components=1 -C . && \
-    ./build.sh --config Release --parallel --cmake_generator Ninja \
-    --skip_submodule_sync --skip_tests --allow_running_as_root
+    ./build.sh --config Release --parallel --skip_submodule_sync --skip_tests --allow_running_as_root
 
 
 FROM pg-onnx-build AS pgrag-pg-build
 
-RUN apt update && apt install --no-install-recommends --no-install-suggests -y protobuf-compiler \
-    && apt clean && rm -rf /var/lib/apt/lists/* && \
+RUN apt-get install -y protobuf-compiler && \
     wget https://github.com/neondatabase-labs/pgrag/archive/refs/tags/v0.0.0.tar.gz -O pgrag.tar.gz &&  \
     echo "2cbe394c1e74fc8bcad9b52d5fbbfb783aef834ca3ce44626cfd770573700bb4 pgrag.tar.gz" | sha256sum --check && \
     mkdir pgrag-src && cd pgrag-src && tar xzf ../pgrag.tar.gz --strip-components=1 -C . && \
@@ -1185,25 +1168,6 @@ RUN case "${PG_VERSION}" in \
     make BUILD_TYPE=release -j $(getconf _NPROCESSORS_ONLN) install && \
     echo 'trusted = true' >> /usr/local/pgsql/share/extension/pg_mooncake.control
 
-#########################################################################################
-#
-# Layer "pg_repack"
-# compile pg_repack extension
-#
-#########################################################################################
-
-FROM build-deps AS pg-repack-build
-ARG PG_VERSION
-COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
-
-ENV PATH="/usr/local/pgsql/bin/:$PATH"
-
-RUN wget https://github.com/reorg/pg_repack/archive/refs/tags/ver_1.5.2.tar.gz -O pg_repack.tar.gz && \
-    echo '4516cad42251ed3ad53ff619733004db47d5755acac83f75924cd94d1c4fb681 pg_repack.tar.gz' | sha256sum --check && \
-    mkdir pg_repack-src && cd pg_repack-src && tar xzf ../pg_repack.tar.gz --strip-components=1 -C . && \
-    make -j $(getconf _NPROCESSORS_ONLN) && \
-    make -j $(getconf _NPROCESSORS_ONLN) install
-
 #########################################################################################
 #
 # Layer "neon-pg-ext-build"
@@ -1249,7 +1213,6 @@ COPY --from=pg-anon-pg-build /usr/local/pgsql/ /usr/local/pgsql/
 COPY --from=pg-ivm-build /usr/local/pgsql/ /usr/local/pgsql/
 COPY --from=pg-partman-build /usr/local/pgsql/ /usr/local/pgsql/
 COPY --from=pg-mooncake-build /usr/local/pgsql/ /usr/local/pgsql/
-COPY --from=pg-repack-build /usr/local/pgsql/ /usr/local/pgsql/
 COPY pgxn/ pgxn/
 
 RUN make -j $(getconf _NPROCESSORS_ONLN) \
@@ -1316,8 +1279,8 @@ COPY --from=compute-tools /home/nonroot/target/release-line-debug-size-lto/fast_
 
 FROM debian:$DEBIAN_FLAVOR AS pgbouncer
 RUN set -e \
-    && apt update \
-    && apt install --no-install-suggests --no-install-recommends -y \
+    && apt-get update \
+    && apt-get install --no-install-recommends -y \
         build-essential \
         git \
         ca-certificates \
@@ -1325,8 +1288,7 @@ RUN set -e \
         automake \
         libevent-dev \
         libtool \
-        pkg-config \
-    && apt clean && rm -rf /var/lib/apt/lists/*
+        pkg-config
 
 # Use `dist_man_MANS=` to skip manpage generation (which requires python3/pandoc)
 ENV PGBOUNCER_TAG=pgbouncer_1_22_1
@@ -1557,7 +1519,7 @@ RUN apt update && \
         procps \
         ca-certificates \
         $VERSION_INSTALLS && \
-    apt clean && rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/* && \
+    rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/* && \
     localedef -i en_US -c -f UTF-8 -A /usr/share/locale/locale.alias en_US.UTF-8
 
 # s5cmd 2.2.2 from https://github.com/peak/s5cmd/releases/tag/v2.2.2

compute/etc/neon_collector.jsonnet

@@ -3,7 +3,7 @@
   metrics: [
     import 'sql_exporter/checkpoints_req.libsonnet',
     import 'sql_exporter/checkpoints_timed.libsonnet',
-    import 'sql_exporter/compute_backpressure_throttling_seconds_total.libsonnet',
+    import 'sql_exporter/compute_backpressure_throttling_seconds.libsonnet',
     import 'sql_exporter/compute_current_lsn.libsonnet',
     import 'sql_exporter/compute_logical_snapshot_files.libsonnet',
     import 'sql_exporter/compute_logical_snapshots_bytes.libsonnet',

compute/etc/pgbouncer.ini

@@ -19,10 +19,3 @@ max_prepared_statements=0
 admin_users=postgres
 unix_socket_dir=/tmp/
 unix_socket_mode=0777
-
-;; 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
-;; the past where the logging significantly stressed the log device or pgbouncer
-;; itself.
-log_connections=0
-log_disconnections=0

compute/etc/sql_exporter/compute_backpressure_throttling_seconds.libsonnet

@@ -1,10 +1,10 @@
 {
-  metric_name: 'compute_backpressure_throttling_seconds_total',
-  type: 'counter',
+  metric_name: 'compute_backpressure_throttling_seconds',
+  type: 'gauge',
   help: 'Time compute has spent throttled',
   key_labels: null,
   values: [
     'throttled',
   ],
-  query: importstr 'sql_exporter/compute_backpressure_throttling_seconds_total.sql',
+  query: importstr 'sql_exporter/compute_backpressure_throttling_seconds.sql',
 }

compute/patches/cloud_regress_pg16.patch

@@ -981,7 +981,7 @@ index fc42d418bf..e38f517574 100644
  CREATE SCHEMA addr_nsp;
  SET search_path TO 'addr_nsp';
 diff --git a/src/test/regress/expected/password.out b/src/test/regress/expected/password.out
-index 8475231735..0653946337 100644
+index 8475231735..1afae5395f 100644
 --- a/src/test/regress/expected/password.out
 +++ b/src/test/regress/expected/password.out
 @@ -12,11 +12,11 @@ SET password_encryption = 'md5'; -- ok
@@ -1006,63 +1006,65 @@ index 8475231735..0653946337 100644
  -----------------+---------------------------------------------------
 - regress_passwd1 | md5783277baca28003b33453252be4dbb34
 - regress_passwd2 | md54044304ba511dd062133eb5b4b84a2a3
-+ regress_passwd1 | NEON_MD5_PLACEHOLDER:regress_passwd1
-+ regress_passwd2 | NEON_MD5_PLACEHOLDER:regress_passwd2
++ regress_passwd1 | NEON_MD5_PLACEHOLDER_regress_passwd1
++ regress_passwd2 | NEON_MD5_PLACEHOLDER_regress_passwd2
   regress_passwd3 | SCRAM-SHA-256$4096:<salt>$<storedkey>:<serverkey>
 - regress_passwd4 | 
 + regress_passwd4 | SCRAM-SHA-256$4096:<salt>$<storedkey>:<serverkey>
  (4 rows)
  
  -- Rename a role
-@@ -54,24 +54,16 @@ ALTER ROLE regress_passwd2_new RENAME TO regress_passwd2;
+@@ -54,24 +54,30 @@ ALTER ROLE regress_passwd2_new RENAME TO regress_passwd2;
  -- passwords.
  SET password_encryption = 'md5';
  -- encrypt with MD5
 -ALTER ROLE regress_passwd2 PASSWORD 'foo';
---- already encrypted, use as they are
--ALTER ROLE regress_passwd1 PASSWORD 'md5cd3578025fe2c3d7ed1b9a9b26238b70';
--ALTER ROLE regress_passwd3 PASSWORD 'SCRAM-SHA-256$4096:VLK4RMaQLCvNtQ==$6YtlR4t69SguDiwFvbVgVZtuz6gpJQQqUMZ7IQJK5yI=:ps75jrHeYU4lXCcXI4O8oIdJ3eO8o2jirjruw9phBTo=';
 +ALTER ROLE regress_passwd2 PASSWORD NEON_PASSWORD_PLACEHOLDER;
+ -- already encrypted, use as they are
+ ALTER ROLE regress_passwd1 PASSWORD 'md5cd3578025fe2c3d7ed1b9a9b26238b70';
++ERROR:  Received HTTP code 400 from control plane: {"error":"Neon only supports being given plaintext passwords"}
+ ALTER ROLE regress_passwd3 PASSWORD 'SCRAM-SHA-256$4096:VLK4RMaQLCvNtQ==$6YtlR4t69SguDiwFvbVgVZtuz6gpJQQqUMZ7IQJK5yI=:ps75jrHeYU4lXCcXI4O8oIdJ3eO8o2jirjruw9phBTo=';
++ERROR:  Received HTTP code 400 from control plane: {"error":"Neon only supports being given plaintext passwords"}
  SET password_encryption = 'scram-sha-256';
  -- create SCRAM secret
 -ALTER ROLE  regress_passwd4 PASSWORD 'foo';
---- already encrypted with MD5, use as it is
--CREATE ROLE regress_passwd5 PASSWORD 'md5e73a4b11df52a6068f8b39f90be36023';
---- This looks like a valid SCRAM-SHA-256 secret, but it is not
---- so it should be hashed with SCRAM-SHA-256.
--CREATE ROLE regress_passwd6 PASSWORD 'SCRAM-SHA-256$1234';
---- These may look like valid MD5 secrets, but they are not, so they
---- should be hashed with SCRAM-SHA-256.
---- trailing garbage at the end
--CREATE ROLE regress_passwd7 PASSWORD 'md5012345678901234567890123456789zz';
---- invalid length
--CREATE ROLE regress_passwd8 PASSWORD 'md501234567890123456789012345678901zz';
 +ALTER ROLE  regress_passwd4 PASSWORD NEON_PASSWORD_PLACEHOLDER;
-+-- Neon does not support encrypted passwords, use unencrypted instead
-+CREATE ROLE regress_passwd5 PASSWORD NEON_PASSWORD_PLACEHOLDER;
-+-- Neon does not support encrypted passwords, use unencrypted instead
-+CREATE ROLE regress_passwd6 PASSWORD NEON_PASSWORD_PLACEHOLDER;
-+CREATE ROLE regress_passwd7 PASSWORD NEON_PASSWORD_PLACEHOLDER;
-+CREATE ROLE regress_passwd8 PASSWORD NEON_PASSWORD_PLACEHOLDER;
+ -- already encrypted with MD5, use as it is
+ CREATE ROLE regress_passwd5 PASSWORD 'md5e73a4b11df52a6068f8b39f90be36023';
++ERROR:  Received HTTP code 400 from control plane: {"error":"Neon only supports being given plaintext passwords"}
+ -- This looks like a valid SCRAM-SHA-256 secret, but it is not
+ -- so it should be hashed with SCRAM-SHA-256.
+ CREATE ROLE regress_passwd6 PASSWORD 'SCRAM-SHA-256$1234';
++ERROR:  Received HTTP code 400 from control plane: {"error":"Neon only supports being given plaintext passwords"}
+ -- These may look like valid MD5 secrets, but they are not, so they
+ -- should be hashed with SCRAM-SHA-256.
+ -- trailing garbage at the end
+ CREATE ROLE regress_passwd7 PASSWORD 'md5012345678901234567890123456789zz';
++ERROR:  Received HTTP code 400 from control plane: {"error":"Neon only supports being given plaintext passwords"}
+ -- invalid length
+ CREATE ROLE regress_passwd8 PASSWORD 'md501234567890123456789012345678901zz';
++ERROR:  Received HTTP code 400 from control plane: {"error":"Neon only supports being given plaintext passwords"}
  -- Changing the SCRAM iteration count
  SET scram_iterations = 1024;
  CREATE ROLE regress_passwd9 PASSWORD 'alterediterationcount';
-@@ -81,11 +73,11 @@ SELECT rolname, regexp_replace(rolpassword, '(SCRAM-SHA-256)\$(\d+):([a-zA-Z0-9+
+@@ -81,63 +87,67 @@ SELECT rolname, regexp_replace(rolpassword, '(SCRAM-SHA-256)\$(\d+):([a-zA-Z0-9+
      ORDER BY rolname, rolpassword;
       rolname     |                rolpassword_masked                 
  -----------------+---------------------------------------------------
 - regress_passwd1 | md5cd3578025fe2c3d7ed1b9a9b26238b70
 - regress_passwd2 | md5dfa155cadd5f4ad57860162f3fab9cdb
-+ regress_passwd1 | NEON_MD5_PLACEHOLDER:regress_passwd1
-+ regress_passwd2 | NEON_MD5_PLACEHOLDER:regress_passwd2
++ regress_passwd1 | NEON_MD5_PLACEHOLDER_regress_passwd1
++ regress_passwd2 | NEON_MD5_PLACEHOLDER_regress_passwd2
   regress_passwd3 | SCRAM-SHA-256$4096:<salt>$<storedkey>:<serverkey>
   regress_passwd4 | SCRAM-SHA-256$4096:<salt>$<storedkey>:<serverkey>
 - regress_passwd5 | md5e73a4b11df52a6068f8b39f90be36023
-+ regress_passwd5 | SCRAM-SHA-256$4096:<salt>$<storedkey>:<serverkey>
-  regress_passwd6 | SCRAM-SHA-256$4096:<salt>$<storedkey>:<serverkey>
-  regress_passwd7 | SCRAM-SHA-256$4096:<salt>$<storedkey>:<serverkey>
-  regress_passwd8 | SCRAM-SHA-256$4096:<salt>$<storedkey>:<serverkey>
-@@ -95,23 +87,20 @@ SELECT rolname, regexp_replace(rolpassword, '(SCRAM-SHA-256)\$(\d+):([a-zA-Z0-9+
+- regress_passwd6 | SCRAM-SHA-256$4096:<salt>$<storedkey>:<serverkey>
+- regress_passwd7 | SCRAM-SHA-256$4096:<salt>$<storedkey>:<serverkey>
+- regress_passwd8 | SCRAM-SHA-256$4096:<salt>$<storedkey>:<serverkey>
+  regress_passwd9 | SCRAM-SHA-256$1024:<salt>$<storedkey>:<serverkey>
+-(9 rows)
++(5 rows)
+ 
  -- An empty password is not allowed, in any form
  CREATE ROLE regress_passwd_empty PASSWORD '';
  NOTICE:  empty string is not a valid password, clearing password
@@ -1080,37 +1082,56 @@ index 8475231735..0653946337 100644
 -(1 row)
 +(0 rows)
  
---- Test with invalid stored and server keys.
----
---- The first is valid, to act as a control. The others have too long
---- stored/server keys. They will be re-hashed.
--CREATE ROLE regress_passwd_sha_len0 PASSWORD 'SCRAM-SHA-256$4096:A6xHKoH/494E941doaPOYg==$Ky+A30sewHIH3VHQLRN9vYsuzlgNyGNKCh37dy96Rqw=:COPdlNiIkrsacU5QoxydEuOH6e/KfiipeETb/bPw8ZI=';
--CREATE ROLE regress_passwd_sha_len1 PASSWORD 'SCRAM-SHA-256$4096:A6xHKoH/494E941doaPOYg==$Ky+A30sewHIH3VHQLRN9vYsuzlgNyGNKCh37dy96RqwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=:COPdlNiIkrsacU5QoxydEuOH6e/KfiipeETb/bPw8ZI=';
--CREATE ROLE regress_passwd_sha_len2 PASSWORD 'SCRAM-SHA-256$4096:A6xHKoH/494E941doaPOYg==$Ky+A30sewHIH3VHQLRN9vYsuzlgNyGNKCh37dy96Rqw=:COPdlNiIkrsacU5QoxydEuOH6e/KfiipeETb/bPw8ZIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=';
-+-- Neon does not support encrypted passwords, use unencrypted instead
-+CREATE ROLE regress_passwd_sha_len0 PASSWORD NEON_PASSWORD_PLACEHOLDER;
-+CREATE ROLE regress_passwd_sha_len1 PASSWORD NEON_PASSWORD_PLACEHOLDER;
-+CREATE ROLE regress_passwd_sha_len2 PASSWORD NEON_PASSWORD_PLACEHOLDER;
+ -- Test with invalid stored and server keys.
+ --
+ -- The first is valid, to act as a control. The others have too long
+ -- stored/server keys. They will be re-hashed.
+ CREATE ROLE regress_passwd_sha_len0 PASSWORD 'SCRAM-SHA-256$4096:A6xHKoH/494E941doaPOYg==$Ky+A30sewHIH3VHQLRN9vYsuzlgNyGNKCh37dy96Rqw=:COPdlNiIkrsacU5QoxydEuOH6e/KfiipeETb/bPw8ZI=';
++ERROR:  Received HTTP code 400 from control plane: {"error":"Neon only supports being given plaintext passwords"}
+ CREATE ROLE regress_passwd_sha_len1 PASSWORD 'SCRAM-SHA-256$4096:A6xHKoH/494E941doaPOYg==$Ky+A30sewHIH3VHQLRN9vYsuzlgNyGNKCh37dy96RqwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=:COPdlNiIkrsacU5QoxydEuOH6e/KfiipeETb/bPw8ZI=';
++ERROR:  Received HTTP code 400 from control plane: {"error":"Neon only supports being given plaintext passwords"}
+ CREATE ROLE regress_passwd_sha_len2 PASSWORD 'SCRAM-SHA-256$4096:A6xHKoH/494E941doaPOYg==$Ky+A30sewHIH3VHQLRN9vYsuzlgNyGNKCh37dy96Rqw=:COPdlNiIkrsacU5QoxydEuOH6e/KfiipeETb/bPw8ZIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=';
++ERROR:  Received HTTP code 400 from control plane: {"error":"Neon only supports being given plaintext passwords"}
  -- Check that the invalid secrets were re-hashed. A re-hashed secret
  -- should not contain the original salt.
  SELECT rolname, rolpassword not like '%A6xHKoH/494E941doaPOYg==%' as is_rolpassword_rehashed
-@@ -120,7 +109,7 @@ SELECT rolname, rolpassword not like '%A6xHKoH/494E941doaPOYg==%' as is_rolpassw
+     FROM pg_authid
+     WHERE rolname LIKE 'regress_passwd_sha_len%'
      ORDER BY rolname;
-          rolname         | is_rolpassword_rehashed 
- -------------------------+-------------------------
+-         rolname         | is_rolpassword_rehashed 
+--------------------------+-------------------------
 - regress_passwd_sha_len0 | f
-+ regress_passwd_sha_len0 | t
-  regress_passwd_sha_len1 | t
-  regress_passwd_sha_len2 | t
- (3 rows)
-@@ -135,6 +124,7 @@ DROP ROLE regress_passwd7;
+- regress_passwd_sha_len1 | t
+- regress_passwd_sha_len2 | t
+-(3 rows)
++ rolname | is_rolpassword_rehashed 
++---------+-------------------------
++(0 rows)
+ 
+ DROP ROLE regress_passwd1;
+ DROP ROLE regress_passwd2;
+ DROP ROLE regress_passwd3;
+ DROP ROLE regress_passwd4;
+ DROP ROLE regress_passwd5;
++ERROR:  role "regress_passwd5" does not exist
+ DROP ROLE regress_passwd6;
++ERROR:  role "regress_passwd6" does not exist
+ DROP ROLE regress_passwd7;
++ERROR:  role "regress_passwd7" does not exist
  DROP ROLE regress_passwd8;
++ERROR:  role "regress_passwd8" does not exist
  DROP ROLE regress_passwd9;
  DROP ROLE regress_passwd_empty;
 +ERROR:  role "regress_passwd_empty" does not exist
  DROP ROLE regress_passwd_sha_len0;
++ERROR:  role "regress_passwd_sha_len0" does not exist
  DROP ROLE regress_passwd_sha_len1;
++ERROR:  role "regress_passwd_sha_len1" does not exist
  DROP ROLE regress_passwd_sha_len2;
++ERROR:  role "regress_passwd_sha_len2" does not exist
+ -- all entries should have been removed
+ SELECT rolname, rolpassword
+     FROM pg_authid
 diff --git a/src/test/regress/expected/privileges.out b/src/test/regress/expected/privileges.out
 index 5b9dba7b32..cc408dad42 100644
 --- a/src/test/regress/expected/privileges.out
@@ -3173,7 +3194,7 @@ index 1a6c61f49d..1c31ac6a53 100644
  -- Test generic object addressing/identification functions
  CREATE SCHEMA addr_nsp;
 diff --git a/src/test/regress/sql/password.sql b/src/test/regress/sql/password.sql
-index 53e86b0b6c..0303fdfe96 100644
+index 53e86b0b6c..f07cf1ec54 100644
 --- a/src/test/regress/sql/password.sql
 +++ b/src/test/regress/sql/password.sql
 @@ -10,11 +10,11 @@ SET password_encryption = 'scram-sha-256'; -- ok
@@ -3192,59 +3213,23 @@ index 53e86b0b6c..0303fdfe96 100644
  
  -- check list of created entries
  --
-@@ -42,26 +42,18 @@ ALTER ROLE regress_passwd2_new RENAME TO regress_passwd2;
+@@ -42,14 +42,14 @@ ALTER ROLE regress_passwd2_new RENAME TO regress_passwd2;
  SET password_encryption = 'md5';
  
  -- encrypt with MD5
 -ALTER ROLE regress_passwd2 PASSWORD 'foo';
---- already encrypted, use as they are
--ALTER ROLE regress_passwd1 PASSWORD 'md5cd3578025fe2c3d7ed1b9a9b26238b70';
--ALTER ROLE regress_passwd3 PASSWORD 'SCRAM-SHA-256$4096:VLK4RMaQLCvNtQ==$6YtlR4t69SguDiwFvbVgVZtuz6gpJQQqUMZ7IQJK5yI=:ps75jrHeYU4lXCcXI4O8oIdJ3eO8o2jirjruw9phBTo=';
 +ALTER ROLE regress_passwd2 PASSWORD NEON_PASSWORD_PLACEHOLDER;
+ -- already encrypted, use as they are
+ ALTER ROLE regress_passwd1 PASSWORD 'md5cd3578025fe2c3d7ed1b9a9b26238b70';
+ ALTER ROLE regress_passwd3 PASSWORD 'SCRAM-SHA-256$4096:VLK4RMaQLCvNtQ==$6YtlR4t69SguDiwFvbVgVZtuz6gpJQQqUMZ7IQJK5yI=:ps75jrHeYU4lXCcXI4O8oIdJ3eO8o2jirjruw9phBTo=';
  
  SET password_encryption = 'scram-sha-256';
  -- create SCRAM secret
 -ALTER ROLE  regress_passwd4 PASSWORD 'foo';
---- already encrypted with MD5, use as it is
--CREATE ROLE regress_passwd5 PASSWORD 'md5e73a4b11df52a6068f8b39f90be36023';
 +ALTER ROLE  regress_passwd4 PASSWORD NEON_PASSWORD_PLACEHOLDER;
-+-- Neon does not support encrypted passwords, use unencrypted instead
-+CREATE ROLE regress_passwd5 PASSWORD NEON_PASSWORD_PLACEHOLDER;
+ -- already encrypted with MD5, use as it is
+ CREATE ROLE regress_passwd5 PASSWORD 'md5e73a4b11df52a6068f8b39f90be36023';
  
---- This looks like a valid SCRAM-SHA-256 secret, but it is not
---- so it should be hashed with SCRAM-SHA-256.
--CREATE ROLE regress_passwd6 PASSWORD 'SCRAM-SHA-256$1234';
---- These may look like valid MD5 secrets, but they are not, so they
---- should be hashed with SCRAM-SHA-256.
---- trailing garbage at the end
--CREATE ROLE regress_passwd7 PASSWORD 'md5012345678901234567890123456789zz';
---- invalid length
--CREATE ROLE regress_passwd8 PASSWORD 'md501234567890123456789012345678901zz';
-+-- Neon does not support encrypted passwords, use unencrypted instead
-+CREATE ROLE regress_passwd6 PASSWORD NEON_PASSWORD_PLACEHOLDER;
-+CREATE ROLE regress_passwd7 PASSWORD NEON_PASSWORD_PLACEHOLDER;
-+CREATE ROLE regress_passwd8 PASSWORD NEON_PASSWORD_PLACEHOLDER;
- 
- -- Changing the SCRAM iteration count
- SET scram_iterations = 1024;
-@@ -78,13 +70,10 @@ ALTER ROLE regress_passwd_empty PASSWORD 'md585939a5ce845f1a1b620742e3c659e0a';
- ALTER ROLE regress_passwd_empty PASSWORD 'SCRAM-SHA-256$4096:hpFyHTUsSWcR7O9P$LgZFIt6Oqdo27ZFKbZ2nV+vtnYM995pDh9ca6WSi120=:qVV5NeluNfUPkwm7Vqat25RjSPLkGeoZBQs6wVv+um4=';
- SELECT rolpassword FROM pg_authid WHERE rolname='regress_passwd_empty';
- 
---- Test with invalid stored and server keys.
----
---- The first is valid, to act as a control. The others have too long
---- stored/server keys. They will be re-hashed.
--CREATE ROLE regress_passwd_sha_len0 PASSWORD 'SCRAM-SHA-256$4096:A6xHKoH/494E941doaPOYg==$Ky+A30sewHIH3VHQLRN9vYsuzlgNyGNKCh37dy96Rqw=:COPdlNiIkrsacU5QoxydEuOH6e/KfiipeETb/bPw8ZI=';
--CREATE ROLE regress_passwd_sha_len1 PASSWORD 'SCRAM-SHA-256$4096:A6xHKoH/494E941doaPOYg==$Ky+A30sewHIH3VHQLRN9vYsuzlgNyGNKCh37dy96RqwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=:COPdlNiIkrsacU5QoxydEuOH6e/KfiipeETb/bPw8ZI=';
--CREATE ROLE regress_passwd_sha_len2 PASSWORD 'SCRAM-SHA-256$4096:A6xHKoH/494E941doaPOYg==$Ky+A30sewHIH3VHQLRN9vYsuzlgNyGNKCh37dy96Rqw=:COPdlNiIkrsacU5QoxydEuOH6e/KfiipeETb/bPw8ZIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=';
-+-- Neon does not support encrypted passwords, use unencrypted instead
-+CREATE ROLE regress_passwd_sha_len0 PASSWORD NEON_PASSWORD_PLACEHOLDER;
-+CREATE ROLE regress_passwd_sha_len1 PASSWORD NEON_PASSWORD_PLACEHOLDER;
-+CREATE ROLE regress_passwd_sha_len2 PASSWORD NEON_PASSWORD_PLACEHOLDER;
- 
- -- Check that the invalid secrets were re-hashed. A re-hashed secret
- -- should not contain the original salt.
 diff --git a/src/test/regress/sql/privileges.sql b/src/test/regress/sql/privileges.sql
 index 249df17a58..b258e7f26a 100644
 --- a/src/test/regress/sql/privileges.sql

compute/patches/cloud_regress_pg17.patch

@@ -1014,10 +1014,10 @@ index fc42d418bf..e38f517574 100644
  CREATE SCHEMA addr_nsp;
  SET search_path TO 'addr_nsp';
 diff --git a/src/test/regress/expected/password.out b/src/test/regress/expected/password.out
-index 924d6e001d..7fdda73439 100644
+index 924d6e001d..5966531db6 100644
 --- a/src/test/regress/expected/password.out
 +++ b/src/test/regress/expected/password.out
-@@ -12,13 +12,11 @@ SET password_encryption = 'md5'; -- ok
+@@ -12,13 +12,13 @@ SET password_encryption = 'md5'; -- ok
  SET password_encryption = 'scram-sha-256'; -- ok
  -- consistency of password entries
  SET password_encryption = 'md5';
@@ -1026,7 +1026,9 @@ index 924d6e001d..7fdda73439 100644
 -CREATE ROLE regress_passwd2;
 -ALTER ROLE regress_passwd2 PASSWORD 'role_pwd2';
 +CREATE ROLE regress_passwd1 PASSWORD NEON_PASSWORD_PLACEHOLDER;
++ALTER ROLE regress_passwd1 PASSWORD NEON_PASSWORD_PLACEHOLDER;
 +CREATE ROLE regress_passwd2 PASSWORD NEON_PASSWORD_PLACEHOLDER;
++ALTER ROLE regress_passwd2 PASSWORD NEON_PASSWORD_PLACEHOLDER;
  SET password_encryption = 'scram-sha-256';
 -CREATE ROLE regress_passwd3 PASSWORD 'role_pwd3';
 -CREATE ROLE regress_passwd4 PASSWORD NULL;
@@ -1035,69 +1037,71 @@ index 924d6e001d..7fdda73439 100644
  -- check list of created entries
  --
  -- The scram secret will look something like:
-@@ -32,10 +30,10 @@ SELECT rolname, regexp_replace(rolpassword, '(SCRAM-SHA-256)\$(\d+):([a-zA-Z0-9+
+@@ -32,10 +32,10 @@ SELECT rolname, regexp_replace(rolpassword, '(SCRAM-SHA-256)\$(\d+):([a-zA-Z0-9+
      ORDER BY rolname, rolpassword;
       rolname     |                rolpassword_masked                 
  -----------------+---------------------------------------------------
 - regress_passwd1 | md5783277baca28003b33453252be4dbb34
 - regress_passwd2 | md54044304ba511dd062133eb5b4b84a2a3
-+ regress_passwd1 | NEON_MD5_PLACEHOLDER:regress_passwd1
-+ regress_passwd2 | NEON_MD5_PLACEHOLDER:regress_passwd2
++ regress_passwd1 | NEON_MD5_PLACEHOLDER_regress_passwd1
++ regress_passwd2 | NEON_MD5_PLACEHOLDER_regress_passwd2
   regress_passwd3 | SCRAM-SHA-256$4096:<salt>$<storedkey>:<serverkey>
 - regress_passwd4 | 
 + regress_passwd4 | SCRAM-SHA-256$4096:<salt>$<storedkey>:<serverkey>
  (4 rows)
  
  -- Rename a role
-@@ -56,24 +54,17 @@ ALTER ROLE regress_passwd2_new RENAME TO regress_passwd2;
+@@ -56,24 +56,30 @@ ALTER ROLE regress_passwd2_new RENAME TO regress_passwd2;
  -- passwords.
  SET password_encryption = 'md5';
  -- encrypt with MD5
 -ALTER ROLE regress_passwd2 PASSWORD 'foo';
---- already encrypted, use as they are
--ALTER ROLE regress_passwd1 PASSWORD 'md5cd3578025fe2c3d7ed1b9a9b26238b70';
--ALTER ROLE regress_passwd3 PASSWORD 'SCRAM-SHA-256$4096:VLK4RMaQLCvNtQ==$6YtlR4t69SguDiwFvbVgVZtuz6gpJQQqUMZ7IQJK5yI=:ps75jrHeYU4lXCcXI4O8oIdJ3eO8o2jirjruw9phBTo=';
 +ALTER ROLE regress_passwd2 PASSWORD NEON_PASSWORD_PLACEHOLDER;
+ -- already encrypted, use as they are
+ ALTER ROLE regress_passwd1 PASSWORD 'md5cd3578025fe2c3d7ed1b9a9b26238b70';
++ERROR:  Received HTTP code 400 from control plane: {"error":"Neon only supports being given plaintext passwords"}
+ ALTER ROLE regress_passwd3 PASSWORD 'SCRAM-SHA-256$4096:VLK4RMaQLCvNtQ==$6YtlR4t69SguDiwFvbVgVZtuz6gpJQQqUMZ7IQJK5yI=:ps75jrHeYU4lXCcXI4O8oIdJ3eO8o2jirjruw9phBTo=';
++ERROR:  Received HTTP code 400 from control plane: {"error":"Neon only supports being given plaintext passwords"}
  SET password_encryption = 'scram-sha-256';
  -- create SCRAM secret
 -ALTER ROLE  regress_passwd4 PASSWORD 'foo';
 +ALTER ROLE  regress_passwd4 PASSWORD NEON_PASSWORD_PLACEHOLDER;
  -- already encrypted with MD5, use as it is
--CREATE ROLE regress_passwd5 PASSWORD 'md5e73a4b11df52a6068f8b39f90be36023';
---- This looks like a valid SCRAM-SHA-256 secret, but it is not
---- so it should be hashed with SCRAM-SHA-256.
--CREATE ROLE regress_passwd6 PASSWORD 'SCRAM-SHA-256$1234';
---- These may look like valid MD5 secrets, but they are not, so they
---- should be hashed with SCRAM-SHA-256.
---- trailing garbage at the end
--CREATE ROLE regress_passwd7 PASSWORD 'md5012345678901234567890123456789zz';
---- invalid length
--CREATE ROLE regress_passwd8 PASSWORD 'md501234567890123456789012345678901zz';
-+-- Neon does not support encrypted passwords, use unencrypted instead
-+CREATE ROLE regress_passwd5 PASSWORD NEON_PASSWORD_PLACEHOLDER;
-+-- Neon does not support encrypted passwords, use unencrypted instead
-+CREATE ROLE regress_passwd6 PASSWORD NEON_PASSWORD_PLACEHOLDER;
-+CREATE ROLE regress_passwd7 PASSWORD NEON_PASSWORD_PLACEHOLDER;
-+CREATE ROLE regress_passwd8 PASSWORD NEON_PASSWORD_PLACEHOLDER;
+ CREATE ROLE regress_passwd5 PASSWORD 'md5e73a4b11df52a6068f8b39f90be36023';
++ERROR:  Received HTTP code 400 from control plane: {"error":"Neon only supports being given plaintext passwords"}
+ -- This looks like a valid SCRAM-SHA-256 secret, but it is not
+ -- so it should be hashed with SCRAM-SHA-256.
+ CREATE ROLE regress_passwd6 PASSWORD 'SCRAM-SHA-256$1234';
++ERROR:  Received HTTP code 400 from control plane: {"error":"Neon only supports being given plaintext passwords"}
+ -- These may look like valid MD5 secrets, but they are not, so they
+ -- should be hashed with SCRAM-SHA-256.
+ -- trailing garbage at the end
+ CREATE ROLE regress_passwd7 PASSWORD 'md5012345678901234567890123456789zz';
++ERROR:  Received HTTP code 400 from control plane: {"error":"Neon only supports being given plaintext passwords"}
+ -- invalid length
+ CREATE ROLE regress_passwd8 PASSWORD 'md501234567890123456789012345678901zz';
++ERROR:  Received HTTP code 400 from control plane: {"error":"Neon only supports being given plaintext passwords"}
  -- Changing the SCRAM iteration count
  SET scram_iterations = 1024;
  CREATE ROLE regress_passwd9 PASSWORD 'alterediterationcount';
-@@ -83,11 +74,11 @@ SELECT rolname, regexp_replace(rolpassword, '(SCRAM-SHA-256)\$(\d+):([a-zA-Z0-9+
+@@ -83,63 +89,67 @@ SELECT rolname, regexp_replace(rolpassword, '(SCRAM-SHA-256)\$(\d+):([a-zA-Z0-9+
      ORDER BY rolname, rolpassword;
       rolname     |                rolpassword_masked                 
  -----------------+---------------------------------------------------
 - regress_passwd1 | md5cd3578025fe2c3d7ed1b9a9b26238b70
 - regress_passwd2 | md5dfa155cadd5f4ad57860162f3fab9cdb
-+ regress_passwd1 | NEON_MD5_PLACEHOLDER:regress_passwd1
-+ regress_passwd2 | NEON_MD5_PLACEHOLDER:regress_passwd2
++ regress_passwd1 | NEON_MD5_PLACEHOLDER_regress_passwd1
++ regress_passwd2 | NEON_MD5_PLACEHOLDER_regress_passwd2
   regress_passwd3 | SCRAM-SHA-256$4096:<salt>$<storedkey>:<serverkey>
   regress_passwd4 | SCRAM-SHA-256$4096:<salt>$<storedkey>:<serverkey>
 - regress_passwd5 | md5e73a4b11df52a6068f8b39f90be36023
-+ regress_passwd5 | SCRAM-SHA-256$4096:<salt>$<storedkey>:<serverkey>
-  regress_passwd6 | SCRAM-SHA-256$4096:<salt>$<storedkey>:<serverkey>
-  regress_passwd7 | SCRAM-SHA-256$4096:<salt>$<storedkey>:<serverkey>
-  regress_passwd8 | SCRAM-SHA-256$4096:<salt>$<storedkey>:<serverkey>
-@@ -97,23 +88,20 @@ SELECT rolname, regexp_replace(rolpassword, '(SCRAM-SHA-256)\$(\d+):([a-zA-Z0-9+
+- regress_passwd6 | SCRAM-SHA-256$4096:<salt>$<storedkey>:<serverkey>
+- regress_passwd7 | SCRAM-SHA-256$4096:<salt>$<storedkey>:<serverkey>
+- regress_passwd8 | SCRAM-SHA-256$4096:<salt>$<storedkey>:<serverkey>
+  regress_passwd9 | SCRAM-SHA-256$1024:<salt>$<storedkey>:<serverkey>
+-(9 rows)
++(5 rows)
+ 
  -- An empty password is not allowed, in any form
  CREATE ROLE regress_passwd_empty PASSWORD '';
  NOTICE:  empty string is not a valid password, clearing password
@@ -1115,37 +1119,56 @@ index 924d6e001d..7fdda73439 100644
 -(1 row)
 +(0 rows)
  
---- Test with invalid stored and server keys.
----
---- The first is valid, to act as a control. The others have too long
---- stored/server keys. They will be re-hashed.
--CREATE ROLE regress_passwd_sha_len0 PASSWORD 'SCRAM-SHA-256$4096:A6xHKoH/494E941doaPOYg==$Ky+A30sewHIH3VHQLRN9vYsuzlgNyGNKCh37dy96Rqw=:COPdlNiIkrsacU5QoxydEuOH6e/KfiipeETb/bPw8ZI=';
--CREATE ROLE regress_passwd_sha_len1 PASSWORD 'SCRAM-SHA-256$4096:A6xHKoH/494E941doaPOYg==$Ky+A30sewHIH3VHQLRN9vYsuzlgNyGNKCh37dy96RqwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=:COPdlNiIkrsacU5QoxydEuOH6e/KfiipeETb/bPw8ZI=';
--CREATE ROLE regress_passwd_sha_len2 PASSWORD 'SCRAM-SHA-256$4096:A6xHKoH/494E941doaPOYg==$Ky+A30sewHIH3VHQLRN9vYsuzlgNyGNKCh37dy96Rqw=:COPdlNiIkrsacU5QoxydEuOH6e/KfiipeETb/bPw8ZIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=';
-+-- Neon does not support encrypted passwords, use unencrypted instead
-+CREATE ROLE regress_passwd_sha_len0 PASSWORD NEON_PASSWORD_PLACEHOLDER;
-+CREATE ROLE regress_passwd_sha_len1 PASSWORD NEON_PASSWORD_PLACEHOLDER;
-+CREATE ROLE regress_passwd_sha_len2 PASSWORD NEON_PASSWORD_PLACEHOLDER;
+ -- Test with invalid stored and server keys.
+ --
+ -- The first is valid, to act as a control. The others have too long
+ -- stored/server keys. They will be re-hashed.
+ CREATE ROLE regress_passwd_sha_len0 PASSWORD 'SCRAM-SHA-256$4096:A6xHKoH/494E941doaPOYg==$Ky+A30sewHIH3VHQLRN9vYsuzlgNyGNKCh37dy96Rqw=:COPdlNiIkrsacU5QoxydEuOH6e/KfiipeETb/bPw8ZI=';
++ERROR:  Received HTTP code 400 from control plane: {"error":"Neon only supports being given plaintext passwords"}
+ CREATE ROLE regress_passwd_sha_len1 PASSWORD 'SCRAM-SHA-256$4096:A6xHKoH/494E941doaPOYg==$Ky+A30sewHIH3VHQLRN9vYsuzlgNyGNKCh37dy96RqwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=:COPdlNiIkrsacU5QoxydEuOH6e/KfiipeETb/bPw8ZI=';
++ERROR:  Received HTTP code 400 from control plane: {"error":"Neon only supports being given plaintext passwords"}
+ CREATE ROLE regress_passwd_sha_len2 PASSWORD 'SCRAM-SHA-256$4096:A6xHKoH/494E941doaPOYg==$Ky+A30sewHIH3VHQLRN9vYsuzlgNyGNKCh37dy96Rqw=:COPdlNiIkrsacU5QoxydEuOH6e/KfiipeETb/bPw8ZIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=';
++ERROR:  Received HTTP code 400 from control plane: {"error":"Neon only supports being given plaintext passwords"}
  -- Check that the invalid secrets were re-hashed. A re-hashed secret
  -- should not contain the original salt.
  SELECT rolname, rolpassword not like '%A6xHKoH/494E941doaPOYg==%' as is_rolpassword_rehashed
-@@ -122,7 +110,7 @@ SELECT rolname, rolpassword not like '%A6xHKoH/494E941doaPOYg==%' as is_rolpassw
+     FROM pg_authid
+     WHERE rolname LIKE 'regress_passwd_sha_len%'
      ORDER BY rolname;
-          rolname         | is_rolpassword_rehashed 
- -------------------------+-------------------------
+-         rolname         | is_rolpassword_rehashed 
+--------------------------+-------------------------
 - regress_passwd_sha_len0 | f
-+ regress_passwd_sha_len0 | t
-  regress_passwd_sha_len1 | t
-  regress_passwd_sha_len2 | t
- (3 rows)
-@@ -137,6 +125,7 @@ DROP ROLE regress_passwd7;
+- regress_passwd_sha_len1 | t
+- regress_passwd_sha_len2 | t
+-(3 rows)
++ rolname | is_rolpassword_rehashed 
++---------+-------------------------
++(0 rows)
+ 
+ DROP ROLE regress_passwd1;
+ DROP ROLE regress_passwd2;
+ DROP ROLE regress_passwd3;
+ DROP ROLE regress_passwd4;
+ DROP ROLE regress_passwd5;
++ERROR:  role "regress_passwd5" does not exist
+ DROP ROLE regress_passwd6;
++ERROR:  role "regress_passwd6" does not exist
+ DROP ROLE regress_passwd7;
++ERROR:  role "regress_passwd7" does not exist
  DROP ROLE regress_passwd8;
++ERROR:  role "regress_passwd8" does not exist
  DROP ROLE regress_passwd9;
  DROP ROLE regress_passwd_empty;
 +ERROR:  role "regress_passwd_empty" does not exist
  DROP ROLE regress_passwd_sha_len0;
++ERROR:  role "regress_passwd_sha_len0" does not exist
  DROP ROLE regress_passwd_sha_len1;
++ERROR:  role "regress_passwd_sha_len1" does not exist
  DROP ROLE regress_passwd_sha_len2;
++ERROR:  role "regress_passwd_sha_len2" does not exist
+ -- all entries should have been removed
+ SELECT rolname, rolpassword
+     FROM pg_authid
 diff --git a/src/test/regress/expected/privileges.out b/src/test/regress/expected/privileges.out
 index 1296da0d57..f43fffa44c 100644
 --- a/src/test/regress/expected/privileges.out
@@ -3226,10 +3249,10 @@ index 1a6c61f49d..1c31ac6a53 100644
  -- Test generic object addressing/identification functions
  CREATE SCHEMA addr_nsp;
 diff --git a/src/test/regress/sql/password.sql b/src/test/regress/sql/password.sql
-index bb82aa4aa2..dd8a05e24d 100644
+index bb82aa4aa2..7424c91b10 100644
 --- a/src/test/regress/sql/password.sql
 +++ b/src/test/regress/sql/password.sql
-@@ -10,13 +10,11 @@ SET password_encryption = 'scram-sha-256'; -- ok
+@@ -10,13 +10,13 @@ SET password_encryption = 'scram-sha-256'; -- ok
  
  -- consistency of password entries
  SET password_encryption = 'md5';
@@ -3238,7 +3261,9 @@ index bb82aa4aa2..dd8a05e24d 100644
 -CREATE ROLE regress_passwd2;
 -ALTER ROLE regress_passwd2 PASSWORD 'role_pwd2';
 +CREATE ROLE regress_passwd1 PASSWORD NEON_PASSWORD_PLACEHOLDER;
++ALTER ROLE regress_passwd1 PASSWORD NEON_PASSWORD_PLACEHOLDER;
 +CREATE ROLE regress_passwd2 PASSWORD NEON_PASSWORD_PLACEHOLDER;
++ALTER ROLE regress_passwd2 PASSWORD NEON_PASSWORD_PLACEHOLDER;
  SET password_encryption = 'scram-sha-256';
 -CREATE ROLE regress_passwd3 PASSWORD 'role_pwd3';
 -CREATE ROLE regress_passwd4 PASSWORD NULL;
@@ -3247,59 +3272,23 @@ index bb82aa4aa2..dd8a05e24d 100644
  
  -- check list of created entries
  --
-@@ -44,26 +42,19 @@ ALTER ROLE regress_passwd2_new RENAME TO regress_passwd2;
+@@ -44,14 +44,14 @@ ALTER ROLE regress_passwd2_new RENAME TO regress_passwd2;
  SET password_encryption = 'md5';
  
  -- encrypt with MD5
 -ALTER ROLE regress_passwd2 PASSWORD 'foo';
---- already encrypted, use as they are
--ALTER ROLE regress_passwd1 PASSWORD 'md5cd3578025fe2c3d7ed1b9a9b26238b70';
--ALTER ROLE regress_passwd3 PASSWORD 'SCRAM-SHA-256$4096:VLK4RMaQLCvNtQ==$6YtlR4t69SguDiwFvbVgVZtuz6gpJQQqUMZ7IQJK5yI=:ps75jrHeYU4lXCcXI4O8oIdJ3eO8o2jirjruw9phBTo=';
 +ALTER ROLE regress_passwd2 PASSWORD NEON_PASSWORD_PLACEHOLDER;
+ -- already encrypted, use as they are
+ ALTER ROLE regress_passwd1 PASSWORD 'md5cd3578025fe2c3d7ed1b9a9b26238b70';
+ ALTER ROLE regress_passwd3 PASSWORD 'SCRAM-SHA-256$4096:VLK4RMaQLCvNtQ==$6YtlR4t69SguDiwFvbVgVZtuz6gpJQQqUMZ7IQJK5yI=:ps75jrHeYU4lXCcXI4O8oIdJ3eO8o2jirjruw9phBTo=';
  
  SET password_encryption = 'scram-sha-256';
  -- create SCRAM secret
 -ALTER ROLE  regress_passwd4 PASSWORD 'foo';
 +ALTER ROLE  regress_passwd4 PASSWORD NEON_PASSWORD_PLACEHOLDER;
  -- already encrypted with MD5, use as it is
--CREATE ROLE regress_passwd5 PASSWORD 'md5e73a4b11df52a6068f8b39f90be36023';
-+-- Neon does not support encrypted passwords, use unencrypted instead
-+CREATE ROLE regress_passwd5 PASSWORD NEON_PASSWORD_PLACEHOLDER;
+ CREATE ROLE regress_passwd5 PASSWORD 'md5e73a4b11df52a6068f8b39f90be36023';
  
---- This looks like a valid SCRAM-SHA-256 secret, but it is not
---- so it should be hashed with SCRAM-SHA-256.
--CREATE ROLE regress_passwd6 PASSWORD 'SCRAM-SHA-256$1234';
---- These may look like valid MD5 secrets, but they are not, so they
---- should be hashed with SCRAM-SHA-256.
---- trailing garbage at the end
--CREATE ROLE regress_passwd7 PASSWORD 'md5012345678901234567890123456789zz';
---- invalid length
--CREATE ROLE regress_passwd8 PASSWORD 'md501234567890123456789012345678901zz';
-+-- Neon does not support encrypted passwords, use unencrypted instead
-+CREATE ROLE regress_passwd6 PASSWORD NEON_PASSWORD_PLACEHOLDER;
-+CREATE ROLE regress_passwd7 PASSWORD NEON_PASSWORD_PLACEHOLDER;
-+CREATE ROLE regress_passwd8 PASSWORD NEON_PASSWORD_PLACEHOLDER;
- 
- -- Changing the SCRAM iteration count
- SET scram_iterations = 1024;
-@@ -80,13 +71,10 @@ ALTER ROLE regress_passwd_empty PASSWORD 'md585939a5ce845f1a1b620742e3c659e0a';
- ALTER ROLE regress_passwd_empty PASSWORD 'SCRAM-SHA-256$4096:hpFyHTUsSWcR7O9P$LgZFIt6Oqdo27ZFKbZ2nV+vtnYM995pDh9ca6WSi120=:qVV5NeluNfUPkwm7Vqat25RjSPLkGeoZBQs6wVv+um4=';
- SELECT rolpassword FROM pg_authid WHERE rolname='regress_passwd_empty';
- 
---- Test with invalid stored and server keys.
----
---- The first is valid, to act as a control. The others have too long
---- stored/server keys. They will be re-hashed.
--CREATE ROLE regress_passwd_sha_len0 PASSWORD 'SCRAM-SHA-256$4096:A6xHKoH/494E941doaPOYg==$Ky+A30sewHIH3VHQLRN9vYsuzlgNyGNKCh37dy96Rqw=:COPdlNiIkrsacU5QoxydEuOH6e/KfiipeETb/bPw8ZI=';
--CREATE ROLE regress_passwd_sha_len1 PASSWORD 'SCRAM-SHA-256$4096:A6xHKoH/494E941doaPOYg==$Ky+A30sewHIH3VHQLRN9vYsuzlgNyGNKCh37dy96RqwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=:COPdlNiIkrsacU5QoxydEuOH6e/KfiipeETb/bPw8ZI=';
--CREATE ROLE regress_passwd_sha_len2 PASSWORD 'SCRAM-SHA-256$4096:A6xHKoH/494E941doaPOYg==$Ky+A30sewHIH3VHQLRN9vYsuzlgNyGNKCh37dy96Rqw=:COPdlNiIkrsacU5QoxydEuOH6e/KfiipeETb/bPw8ZIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=';
-+-- Neon does not support encrypted passwords, use unencrypted instead
-+CREATE ROLE regress_passwd_sha_len0 PASSWORD NEON_PASSWORD_PLACEHOLDER;
-+CREATE ROLE regress_passwd_sha_len1 PASSWORD NEON_PASSWORD_PLACEHOLDER;
-+CREATE ROLE regress_passwd_sha_len2 PASSWORD NEON_PASSWORD_PLACEHOLDER;
- 
- -- Check that the invalid secrets were re-hashed. A re-hashed secret
- -- should not contain the original salt.
 diff --git a/src/test/regress/sql/privileges.sql b/src/test/regress/sql/privileges.sql
 index 5880bc018d..27aa952b18 100644
 --- a/src/test/regress/sql/privileges.sql

compute_tools/src/bin/compute_ctl.rs

@@ -246,48 +246,47 @@ fn try_spec_from_cli(
     let compute_id = matches.get_one::<String>("compute-id");
     let control_plane_uri = matches.get_one::<String>("control-plane-uri");
 
-    // First, try to get cluster spec from the cli argument
-    if let Some(spec_json) = spec_json {
-        info!("got spec from cli argument {}", spec_json);
-        return Ok(CliSpecParams {
-            spec: Some(serde_json::from_str(spec_json)?),
-            live_config_allowed: false,
-        });
-    }
-
-    // Second, try to read it from the file if path is provided
-    if let Some(spec_path) = spec_path {
-        let file = File::open(Path::new(spec_path))?;
-        return Ok(CliSpecParams {
-            spec: Some(serde_json::from_reader(file)?),
-            live_config_allowed: true,
-        });
-    }
-
-    let Some(compute_id) = compute_id else {
-        panic!(
-            "compute spec should be provided by one of the following ways: \
-                --spec OR --spec-path OR --control-plane-uri and --compute-id"
-        );
-    };
-    let Some(control_plane_uri) = control_plane_uri else {
-        panic!("must specify both --control-plane-uri and --compute-id or none");
-    };
-
-    match get_spec_from_control_plane(control_plane_uri, compute_id) {
-        Ok(spec) => Ok(CliSpecParams {
-            spec,
-            live_config_allowed: true,
-        }),
-        Err(e) => {
-            error!(
-                "cannot get response from control plane: {}\n\
-                neither spec nor confirmation that compute is in the Empty state was received",
-                e
-            );
-            Err(e)
+    let spec;
+    let mut live_config_allowed = false;
+    match spec_json {
+        // First, try to get cluster spec from the cli argument
+        Some(json) => {
+            info!("got spec from cli argument {}", json);
+            spec = Some(serde_json::from_str(json)?);
         }
-    }
+        None => {
+            // Second, try to read it from the file if path is provided
+            if let Some(sp) = spec_path {
+                let path = Path::new(sp);
+                let file = File::open(path)?;
+                spec = Some(serde_json::from_reader(file)?);
+                live_config_allowed = true;
+            } else if let Some(id) = compute_id {
+                if let Some(cp_base) = control_plane_uri {
+                    live_config_allowed = true;
+                    spec = match get_spec_from_control_plane(cp_base, id) {
+                        Ok(s) => s,
+                        Err(e) => {
+                            error!("cannot get response from control plane: {}", e);
+                            panic!("neither spec nor confirmation that compute is in the Empty state was received");
+                        }
+                    };
+                } else {
+                    panic!("must specify both --control-plane-uri and --compute-id or none");
+                }
+            } else {
+                panic!(
+                    "compute spec should be provided by one of the following ways: \
+                    --spec OR --spec-path OR --control-plane-uri and --compute-id"
+                );
+            }
+        }
+    };
+
+    Ok(CliSpecParams {
+        spec,
+        live_config_allowed,
+    })
 }
 
 struct CliSpecParams {

compute_tools/src/http/openapi_spec.yaml

@@ -537,14 +537,12 @@ components:
             properties:
               extname:
                 type: string
-              version:
-                type: string
+              versions:
+                type: array
                 items:
                   type: string
               n_databases:
                 type: integer
-              owned_by_superuser:
-                type: integer
 
     SetRoleGrantsRequest:
       type: object

compute_tools/src/installed_extensions.rs

@@ -1,6 +1,7 @@
 use compute_api::responses::{InstalledExtension, InstalledExtensions};
 use metrics::proto::MetricFamily;
 use std::collections::HashMap;
+use std::collections::HashSet;
 
 use anyhow::Result;
 use postgres::{Client, NoTls};
@@ -37,77 +38,61 @@ fn list_dbs(client: &mut Client) -> Result<Vec<String>> {
 /// Connect to every database (see list_dbs above) and get the list of installed extensions.
 ///
 /// Same extension can be installed in multiple databases with different versions,
-/// so we report a separate metric (number of databases where it is installed)
-/// for each extension version.
+/// we only keep the highest and lowest version across all databases.
 pub fn get_installed_extensions(mut conf: postgres::config::Config) -> Result<InstalledExtensions> {
     conf.application_name("compute_ctl:get_installed_extensions");
     let mut client = conf.connect(NoTls)?;
+
     let databases: Vec<String> = list_dbs(&mut client)?;
 
-    let mut extensions_map: HashMap<(String, String, String), InstalledExtension> = HashMap::new();
+    let mut extensions_map: HashMap<String, InstalledExtension> = HashMap::new();
     for db in databases.iter() {
         conf.dbname(db);
         let mut db_client = conf.connect(NoTls)?;
-        let extensions: Vec<(String, String, i32)> = db_client
+        let extensions: Vec<(String, String)> = db_client
             .query(
-                "SELECT extname, extversion, extowner::integer FROM pg_catalog.pg_extension",
+                "SELECT extname, extversion FROM pg_catalog.pg_extension;",
                 &[],
             )?
             .iter()
-            .map(|row| {
-                (
-                    row.get("extname"),
-                    row.get("extversion"),
-                    row.get("extowner"),
-                )
-            })
+            .map(|row| (row.get("extname"), row.get("extversion")))
             .collect();
 
-        for (extname, v, extowner) in extensions.iter() {
+        for (extname, v) in extensions.iter() {
             let version = v.to_string();
 
-            // check if the extension is owned by superuser
-            // 10 is the oid of superuser
-            let owned_by_superuser = if *extowner == 10 { "1" } else { "0" };
+            // increment the number of databases where the version of extension is installed
+            INSTALLED_EXTENSIONS
+                .with_label_values(&[extname, &version])
+                .inc();
 
             extensions_map
-                .entry((
-                    extname.to_string(),
-                    version.clone(),
-                    owned_by_superuser.to_string(),
-                ))
+                .entry(extname.to_string())
                 .and_modify(|e| {
+                    e.versions.insert(version.clone());
                     // count the number of databases where the extension is installed
                     e.n_databases += 1;
                 })
                 .or_insert(InstalledExtension {
                     extname: extname.to_string(),
-                    version: version.clone(),
+                    versions: HashSet::from([version.clone()]),
                     n_databases: 1,
-                    owned_by_superuser: owned_by_superuser.to_string(),
                 });
         }
     }
 
-    for (key, ext) in extensions_map.iter() {
-        let (extname, version, owned_by_superuser) = key;
-        let n_databases = ext.n_databases as u64;
-
-        INSTALLED_EXTENSIONS
-            .with_label_values(&[extname, version, owned_by_superuser])
-            .set(n_databases);
-    }
-
-    Ok(InstalledExtensions {
+    let res = InstalledExtensions {
         extensions: extensions_map.into_values().collect(),
-    })
+    };
+
+    Ok(res)
 }
 
 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"]
+        &["extension_name", "version"]
     )
     .expect("failed to define a metric")
 });

control_plane/src/background_process.rs

@@ -274,7 +274,6 @@ fn fill_remote_storage_secrets_vars(mut cmd: &mut Command) -> &mut Command {
     for env_key in [
         "AWS_ACCESS_KEY_ID",
         "AWS_SECRET_ACCESS_KEY",
-        "AWS_SESSION_TOKEN",
         "AWS_PROFILE",
         // HOME is needed in combination with `AWS_PROFILE` to pick up the SSO sessions.
         "HOME",

control_plane/src/endpoint.rs

@@ -810,7 +810,7 @@ impl Endpoint {
         }
 
         let client = reqwest::Client::builder()
-            .timeout(Duration::from_secs(120))
+            .timeout(Duration::from_secs(30))
             .build()
             .unwrap();
         let response = client

control_plane/src/pageserver.rs

@@ -435,7 +435,7 @@ impl PageServerNode {
     ) -> anyhow::Result<()> {
         let config = Self::parse_config(settings)?;
         self.http_client
-            .set_tenant_config(&models::TenantConfigRequest { tenant_id, config })
+            .tenant_config(&models::TenantConfigRequest { tenant_id, config })
             .await?;
 
         Ok(())

control_plane/storcon_cli/src/main.rs

@@ -9,8 +9,8 @@ use pageserver_api::{
     },
     models::{
         EvictionPolicy, EvictionPolicyLayerAccessThreshold, LocationConfigSecondary,
-        ShardParameters, TenantConfig, TenantConfigPatchRequest, TenantConfigRequest,
-        TenantShardSplitRequest, TenantShardSplitResponse,
+        ShardParameters, TenantConfig, TenantConfigRequest, TenantShardSplitRequest,
+        TenantShardSplitResponse,
     },
     shard::{ShardStripeSize, TenantShardId},
 };
@@ -116,19 +116,9 @@ enum Command {
         #[arg(long)]
         tenant_shard_id: TenantShardId,
     },
-    /// Set the pageserver tenant configuration of a tenant: this is the configuration structure
+    /// Modify the pageserver tenant configuration of a tenant: this is the configuration structure
     /// that is passed through to pageservers, and does not affect storage controller behavior.
-    /// Any previous tenant configs are overwritten.
-    SetTenantConfig {
-        #[arg(long)]
-        tenant_id: TenantId,
-        #[arg(long)]
-        config: String,
-    },
-    /// Patch the pageserver tenant configuration of a tenant. Any fields with null values in the
-    /// provided JSON are unset from the tenant config and all fields with non-null values are set.
-    /// Unspecified fields are not changed.
-    PatchTenantConfig {
+    TenantConfig {
         #[arg(long)]
         tenant_id: TenantId,
         #[arg(long)]
@@ -559,21 +549,11 @@ async fn main() -> anyhow::Result<()> {
                 )
                 .await?;
         }
-        Command::SetTenantConfig { tenant_id, config } => {
+        Command::TenantConfig { tenant_id, config } => {
             let tenant_conf = serde_json::from_str(&config)?;
 
             vps_client
-                .set_tenant_config(&TenantConfigRequest {
-                    tenant_id,
-                    config: tenant_conf,
-                })
-                .await?;
-        }
-        Command::PatchTenantConfig { tenant_id, config } => {
-            let tenant_conf = serde_json::from_str(&config)?;
-
-            vps_client
-                .patch_tenant_config(&TenantConfigPatchRequest {
+                .tenant_config(&TenantConfigRequest {
                     tenant_id,
                     config: tenant_conf,
                 })
@@ -756,7 +736,7 @@ async fn main() -> anyhow::Result<()> {
             threshold,
         } => {
             vps_client
-                .set_tenant_config(&TenantConfigRequest {
+                .tenant_config(&TenantConfigRequest {
                     tenant_id,
                     config: TenantConfig {
                         eviction_policy: Some(EvictionPolicy::LayerAccessThreshold(

docker-compose/compute_wrapper/var/db/postgres/specs/spec.json

@@ -132,6 +132,11 @@
                 "name": "cron.database",
                 "value": "postgres",
                 "vartype": "string"
+            },
+            {
+                "name": "session_preload_libraries",
+                "value": "anon",
+                "vartype": "string"
             }
         ]
     },

docker-compose/docker_compose_test.sh

@@ -35,11 +35,11 @@ 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
+    # The support of pg_anon not yet added to PG17, so we have to remove the corresponding option
+    if [ $pg_version -eq 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"
+      jq 'del(.cluster.settings[] | select (.name == "session_preload_libraries"))' $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
 
@@ -106,8 +106,8 @@ for pg_version in ${TEST_VERSION_ONLY-14 15 16 17}; do
         fi
     fi
     cleanup
-    # Restore the original spec.json
-    if [ "$pg_version" -ne 17 ]; then
-      mv "$SPEC_PATH/spec.bak" "$SPEC_PATH/spec.json"
+    # The support of pg_anon not yet added to PG17, so we have to remove the corresponding option
+    if [ $pg_version -eq 17 ]; then
+      mv $SPEC_PATH/spec.bak $SPEC_PATH/spec.json
     fi
 done

libs/compute_api/src/responses.rs

@@ -1,5 +1,6 @@
 //! Structs representing the JSON formats used in the compute_ctl's HTTP API.
 
+use std::collections::HashSet;
 use std::fmt::Display;
 
 use chrono::{DateTime, Utc};
@@ -162,9 +163,8 @@ pub enum ControlPlaneComputeStatus {
 #[derive(Clone, Debug, Default, Serialize)]
 pub struct InstalledExtension {
     pub extname: String,
-    pub version: String,
+    pub versions: HashSet<String>,
     pub n_databases: u32, // Number of databases using this extension
-    pub owned_by_superuser: String,
 }
 
 #[derive(Clone, Debug, Default, Serialize)]

libs/desim/src/time.rs

@@ -91,7 +91,7 @@ impl Timing {
 
     /// Return true if there is a ready event.
     fn is_event_ready(&self, queue: &mut BinaryHeap<Pending>) -> bool {
-        queue.peek().is_some_and(|x| x.time <= self.now())
+        queue.peek().map_or(false, |x| x.time <= self.now())
     }
 
     /// Clear all pending events.

libs/pageserver_api/src/controller_api.rs

@@ -75,7 +75,7 @@ pub struct TenantPolicyRequest {
     pub scheduling: Option<ShardSchedulingPolicy>,
 }
 
-#[derive(Clone, Serialize, Deserialize, PartialEq, Eq, Hash, Debug, PartialOrd, Ord)]
+#[derive(Clone, Serialize, Deserialize, PartialEq, Eq, Hash, Debug)]
 pub struct AvailabilityZone(pub String);
 
 impl Display for AvailabilityZone {

libs/pageserver_api/src/key.rs

@@ -24,7 +24,7 @@ pub struct Key {
 
 /// When working with large numbers of Keys in-memory, it is more efficient to handle them as i128 than as
 /// a struct of fields.
-#[derive(Clone, Copy, Hash, PartialEq, Eq, Ord, PartialOrd, Serialize, Deserialize, Debug)]
+#[derive(Clone, Copy, Hash, PartialEq, Eq, Ord, PartialOrd, Serialize, Deserialize)]
 pub struct CompactKey(i128);
 
 /// The storage key size.
@@ -565,10 +565,6 @@ impl Key {
             && self.field5 == 0
             && self.field6 == u32::MAX
     }
-
-    pub fn is_slru_dir_key(&self) -> bool {
-        slru_dir_kind(self).is_some()
-    }
 }
 
 #[inline(always)]

libs/pageserver_api/src/models.rs

@@ -17,7 +17,7 @@ use std::{
 
 use byteorder::{BigEndian, ReadBytesExt};
 use postgres_ffi::BLCKSZ;
-use serde::{Deserialize, Deserializer, Serialize, Serializer};
+use serde::{Deserialize, Serialize};
 use serde_with::serde_as;
 use utils::{
     completion,
@@ -325,115 +325,6 @@ impl Default for ShardParameters {
     }
 }
 
-#[derive(Debug, Default, Clone, Eq, PartialEq)]
-pub enum FieldPatch<T> {
-    Upsert(T),
-    Remove,
-    #[default]
-    Noop,
-}
-
-impl<T> FieldPatch<T> {
-    fn is_noop(&self) -> bool {
-        matches!(self, FieldPatch::Noop)
-    }
-
-    pub fn apply(self, target: &mut Option<T>) {
-        match self {
-            Self::Upsert(v) => *target = Some(v),
-            Self::Remove => *target = None,
-            Self::Noop => {}
-        }
-    }
-
-    pub fn map<U, E, F: FnOnce(T) -> Result<U, E>>(self, map: F) -> Result<FieldPatch<U>, E> {
-        match self {
-            Self::Upsert(v) => Ok(FieldPatch::<U>::Upsert(map(v)?)),
-            Self::Remove => Ok(FieldPatch::<U>::Remove),
-            Self::Noop => Ok(FieldPatch::<U>::Noop),
-        }
-    }
-}
-
-impl<'de, T: Deserialize<'de>> Deserialize<'de> for FieldPatch<T> {
-    fn deserialize<D>(deserializer: D) -> Result<Self, D::Error>
-    where
-        D: Deserializer<'de>,
-    {
-        Option::deserialize(deserializer).map(|opt| match opt {
-            None => FieldPatch::Remove,
-            Some(val) => FieldPatch::Upsert(val),
-        })
-    }
-}
-
-impl<T: Serialize> Serialize for FieldPatch<T> {
-    fn serialize<S>(&self, serializer: S) -> Result<S::Ok, S::Error>
-    where
-        S: Serializer,
-    {
-        match self {
-            FieldPatch::Upsert(val) => serializer.serialize_some(val),
-            FieldPatch::Remove => serializer.serialize_none(),
-            FieldPatch::Noop => unreachable!(),
-        }
-    }
-}
-
-#[derive(Serialize, Deserialize, Debug, Default, Clone, Eq, PartialEq)]
-#[serde(default)]
-pub struct TenantConfigPatch {
-    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
-    pub checkpoint_distance: FieldPatch<u64>,
-    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
-    pub checkpoint_timeout: FieldPatch<String>,
-    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
-    pub compaction_target_size: FieldPatch<u64>,
-    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
-    pub compaction_period: FieldPatch<String>,
-    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
-    pub compaction_threshold: 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 gc_horizon: FieldPatch<u64>,
-    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
-    pub gc_period: FieldPatch<String>,
-    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
-    pub image_creation_threshold: FieldPatch<usize>,
-    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
-    pub pitr_interval: FieldPatch<String>,
-    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
-    pub walreceiver_connect_timeout: FieldPatch<String>,
-    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
-    pub lagging_wal_timeout: FieldPatch<String>,
-    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
-    pub max_lsn_wal_lag: FieldPatch<NonZeroU64>,
-    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
-    pub eviction_policy: FieldPatch<EvictionPolicy>,
-    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
-    pub min_resident_size_override: FieldPatch<u64>,
-    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
-    pub evictions_low_residence_duration_metric_threshold: FieldPatch<String>,
-    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
-    pub heatmap_period: FieldPatch<String>,
-    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
-    pub lazy_slru_download: FieldPatch<bool>,
-    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
-    pub timeline_get_throttle: FieldPatch<ThrottleConfig>,
-    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
-    pub image_layer_creation_check_threshold: FieldPatch<u8>,
-    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
-    pub lsn_lease_length: FieldPatch<String>,
-    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
-    pub lsn_lease_length_for_ts: FieldPatch<String>,
-    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
-    pub timeline_offloading: FieldPatch<bool>,
-    #[serde(skip_serializing_if = "FieldPatch::is_noop")]
-    pub wal_receiver_protocol_override: FieldPatch<PostgresClientProtocol>,
-}
-
 /// An alternative representation of `pageserver::tenant::TenantConf` with
 /// simpler types.
 #[derive(Serialize, Deserialize, Debug, Default, Clone, Eq, PartialEq)]
@@ -465,107 +356,6 @@ pub struct TenantConfig {
     pub wal_receiver_protocol_override: Option<PostgresClientProtocol>,
 }
 
-impl TenantConfig {
-    pub fn apply_patch(self, patch: TenantConfigPatch) -> TenantConfig {
-        let Self {
-            mut checkpoint_distance,
-            mut checkpoint_timeout,
-            mut compaction_target_size,
-            mut compaction_period,
-            mut compaction_threshold,
-            mut compaction_algorithm,
-            mut gc_horizon,
-            mut gc_period,
-            mut image_creation_threshold,
-            mut pitr_interval,
-            mut walreceiver_connect_timeout,
-            mut lagging_wal_timeout,
-            mut max_lsn_wal_lag,
-            mut eviction_policy,
-            mut min_resident_size_override,
-            mut evictions_low_residence_duration_metric_threshold,
-            mut heatmap_period,
-            mut lazy_slru_download,
-            mut timeline_get_throttle,
-            mut image_layer_creation_check_threshold,
-            mut lsn_lease_length,
-            mut lsn_lease_length_for_ts,
-            mut timeline_offloading,
-            mut wal_receiver_protocol_override,
-        } = self;
-
-        patch.checkpoint_distance.apply(&mut checkpoint_distance);
-        patch.checkpoint_timeout.apply(&mut checkpoint_timeout);
-        patch
-            .compaction_target_size
-            .apply(&mut compaction_target_size);
-        patch.compaction_period.apply(&mut compaction_period);
-        patch.compaction_threshold.apply(&mut compaction_threshold);
-        patch.compaction_algorithm.apply(&mut compaction_algorithm);
-        patch.gc_horizon.apply(&mut gc_horizon);
-        patch.gc_period.apply(&mut gc_period);
-        patch
-            .image_creation_threshold
-            .apply(&mut image_creation_threshold);
-        patch.pitr_interval.apply(&mut pitr_interval);
-        patch
-            .walreceiver_connect_timeout
-            .apply(&mut walreceiver_connect_timeout);
-        patch.lagging_wal_timeout.apply(&mut lagging_wal_timeout);
-        patch.max_lsn_wal_lag.apply(&mut max_lsn_wal_lag);
-        patch.eviction_policy.apply(&mut eviction_policy);
-        patch
-            .min_resident_size_override
-            .apply(&mut min_resident_size_override);
-        patch
-            .evictions_low_residence_duration_metric_threshold
-            .apply(&mut evictions_low_residence_duration_metric_threshold);
-        patch.heatmap_period.apply(&mut heatmap_period);
-        patch.lazy_slru_download.apply(&mut lazy_slru_download);
-        patch
-            .timeline_get_throttle
-            .apply(&mut timeline_get_throttle);
-        patch
-            .image_layer_creation_check_threshold
-            .apply(&mut image_layer_creation_check_threshold);
-        patch.lsn_lease_length.apply(&mut lsn_lease_length);
-        patch
-            .lsn_lease_length_for_ts
-            .apply(&mut lsn_lease_length_for_ts);
-        patch.timeline_offloading.apply(&mut timeline_offloading);
-        patch
-            .wal_receiver_protocol_override
-            .apply(&mut wal_receiver_protocol_override);
-
-        Self {
-            checkpoint_distance,
-            checkpoint_timeout,
-            compaction_target_size,
-            compaction_period,
-            compaction_threshold,
-            compaction_algorithm,
-            gc_horizon,
-            gc_period,
-            image_creation_threshold,
-            pitr_interval,
-            walreceiver_connect_timeout,
-            lagging_wal_timeout,
-            max_lsn_wal_lag,
-            eviction_policy,
-            min_resident_size_override,
-            evictions_low_residence_duration_metric_threshold,
-            heatmap_period,
-            lazy_slru_download,
-            timeline_get_throttle,
-            image_layer_creation_check_threshold,
-            lsn_lease_length,
-            lsn_lease_length_for_ts,
-            timeline_offloading,
-            wal_receiver_protocol_override,
-        }
-    }
-}
-
 /// The policy for the aux file storage.
 ///
 /// It can be switched through `switch_aux_file_policy` tenant config.
@@ -896,14 +686,6 @@ impl TenantConfigRequest {
     }
 }
 
-#[derive(Serialize, Deserialize, Debug)]
-#[serde(deny_unknown_fields)]
-pub struct TenantConfigPatchRequest {
-    pub tenant_id: TenantId,
-    #[serde(flatten)]
-    pub config: TenantConfigPatch, // as we have a flattened field, we should reject all unknown fields in it
-}
-
 /// See [`TenantState::attachment_status`] and the OpenAPI docs for context.
 #[derive(Serialize, Deserialize, Clone)]
 #[serde(tag = "slug", content = "data", rename_all = "snake_case")]
@@ -1917,45 +1699,4 @@ mod tests {
             );
         }
     }
-
-    #[test]
-    fn test_tenant_config_patch_request_serde() {
-        let patch_request = TenantConfigPatchRequest {
-            tenant_id: TenantId::from_str("17c6d121946a61e5ab0fe5a2fd4d8215").unwrap(),
-            config: TenantConfigPatch {
-                checkpoint_distance: FieldPatch::Upsert(42),
-                gc_horizon: FieldPatch::Remove,
-                compaction_threshold: FieldPatch::Noop,
-                ..TenantConfigPatch::default()
-            },
-        };
-
-        let json = serde_json::to_string(&patch_request).unwrap();
-
-        let expected = r#"{"tenant_id":"17c6d121946a61e5ab0fe5a2fd4d8215","checkpoint_distance":42,"gc_horizon":null}"#;
-        assert_eq!(json, expected);
-
-        let decoded: TenantConfigPatchRequest = serde_json::from_str(&json).unwrap();
-        assert_eq!(decoded.tenant_id, patch_request.tenant_id);
-        assert_eq!(decoded.config, patch_request.config);
-
-        // Now apply the patch to a config to demonstrate semantics
-
-        let base = TenantConfig {
-            checkpoint_distance: Some(28),
-            gc_horizon: Some(100),
-            compaction_target_size: Some(1024),
-            ..Default::default()
-        };
-
-        let expected = TenantConfig {
-            checkpoint_distance: Some(42),
-            gc_horizon: None,
-            ..base.clone()
-        };
-
-        let patched = base.apply_patch(decoded.config);
-
-        assert_eq!(patched, expected);
-    }
 }

libs/pageserver_api/src/shard.rs

@@ -173,11 +173,7 @@ impl ShardIdentity {
 
     /// Return true if the key should be stored on all shards, not just one.
     pub fn is_key_global(&self, key: &Key) -> bool {
-        if key.is_slru_block_key()
-            || key.is_slru_segment_size_key()
-            || key.is_aux_file_key()
-            || key.is_slru_dir_key()
-        {
+        if key.is_slru_block_key() || key.is_slru_segment_size_key() || key.is_aux_file_key() {
             // Special keys that are only stored on shard 0
             false
         } else if key.is_rel_block_key() {

libs/postgres_ffi/Cargo.toml

@@ -9,11 +9,9 @@ regex.workspace = true
 bytes.workspace = true
 anyhow.workspace = true
 crc32c.workspace = true
-criterion.workspace = true
 once_cell.workspace = true
 log.workspace = true
 memoffset.workspace = true
-pprof.workspace = true
 thiserror.workspace = true
 serde.workspace = true
 utils.workspace = true
@@ -26,7 +24,3 @@ postgres.workspace = true
 [build-dependencies]
 anyhow.workspace = true
 bindgen.workspace = true
-
-[[bench]]
-name = "waldecoder"
-harness = false

libs/postgres_ffi/benches/README.md

@@ -1,26 +0,0 @@
-## Benchmarks
-
-To run benchmarks:
-
-```sh
-# All benchmarks.
-cargo bench --package postgres_ffi
-
-# Specific file.
-cargo bench --package postgres_ffi --bench waldecoder
-
-# Specific benchmark.
-cargo bench --package postgres_ffi --bench waldecoder complete_record/size=1024
-
-# List available benchmarks.
-cargo bench --package postgres_ffi --benches -- --list
-
-# Generate flamegraph profiles using pprof-rs, profiling for 10 seconds.
-# Output in target/criterion/*/profile/flamegraph.svg.
-cargo bench --package postgres_ffi --bench waldecoder complete_record/size=1024 -- --profile-time 10
-```
-
-Additional charts and statistics are available in `target/criterion/report/index.html`.
-
-Benchmarks are automatically compared against the previous run. To compare against other runs, see
-`--baseline` and `--save-baseline`.

libs/postgres_ffi/benches/waldecoder.rs

@@ -1,49 +0,0 @@
-use std::ffi::CStr;
-
-use criterion::{criterion_group, criterion_main, Bencher, Criterion};
-use postgres_ffi::v17::wal_generator::LogicalMessageGenerator;
-use postgres_ffi::v17::waldecoder_handler::WalStreamDecoderHandler;
-use postgres_ffi::waldecoder::WalStreamDecoder;
-use pprof::criterion::{Output, PProfProfiler};
-use utils::lsn::Lsn;
-
-const KB: usize = 1024;
-
-// Register benchmarks with Criterion.
-criterion_group!(
-    name = benches;
-    config = Criterion::default().with_profiler(PProfProfiler::new(100, Output::Flamegraph(None)));
-    targets = bench_complete_record,
-);
-criterion_main!(benches);
-
-/// Benchmarks WalStreamDecoder::complete_record() for a logical message of varying size.
-fn bench_complete_record(c: &mut Criterion) {
-    let mut g = c.benchmark_group("complete_record");
-    for size in [64, KB, 8 * KB, 128 * KB] {
-        // Kind of weird to change the group throughput per benchmark, but it's the only way
-        // to vary it per benchmark. It works.
-        g.throughput(criterion::Throughput::Bytes(size as u64));
-        g.bench_function(format!("size={size}"), |b| run_bench(b, size).unwrap());
-    }
-
-    fn run_bench(b: &mut Bencher, size: usize) -> anyhow::Result<()> {
-        const PREFIX: &CStr = c"";
-        let value_size = LogicalMessageGenerator::make_value_size(size, PREFIX);
-        let value = vec![1; value_size];
-
-        let mut decoder = WalStreamDecoder::new(Lsn(0), 170000);
-        let msg = LogicalMessageGenerator::new(PREFIX, &value)
-            .next()
-            .unwrap()
-            .encode(Lsn(0));
-        assert_eq!(msg.len(), size);
-
-        b.iter(|| {
-            let msg = msg.clone(); // Bytes::clone() is cheap
-            decoder.complete_record(msg).unwrap();
-        });
-
-        Ok(())
-    }
-}

libs/postgres_ffi/src/wal_generator.rs

@@ -231,22 +231,6 @@ impl LogicalMessageGenerator {
         };
         [&header.encode(), prefix, message].concat().into()
     }
-
-    /// Computes how large a value must be to get a record of the given size. Convenience method to
-    /// construct records of pre-determined size. Panics if the record size is too small.
-    pub fn make_value_size(record_size: usize, prefix: &CStr) -> usize {
-        let xlog_header_size = XLOG_SIZE_OF_XLOG_RECORD;
-        let lm_header_size = size_of::<XlLogicalMessage>();
-        let prefix_size = prefix.to_bytes_with_nul().len();
-        let data_header_size = match record_size - xlog_header_size - 2 {
-            0..=255 => 2,
-            256..=258 => panic!("impossible record_size {record_size}"),
-            259.. => 5,
-        };
-        record_size
-            .checked_sub(xlog_header_size + lm_header_size + prefix_size + data_header_size)
-            .expect("record_size too small")
-    }
 }
 
 impl Iterator for LogicalMessageGenerator {

libs/postgres_ffi/wal_craft/src/xlog_utils_test.rs

@@ -81,7 +81,7 @@ fn test_end_of_wal<C: crate::Crafter>(test_name: &str) {
                 continue;
             }
             let mut f = File::options().write(true).open(file.path()).unwrap();
-            static ZEROS: [u8; WAL_SEGMENT_SIZE] = [0u8; WAL_SEGMENT_SIZE];
+            const ZEROS: [u8; WAL_SEGMENT_SIZE] = [0u8; WAL_SEGMENT_SIZE];
             f.write_all(
                 &ZEROS[0..min(
                     WAL_SEGMENT_SIZE,

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

@@ -11,7 +11,7 @@ mod private {
         Query(&'a str),
     }
 
-    impl ToStatementType<'_> {
+    impl<'a> ToStatementType<'a> {
         pub async fn into_statement(self, client: &Client) -> Result<Statement, Error> {
             match self {
                 ToStatementType::Statement(s) => Ok(s.clone()),

libs/remote_storage/Cargo.toml

@@ -18,7 +18,6 @@ camino = { workspace = true, features = ["serde1"] }
 humantime-serde.workspace = true
 hyper = { workspace = true, features = ["client"] }
 futures.workspace = true
-reqwest.workspace = true
 serde.workspace = true
 serde_json.workspace = true
 tokio = { workspace = true, features = ["sync", "fs", "io-util"] }

libs/remote_storage/src/azure_blob.rs

@@ -8,7 +8,6 @@ use std::io;
 use std::num::NonZeroU32;
 use std::pin::Pin;
 use std::str::FromStr;
-use std::sync::Arc;
 use std::time::Duration;
 use std::time::SystemTime;
 
@@ -16,8 +15,6 @@ use super::REMOTE_STORAGE_PREFIX_SEPARATOR;
 use anyhow::Context;
 use anyhow::Result;
 use azure_core::request_options::{IfMatchCondition, MaxResults, Metadata, Range};
-use azure_core::HttpClient;
-use azure_core::TransportOptions;
 use azure_core::{Continuable, RetryOptions};
 use azure_storage::StorageCredentials;
 use azure_storage_blobs::blob::CopyStatus;
@@ -83,13 +80,8 @@ impl AzureBlobStorage {
             StorageCredentials::token_credential(token_credential)
         };
 
-        let builder = ClientBuilder::new(account, credentials)
-            // we have an outer retry
-            .retry(RetryOptions::none())
-            // Customize transport to configure conneciton pooling
-            .transport(TransportOptions::new(Self::reqwest_client(
-                azure_config.conn_pool_size,
-            )));
+        // we have an outer retry
+        let builder = ClientBuilder::new(account, credentials).retry(RetryOptions::none());
 
         let client = builder.container_client(azure_config.container_name.to_owned());
 
@@ -114,14 +106,6 @@ impl AzureBlobStorage {
         })
     }
 
-    fn reqwest_client(conn_pool_size: usize) -> Arc<dyn HttpClient> {
-        let client = reqwest::ClientBuilder::new()
-            .pool_max_idle_per_host(conn_pool_size)
-            .build()
-            .expect("failed to build `reqwest` client");
-        Arc::new(client)
-    }
-
     pub fn relative_path_to_name(&self, path: &RemotePath) -> String {
         assert_eq!(std::path::MAIN_SEPARATOR, REMOTE_STORAGE_PREFIX_SEPARATOR);
         let path_string = path.get_path().as_str();
@@ -560,9 +544,9 @@ impl RemoteStorage for AzureBlobStorage {
             .await
     }
 
-    async fn delete_objects(
+    async fn delete_objects<'a>(
         &self,
-        paths: &[RemotePath],
+        paths: &'a [RemotePath],
         cancel: &CancellationToken,
     ) -> anyhow::Result<()> {
         let kind = RequestKind::Delete;
@@ -572,7 +556,7 @@ impl RemoteStorage for AzureBlobStorage {
         let op = async {
             // TODO batch requests are not supported by the SDK
             // https://github.com/Azure/azure-sdk-for-rust/issues/1068
-            for path in paths {
+            for path_chunk in paths.chunks(256) {
                 #[derive(Debug)]
                 enum AzureOrTimeout {
                     AzureError(azure_core::Error),
@@ -588,13 +572,20 @@ impl RemoteStorage for AzureBlobStorage {
                 let max_retries = 5;
                 backoff::retry(
                     || async {
-                        let blob_client = self.client.blob_client(self.relative_path_to_name(path));
+                        let mut batch_client = self.client.blob_batch();
+                        for path in path_chunk {
+                            batch_client = match batch_client.delete(self.relative_path_to_name(path)) {
+                                Ok(batch_client) => batch_client,
+                                Err(e) => return Err(AzureOrTimeout::AzureError(e)),
+                            };
+                        }
 
-                        let request = blob_client.delete().into_future();
+                        let request = batch_client.into_future();
 
                         let res = tokio::time::timeout(self.timeout, request).await;
 
                         match res {
+                            // TODO: validate that all deletions were successful
                             Ok(Ok(_v)) => Ok(()),
                             Ok(Err(azure_err)) => {
                                 if let Some(http_err) = azure_err.as_http_error() {
@@ -640,10 +631,6 @@ impl RemoteStorage for AzureBlobStorage {
         res
     }
 
-    fn max_keys_per_delete(&self) -> usize {
-        super::MAX_KEYS_PER_DELETE_AZURE
-    }
-
     async fn copy(
         &self,
         from: &RemotePath,

libs/remote_storage/src/config.rs

@@ -114,16 +114,6 @@ fn default_max_keys_per_list_response() -> Option<i32> {
     DEFAULT_MAX_KEYS_PER_LIST_RESPONSE
 }
 
-fn default_azure_conn_pool_size() -> usize {
-    // Conservative default: no connection pooling.  At time of writing this is the Azure
-    // SDK's default as well, due to historic reports of hard-to-reproduce issues
-    // (https://github.com/hyperium/hyper/issues/2312)
-    //
-    // However, using connection pooling is important to avoid exhausting client ports when
-    // doing huge numbers of requests (https://github.com/neondatabase/cloud/issues/20971)
-    0
-}
-
 impl Debug for S3Config {
     fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
         f.debug_struct("S3Config")
@@ -156,8 +146,6 @@ pub struct AzureConfig {
     pub concurrency_limit: NonZeroUsize,
     #[serde(default = "default_max_keys_per_list_response")]
     pub max_keys_per_list_response: Option<i32>,
-    #[serde(default = "default_azure_conn_pool_size")]
-    pub conn_pool_size: usize,
 }
 
 fn default_remote_storage_azure_concurrency_limit() -> NonZeroUsize {
@@ -314,7 +302,6 @@ timeout = '5s'";
     container_region = 'westeurope'
     upload_storage_class = 'INTELLIGENT_TIERING'
     timeout = '7s'
-    conn_pool_size = 8
     ";
 
         let config = parse(toml).unwrap();
@@ -329,7 +316,6 @@ timeout = '5s'";
                     prefix_in_container: None,
                     concurrency_limit: default_remote_storage_azure_concurrency_limit(),
                     max_keys_per_list_response: DEFAULT_MAX_KEYS_PER_LIST_RESPONSE,
-                    conn_pool_size: 8,
                 }),
                 timeout: Duration::from_secs(7),
                 small_timeout: RemoteStorageConfig::DEFAULT_SMALL_TIMEOUT

libs/remote_storage/src/lib.rs

@@ -70,14 +70,7 @@ pub const DEFAULT_REMOTE_STORAGE_AZURE_CONCURRENCY_LIMIT: usize = 100;
 pub const DEFAULT_MAX_KEYS_PER_LIST_RESPONSE: Option<i32> = None;
 
 /// As defined in S3 docs
-///
-/// <https://docs.aws.amazon.com/AmazonS3/latest/API/API_DeleteObjects.html>
-pub const MAX_KEYS_PER_DELETE_S3: usize = 1000;
-
-/// As defined in Azure docs
-///
-/// <https://learn.microsoft.com/en-us/rest/api/storageservices/blob-batch>
-pub const MAX_KEYS_PER_DELETE_AZURE: usize = 256;
+pub const MAX_KEYS_PER_DELETE: usize = 1000;
 
 const REMOTE_STORAGE_PREFIX_SEPARATOR: char = '/';
 
@@ -341,20 +334,12 @@ pub trait RemoteStorage: Send + Sync + 'static {
     /// If the operation fails because of timeout or cancellation, the root cause of the error will be
     /// set to `TimeoutOrCancel`. In such situation it is unknown which deletions, if any, went
     /// through.
-    async fn delete_objects(
+    async fn delete_objects<'a>(
         &self,
-        paths: &[RemotePath],
+        paths: &'a [RemotePath],
         cancel: &CancellationToken,
     ) -> anyhow::Result<()>;
 
-    /// Returns the maximum number of keys that a call to [`Self::delete_objects`] can delete without chunking
-    ///
-    /// The value returned is only an optimization hint, One can pass larger number of objects to
-    /// `delete_objects` as well.
-    ///
-    /// The value is guaranteed to be >= 1.
-    fn max_keys_per_delete(&self) -> usize;
-
     /// Deletes all objects matching the given prefix.
     ///
     /// NB: this uses NoDelimiter and will match partial prefixes. For example, the prefix /a/b will
@@ -548,16 +533,6 @@ impl<Other: RemoteStorage> GenericRemoteStorage<Arc<Other>> {
         }
     }
 
-    /// [`RemoteStorage::max_keys_per_delete`]
-    pub fn max_keys_per_delete(&self) -> usize {
-        match self {
-            Self::LocalFs(s) => s.max_keys_per_delete(),
-            Self::AwsS3(s) => s.max_keys_per_delete(),
-            Self::AzureBlob(s) => s.max_keys_per_delete(),
-            Self::Unreliable(s) => s.max_keys_per_delete(),
-        }
-    }
-
     /// See [`RemoteStorage::delete_prefix`]
     pub async fn delete_prefix(
         &self,

libs/remote_storage/src/local_fs.rs

@@ -562,9 +562,9 @@ impl RemoteStorage for LocalFs {
         }
     }
 
-    async fn delete_objects(
+    async fn delete_objects<'a>(
         &self,
-        paths: &[RemotePath],
+        paths: &'a [RemotePath],
         cancel: &CancellationToken,
     ) -> anyhow::Result<()> {
         for path in paths {
@@ -573,10 +573,6 @@ impl RemoteStorage for LocalFs {
         Ok(())
     }
 
-    fn max_keys_per_delete(&self) -> usize {
-        super::MAX_KEYS_PER_DELETE_S3
-    }
-
     async fn copy(
         &self,
         from: &RemotePath,

libs/remote_storage/src/s3_bucket.rs

@@ -48,7 +48,7 @@ use crate::{
     metrics::{start_counting_cancelled_wait, start_measuring_requests},
     support::PermitCarrying,
     ConcurrencyLimiter, Download, DownloadError, DownloadOpts, Listing, ListingMode, ListingObject,
-    RemotePath, RemoteStorage, TimeTravelError, TimeoutOrCancel, MAX_KEYS_PER_DELETE_S3,
+    RemotePath, RemoteStorage, TimeTravelError, TimeoutOrCancel, MAX_KEYS_PER_DELETE,
     REMOTE_STORAGE_PREFIX_SEPARATOR,
 };
 
@@ -355,7 +355,7 @@ impl S3Bucket {
         let kind = RequestKind::Delete;
         let mut cancel = std::pin::pin!(cancel.cancelled());
 
-        for chunk in delete_objects.chunks(MAX_KEYS_PER_DELETE_S3) {
+        for chunk in delete_objects.chunks(MAX_KEYS_PER_DELETE) {
             let started_at = start_measuring_requests(kind);
 
             let req = self
@@ -813,9 +813,9 @@ impl RemoteStorage for S3Bucket {
         .await
     }
 
-    async fn delete_objects(
+    async fn delete_objects<'a>(
         &self,
-        paths: &[RemotePath],
+        paths: &'a [RemotePath],
         cancel: &CancellationToken,
     ) -> anyhow::Result<()> {
         let kind = RequestKind::Delete;
@@ -832,10 +832,6 @@ impl RemoteStorage for S3Bucket {
         self.delete_oids(&permit, &delete_objects, cancel).await
     }
 
-    fn max_keys_per_delete(&self) -> usize {
-        MAX_KEYS_PER_DELETE_S3
-    }
-
     async fn delete(&self, path: &RemotePath, cancel: &CancellationToken) -> anyhow::Result<()> {
         let paths = std::array::from_ref(path);
         self.delete_objects(paths, cancel).await

libs/remote_storage/src/simulate_failures.rs

@@ -181,9 +181,9 @@ impl RemoteStorage for UnreliableWrapper {
         self.delete_inner(path, true, cancel).await
     }
 
-    async fn delete_objects(
+    async fn delete_objects<'a>(
         &self,
-        paths: &[RemotePath],
+        paths: &'a [RemotePath],
         cancel: &CancellationToken,
     ) -> anyhow::Result<()> {
         self.attempt(RemoteOp::DeleteObjects(paths.to_vec()))?;
@@ -203,10 +203,6 @@ impl RemoteStorage for UnreliableWrapper {
         Ok(())
     }
 
-    fn max_keys_per_delete(&self) -> usize {
-        self.inner.max_keys_per_delete()
-    }
-
     async fn copy(
         &self,
         from: &RemotePath,

libs/remote_storage/tests/test_real_azure.rs

@@ -218,7 +218,6 @@ async fn create_azure_client(
             prefix_in_container: Some(format!("test_{millis}_{random:08x}/")),
             concurrency_limit: NonZeroUsize::new(100).unwrap(),
             max_keys_per_list_response,
-            conn_pool_size: 8,
         }),
         timeout: RemoteStorageConfig::DEFAULT_TIMEOUT,
         small_timeout: RemoteStorageConfig::DEFAULT_SMALL_TIMEOUT,

libs/safekeeper_api/Cargo.toml

@@ -5,9 +5,6 @@ edition.workspace = true
 license.workspace = true
 
 [dependencies]
-const_format.workspace = true
 serde.workspace = true
-postgres_ffi.workspace = true
-pq_proto.workspace = true
-tokio.workspace = true
+const_format.workspace = true
 utils.workspace = true

libs/safekeeper_api/src/lib.rs

@@ -1,27 +1,10 @@
 #![deny(unsafe_code)]
 #![deny(clippy::undocumented_unsafe_blocks)]
 use const_format::formatcp;
-use pq_proto::SystemId;
-use serde::{Deserialize, Serialize};
 
 /// Public API types
 pub mod models;
 
-/// Consensus logical timestamp. Note: it is a part of sk control file.
-pub type Term = u64;
-pub const INVALID_TERM: Term = 0;
-
-/// Information about Postgres. Safekeeper gets it once and then verifies all
-/// further connections from computes match. Note: it is a part of sk control
-/// file.
-#[derive(Debug, Clone, PartialEq, Eq, Serialize, Deserialize)]
-pub struct ServerInfo {
-    /// Postgres server version
-    pub pg_version: u32,
-    pub system_id: SystemId,
-    pub wal_seg_size: u32,
-}
-
 pub const DEFAULT_PG_LISTEN_PORT: u16 = 5454;
 pub const DEFAULT_PG_LISTEN_ADDR: &str = formatcp!("127.0.0.1:{DEFAULT_PG_LISTEN_PORT}");
 

libs/safekeeper_api/src/models.rs

@@ -1,23 +1,10 @@
-//! Types used in safekeeper http API. Many of them are also reused internally.
-
-use postgres_ffi::TimestampTz;
 use serde::{Deserialize, Serialize};
-use std::net::SocketAddr;
-use tokio::time::Instant;
 
 use utils::{
-    id::{NodeId, TenantId, TenantTimelineId, TimelineId},
+    id::{NodeId, TenantId, TimelineId},
     lsn::Lsn,
-    pageserver_feedback::PageserverFeedback,
 };
 
-use crate::{ServerInfo, Term};
-
-#[derive(Debug, Serialize)]
-pub struct SafekeeperStatus {
-    pub id: NodeId,
-}
-
 #[derive(Serialize, Deserialize)]
 pub struct TimelineCreateRequest {
     pub tenant_id: TenantId,
@@ -31,161 +18,6 @@ pub struct TimelineCreateRequest {
     pub local_start_lsn: Option<Lsn>,
 }
 
-/// Same as TermLsn, but serializes LSN using display serializer
-/// in Postgres format, i.e. 0/FFFFFFFF. Used only for the API response.
-#[derive(Debug, Clone, Copy, Serialize, Deserialize)]
-pub struct TermSwitchApiEntry {
-    pub term: Term,
-    pub lsn: Lsn,
-}
-
-/// Augment AcceptorState with last_log_term for convenience
-#[derive(Debug, Serialize, Deserialize)]
-pub struct AcceptorStateStatus {
-    pub term: Term,
-    pub epoch: Term, // aka last_log_term, old `epoch` name is left for compatibility
-    pub term_history: Vec<TermSwitchApiEntry>,
-}
-
-/// Things safekeeper should know about timeline state on peers.
-/// Used as both model and internally.
-#[derive(Debug, Clone, Serialize, Deserialize)]
-pub struct PeerInfo {
-    pub sk_id: NodeId,
-    pub term: Term,
-    /// Term of the last entry.
-    pub last_log_term: Term,
-    /// LSN of the last record.
-    pub flush_lsn: Lsn,
-    pub commit_lsn: Lsn,
-    /// Since which LSN safekeeper has WAL.
-    pub local_start_lsn: Lsn,
-    /// When info was received. Serde annotations are not very useful but make
-    /// the code compile -- we don't rely on this field externally.
-    #[serde(skip)]
-    #[serde(default = "Instant::now")]
-    pub ts: Instant,
-    pub pg_connstr: String,
-    pub http_connstr: String,
-}
-
-pub type FullTransactionId = u64;
-
-/// Hot standby feedback received from replica
-#[derive(Debug, Clone, Copy, Serialize, Deserialize)]
-pub struct HotStandbyFeedback {
-    pub ts: TimestampTz,
-    pub xmin: FullTransactionId,
-    pub catalog_xmin: FullTransactionId,
-}
-
-pub const INVALID_FULL_TRANSACTION_ID: FullTransactionId = 0;
-
-impl HotStandbyFeedback {
-    pub fn empty() -> HotStandbyFeedback {
-        HotStandbyFeedback {
-            ts: 0,
-            xmin: 0,
-            catalog_xmin: 0,
-        }
-    }
-}
-
-/// Standby status update
-#[derive(Debug, Clone, Copy, Serialize, Deserialize)]
-pub struct StandbyReply {
-    pub write_lsn: Lsn, // The location of the last WAL byte + 1 received and written to disk in the standby.
-    pub flush_lsn: Lsn, // The location of the last WAL byte + 1 flushed to disk in the standby.
-    pub apply_lsn: Lsn, // The location of the last WAL byte + 1 applied in the standby.
-    pub reply_ts: TimestampTz, // The client's system clock at the time of transmission, as microseconds since midnight on 2000-01-01.
-    pub reply_requested: bool,
-}
-
-impl StandbyReply {
-    pub fn empty() -> Self {
-        StandbyReply {
-            write_lsn: Lsn::INVALID,
-            flush_lsn: Lsn::INVALID,
-            apply_lsn: Lsn::INVALID,
-            reply_ts: 0,
-            reply_requested: false,
-        }
-    }
-}
-
-#[derive(Debug, Clone, Copy, Serialize, Deserialize)]
-pub struct StandbyFeedback {
-    pub reply: StandbyReply,
-    pub hs_feedback: HotStandbyFeedback,
-}
-
-impl StandbyFeedback {
-    pub fn empty() -> Self {
-        StandbyFeedback {
-            reply: StandbyReply::empty(),
-            hs_feedback: HotStandbyFeedback::empty(),
-        }
-    }
-}
-
-/// Receiver is either pageserver or regular standby, which have different
-/// feedbacks.
-/// Used as both model and internally.
-#[derive(Debug, Clone, Copy, Serialize, Deserialize)]
-pub enum ReplicationFeedback {
-    Pageserver(PageserverFeedback),
-    Standby(StandbyFeedback),
-}
-
-/// Uniquely identifies a WAL service connection. Logged in spans for
-/// observability.
-pub type ConnectionId = u32;
-
-/// Serialize is used only for json'ing in API response. Also used internally.
-#[derive(Debug, Clone, Serialize, Deserialize)]
-pub struct WalSenderState {
-    pub ttid: TenantTimelineId,
-    pub addr: SocketAddr,
-    pub conn_id: ConnectionId,
-    // postgres application_name
-    pub appname: Option<String>,
-    pub feedback: ReplicationFeedback,
-}
-
-#[derive(Debug, Clone, Serialize, Deserialize)]
-pub struct WalReceiverState {
-    /// None means it is recovery initiated by us (this safekeeper).
-    pub conn_id: Option<ConnectionId>,
-    pub status: WalReceiverStatus,
-}
-
-/// Walreceiver status. Currently only whether it passed voting stage and
-/// started receiving the stream, but it is easy to add more if needed.
-#[derive(Debug, Clone, Serialize, Deserialize)]
-pub enum WalReceiverStatus {
-    Voting,
-    Streaming,
-}
-
-/// Info about timeline on safekeeper ready for reporting.
-#[derive(Debug, Serialize, Deserialize)]
-pub struct TimelineStatus {
-    pub tenant_id: TenantId,
-    pub timeline_id: TimelineId,
-    pub acceptor_state: AcceptorStateStatus,
-    pub pg_info: ServerInfo,
-    pub flush_lsn: Lsn,
-    pub timeline_start_lsn: Lsn,
-    pub local_start_lsn: Lsn,
-    pub commit_lsn: Lsn,
-    pub backup_lsn: Lsn,
-    pub peer_horizon_lsn: Lsn,
-    pub remote_consistent_lsn: Lsn,
-    pub peers: Vec<PeerInfo>,
-    pub walsenders: Vec<WalSenderState>,
-    pub walreceivers: Vec<WalReceiverState>,
-}
-
 fn lsn_invalid() -> Lsn {
     Lsn::INVALID
 }

libs/utils/Cargo.toml

@@ -15,20 +15,17 @@ arc-swap.workspace = true
 sentry.workspace = true
 async-compression.workspace = true
 anyhow.workspace = true
-backtrace.workspace = true
 bincode.workspace = true
 bytes.workspace = true
 camino.workspace = true
 chrono.workspace = true
 diatomic-waker.workspace = true
-flate2.workspace = true
 git-version.workspace = true
 hex = { workspace = true, features = ["serde"] }
 humantime.workspace = true
 hyper0 = { workspace = true, features = ["full"] }
-itertools.workspace = true
 fail.workspace = true
-futures = { workspace = true }
+futures = { workspace = true}
 jemalloc_pprof.workspace = true
 jsonwebtoken.workspace = true
 nix.workspace = true

libs/utils/src/http/endpoint.rs

@@ -1,22 +1,15 @@
 use crate::auth::{AuthError, Claims, SwappableJwtAuth};
 use crate::http::error::{api_error_handler, route_error_handler, ApiError};
 use crate::http::request::{get_query_param, parse_query_param};
-use crate::pprof;
-use ::pprof::protos::Message as _;
-use ::pprof::ProfilerGuardBuilder;
 use anyhow::{anyhow, Context};
-use bytes::{Bytes, BytesMut};
 use hyper::header::{HeaderName, AUTHORIZATION, CONTENT_DISPOSITION};
 use hyper::http::HeaderValue;
 use hyper::Method;
 use hyper::{header::CONTENT_TYPE, Body, Request, Response};
 use metrics::{register_int_counter, Encoder, IntCounter, TextEncoder};
 use once_cell::sync::Lazy;
-use regex::Regex;
 use routerify::ext::RequestExt;
 use routerify::{Middleware, RequestInfo, Router, RouterBuilder};
-use tokio::sync::{mpsc, Mutex};
-use tokio_stream::wrappers::ReceiverStream;
 use tokio_util::io::ReaderStream;
 use tracing::{debug, info, info_span, warn, Instrument};
 
@@ -25,6 +18,11 @@ use std::io::Write as _;
 use std::str::FromStr;
 use std::time::Duration;
 
+use bytes::{Bytes, BytesMut};
+use pprof::protos::Message as _;
+use tokio::sync::{mpsc, Mutex};
+use tokio_stream::wrappers::ReceiverStream;
+
 static SERVE_METRICS_COUNT: Lazy<IntCounter> = Lazy::new(|| {
     register_int_counter!(
         "libmetrics_metric_handler_requests_total",
@@ -367,7 +365,7 @@ pub async fn profile_cpu_handler(req: Request<Body>) -> Result<Response<Body>, A
 
     // Take the profile.
     let report = tokio::task::spawn_blocking(move || {
-        let guard = ProfilerGuardBuilder::default()
+        let guard = pprof::ProfilerGuardBuilder::default()
             .frequency(frequency_hz)
             .blocklist(&["libc", "libgcc", "pthread", "vdso"])
             .build()?;
@@ -459,34 +457,10 @@ pub async fn profile_heap_handler(req: Request<Body>) -> Result<Response<Body>,
         }
 
         Format::Pprof => {
-            let data = tokio::task::spawn_blocking(move || {
-                let bytes = prof_ctl.dump_pprof()?;
-                // Symbolize the profile.
-                // TODO: consider moving this upstream to jemalloc_pprof and avoiding the
-                // serialization roundtrip.
-                static STRIP_FUNCTIONS: Lazy<Vec<(Regex, bool)>> = Lazy::new(|| {
-                    // Functions to strip from profiles. If true, also remove child frames.
-                    vec![
-                        (Regex::new("^__rust").unwrap(), false),
-                        (Regex::new("^_start$").unwrap(), false),
-                        (Regex::new("^irallocx_prof").unwrap(), true),
-                        (Regex::new("^prof_alloc_prep").unwrap(), true),
-                        (Regex::new("^std::rt::lang_start").unwrap(), false),
-                        (Regex::new("^std::sys::backtrace::__rust").unwrap(), false),
-                    ]
-                });
-                let profile = pprof::decode(&bytes)?;
-                let profile = pprof::symbolize(profile)?;
-                let profile = pprof::strip_locations(
-                    profile,
-                    &["libc", "libgcc", "pthread", "vdso"],
-                    &STRIP_FUNCTIONS,
-                );
-                pprof::encode(&profile)
-            })
-            .await
-            .map_err(|join_err| ApiError::InternalServerError(join_err.into()))?
-            .map_err(ApiError::InternalServerError)?;
+            let data = tokio::task::spawn_blocking(move || prof_ctl.dump_pprof())
+                .await
+                .map_err(|join_err| ApiError::InternalServerError(join_err.into()))?
+                .map_err(ApiError::InternalServerError)?;
             Response::builder()
                 .status(200)
                 .header(CONTENT_TYPE, "application/octet-stream")

libs/utils/src/lib.rs

@@ -94,10 +94,6 @@ pub mod toml_edit_ext;
 
 pub mod circuit_breaker;
 
-pub mod try_rcu;
-
-pub mod pprof;
-
 // Re-export used in macro. Avoids adding git-version as dep in target crates.
 #[doc(hidden)]
 pub use git_version;

libs/utils/src/pprof.rs

@@ -1,190 +0,0 @@
-use flate2::write::{GzDecoder, GzEncoder};
-use flate2::Compression;
-use itertools::Itertools as _;
-use once_cell::sync::Lazy;
-use pprof::protos::{Function, Line, Message as _, Profile};
-use regex::Regex;
-
-use std::borrow::Cow;
-use std::collections::{HashMap, HashSet};
-use std::ffi::c_void;
-use std::io::Write as _;
-
-/// Decodes a gzip-compressed Protobuf-encoded pprof profile.
-pub fn decode(bytes: &[u8]) -> anyhow::Result<Profile> {
-    let mut gz = GzDecoder::new(Vec::new());
-    gz.write_all(bytes)?;
-    Ok(Profile::parse_from_bytes(&gz.finish()?)?)
-}
-
-/// Encodes a pprof profile as gzip-compressed Protobuf.
-pub fn encode(profile: &Profile) -> anyhow::Result<Vec<u8>> {
-    let mut gz = GzEncoder::new(Vec::new(), Compression::default());
-    profile.write_to_writer(&mut gz)?;
-    Ok(gz.finish()?)
-}
-
-/// Symbolizes a pprof profile using the current binary.
-pub fn symbolize(mut profile: Profile) -> anyhow::Result<Profile> {
-    if !profile.function.is_empty() {
-        return Ok(profile); // already symbolized
-    }
-
-    // Collect function names.
-    let mut functions: HashMap<String, Function> = HashMap::new();
-    let mut strings: HashMap<String, i64> = profile
-        .string_table
-        .into_iter()
-        .enumerate()
-        .map(|(i, s)| (s, i as i64))
-        .collect();
-
-    // Helper to look up or register a string.
-    let mut string_id = |s: &str| -> i64 {
-        // Don't use .entry() to avoid unnecessary allocations.
-        if let Some(id) = strings.get(s) {
-            return *id;
-        }
-        let id = strings.len() as i64;
-        strings.insert(s.to_string(), id);
-        id
-    };
-
-    for loc in &mut profile.location {
-        if !loc.line.is_empty() {
-            continue;
-        }
-
-        // Resolve the line and function for each location.
-        backtrace::resolve(loc.address as *mut c_void, |symbol| {
-            let Some(symname) = symbol.name() else {
-                return;
-            };
-            let mut name = symname.to_string();
-
-            // Strip the Rust monomorphization suffix from the symbol name.
-            static SUFFIX_REGEX: Lazy<Regex> =
-                Lazy::new(|| Regex::new("::h[0-9a-f]{16}$").expect("invalid regex"));
-            if let Some(m) = SUFFIX_REGEX.find(&name) {
-                name.truncate(m.start());
-            }
-
-            let function_id = match functions.get(&name) {
-                Some(function) => function.id,
-                None => {
-                    let id = functions.len() as u64 + 1;
-                    let system_name = String::from_utf8_lossy(symname.as_bytes());
-                    let filename = symbol
-                        .filename()
-                        .map(|path| path.to_string_lossy())
-                        .unwrap_or(Cow::Borrowed(""));
-                    let function = Function {
-                        id,
-                        name: string_id(&name),
-                        system_name: string_id(&system_name),
-                        filename: string_id(&filename),
-                        ..Default::default()
-                    };
-                    functions.insert(name, function);
-                    id
-                }
-            };
-            loc.line.push(Line {
-                function_id,
-                line: symbol.lineno().unwrap_or(0) as i64,
-                ..Default::default()
-            });
-        });
-    }
-
-    // Store the resolved functions, and mark the mapping as resolved.
-    profile.function = functions.into_values().sorted_by_key(|f| f.id).collect();
-    profile.string_table = strings
-        .into_iter()
-        .sorted_by_key(|(_, i)| *i)
-        .map(|(s, _)| s)
-        .collect();
-
-    for mapping in &mut profile.mapping {
-        mapping.has_functions = true;
-        mapping.has_filenames = true;
-    }
-
-    Ok(profile)
-}
-
-/// Strips locations (stack frames) matching the given mappings (substring) or function names
-/// (regex). The function bool specifies whether child frames should be stripped as well.
-///
-/// The string definitions are left behind in the profile for simplicity, to avoid rewriting all
-/// string references.
-pub fn strip_locations(
-    mut profile: Profile,
-    mappings: &[&str],
-    functions: &[(Regex, bool)],
-) -> Profile {
-    // Strip mappings.
-    let mut strip_mappings: HashSet<u64> = HashSet::new();
-
-    profile.mapping.retain(|mapping| {
-        let Some(name) = profile.string_table.get(mapping.filename as usize) else {
-            return true;
-        };
-        if mappings.iter().any(|substr| name.contains(substr)) {
-            strip_mappings.insert(mapping.id);
-            return false;
-        }
-        true
-    });
-
-    // Strip functions.
-    let mut strip_functions: HashMap<u64, bool> = HashMap::new();
-
-    profile.function.retain(|function| {
-        let Some(name) = profile.string_table.get(function.name as usize) else {
-            return true;
-        };
-        for (regex, strip_children) in functions {
-            if regex.is_match(name) {
-                strip_functions.insert(function.id, *strip_children);
-                return false;
-            }
-        }
-        true
-    });
-
-    // Strip locations. The bool specifies whether child frames should be stripped too.
-    let mut strip_locations: HashMap<u64, bool> = HashMap::new();
-
-    profile.location.retain(|location| {
-        for line in &location.line {
-            if let Some(strip_children) = strip_functions.get(&line.function_id) {
-                strip_locations.insert(location.id, *strip_children);
-                return false;
-            }
-        }
-        if strip_mappings.contains(&location.mapping_id) {
-            strip_locations.insert(location.id, false);
-            return false;
-        }
-        true
-    });
-
-    // Strip sample locations.
-    for sample in &mut profile.sample {
-        // First, find the uppermost function with child removal and truncate the stack.
-        if let Some(truncate) = sample
-            .location_id
-            .iter()
-            .rposition(|id| strip_locations.get(id) == Some(&true))
-        {
-            sample.location_id.drain(..=truncate);
-        }
-        // Next, strip any individual frames without child removal.
-        sample
-            .location_id
-            .retain(|id| !strip_locations.contains_key(id));
-    }
-
-    profile
-}

libs/utils/src/try_rcu.rs

@@ -1,77 +0,0 @@
-//! Try RCU extension lifted from <https://github.com/vorner/arc-swap/issues/94#issuecomment-1987154023>
-
-pub trait ArcSwapExt<T> {
-    /// [`ArcSwap::rcu`](arc_swap::ArcSwap::rcu), but with Result that short-circuits on error.
-    fn try_rcu<R, F, E>(&self, f: F) -> Result<T, E>
-    where
-        F: FnMut(&T) -> Result<R, E>,
-        R: Into<T>;
-}
-
-impl<T, S> ArcSwapExt<T> for arc_swap::ArcSwapAny<T, S>
-where
-    T: arc_swap::RefCnt,
-    S: arc_swap::strategy::CaS<T>,
-{
-    fn try_rcu<R, F, E>(&self, mut f: F) -> Result<T, E>
-    where
-        F: FnMut(&T) -> Result<R, E>,
-        R: Into<T>,
-    {
-        fn ptr_eq<Base, A, B>(a: A, b: B) -> bool
-        where
-            A: arc_swap::AsRaw<Base>,
-            B: arc_swap::AsRaw<Base>,
-        {
-            let a = a.as_raw();
-            let b = b.as_raw();
-            std::ptr::eq(a, b)
-        }
-
-        let mut cur = self.load();
-        loop {
-            let new = f(&cur)?.into();
-            let prev = self.compare_and_swap(&*cur, new);
-            let swapped = ptr_eq(&*cur, &*prev);
-            if swapped {
-                return Ok(arc_swap::Guard::into_inner(prev));
-            } else {
-                cur = prev;
-            }
-        }
-    }
-}
-
-#[cfg(test)]
-mod tests {
-    use super::*;
-    use arc_swap::ArcSwap;
-    use std::sync::Arc;
-
-    #[test]
-    fn test_try_rcu_success() {
-        let swap = ArcSwap::from(Arc::new(42));
-
-        let result = swap.try_rcu(|value| -> Result<_, String> { Ok(**value + 1) });
-
-        assert!(result.is_ok());
-        assert_eq!(**swap.load(), 43);
-    }
-
-    #[test]
-    fn test_try_rcu_error() {
-        let swap = ArcSwap::from(Arc::new(42));
-
-        let result = swap.try_rcu(|value| -> Result<i32, _> {
-            if **value == 42 {
-                Err("err")
-            } else {
-                Ok(**value + 1)
-            }
-        });
-
-        assert!(result.is_err());
-        assert_eq!(result.unwrap_err(), "err");
-        assert_eq!(**swap.load(), 42);
-    }
-}

libs/wal_decoder/proto/interpreted_wal.proto

@@ -37,7 +37,7 @@ message ValueMeta {
 }
 
 message CompactKey {
-  uint64 high = 1;
-  uint64 low = 2;
+  int64 high = 1;
+  int64 low = 2;
 }
 

libs/wal_decoder/src/wire_format.rs

@@ -236,8 +236,8 @@ impl From<ValueMeta> for proto::ValueMeta {
 impl From<CompactKey> for proto::CompactKey {
     fn from(value: CompactKey) -> Self {
         proto::CompactKey {
-            high: (value.raw() >> 64) as u64,
-            low: value.raw() as u64,
+            high: (value.raw() >> 64) as i64,
+            low: value.raw() as i64,
         }
     }
 }
@@ -354,64 +354,3 @@ impl From<proto::CompactKey> for CompactKey {
         (((value.high as i128) << 64) | (value.low as i128)).into()
     }
 }
-
-#[test]
-fn test_compact_key_with_large_relnode() {
-    use pageserver_api::key::Key;
-
-    let inputs = vec![
-        Key {
-            field1: 0,
-            field2: 0x100,
-            field3: 0x200,
-            field4: 0,
-            field5: 0x10,
-            field6: 0x5,
-        },
-        Key {
-            field1: 0,
-            field2: 0x100,
-            field3: 0x200,
-            field4: 0x007FFFFF,
-            field5: 0x10,
-            field6: 0x5,
-        },
-        Key {
-            field1: 0,
-            field2: 0x100,
-            field3: 0x200,
-            field4: 0x00800000,
-            field5: 0x10,
-            field6: 0x5,
-        },
-        Key {
-            field1: 0,
-            field2: 0x100,
-            field3: 0x200,
-            field4: 0x00800001,
-            field5: 0x10,
-            field6: 0x5,
-        },
-        Key {
-            field1: 0,
-            field2: 0xFFFFFFFF,
-            field3: 0xFFFFFFFF,
-            field4: 0xFFFFFFFF,
-            field5: 0x0,
-            field6: 0x0,
-        },
-    ];
-
-    for input in inputs {
-        assert!(input.is_valid_key_on_write_path());
-        let compact = input.to_compact();
-        let proto: proto::CompactKey = compact.into();
-        let from_proto: CompactKey = proto.into();
-
-        assert_eq!(
-            compact, from_proto,
-            "Round trip failed for key with relnode={:#x}",
-            input.field4
-        );
-    }
-}

libs/walproposer/build.rs

@@ -30,9 +30,9 @@ fn main() -> anyhow::Result<()> {
     let pgxn_neon = std::fs::canonicalize(pgxn_neon)?;
     let pgxn_neon = pgxn_neon.to_str().ok_or(anyhow!("Bad non-UTF path"))?;
 
-    println!("cargo:rustc-link-lib=static=walproposer");
     println!("cargo:rustc-link-lib=static=pgport");
     println!("cargo:rustc-link-lib=static=pgcommon");
+    println!("cargo:rustc-link-lib=static=walproposer");
     println!("cargo:rustc-link-search={walproposer_lib_search_str}");
 
     // Rebuild crate when libwalproposer.a changes

pageserver/client/src/mgmt_api.rs

@@ -270,18 +270,12 @@ impl Client {
         Ok(body)
     }
 
-    pub async fn set_tenant_config(&self, req: &TenantConfigRequest) -> Result<()> {
+    pub async fn tenant_config(&self, req: &TenantConfigRequest) -> Result<()> {
         let uri = format!("{}/v1/tenant/config", self.mgmt_api_endpoint);
         self.request(Method::PUT, &uri, req).await?;
         Ok(())
     }
 
-    pub async fn patch_tenant_config(&self, req: &TenantConfigPatchRequest) -> Result<()> {
-        let uri = format!("{}/v1/tenant/config", self.mgmt_api_endpoint);
-        self.request(Method::PATCH, &uri, req).await?;
-        Ok(())
-    }
-
     pub async fn tenant_secondary_download(
         &self,
         tenant_id: TenantShardId,

pageserver/compaction/src/compact_tiered.rs

@@ -272,7 +272,7 @@ struct CompactionJob<E: CompactionJobExecutor> {
     completed: bool,
 }
 
-impl<E> LevelCompactionState<'_, E>
+impl<'a, E> LevelCompactionState<'a, E>
 where
     E: CompactionJobExecutor,
 {

pageserver/compaction/src/identify_levels.rs

@@ -224,8 +224,9 @@ impl<L> Level<L> {
             }
 
             // recalculate depth if this was the last event at this point
-            let more_events_at_this_key =
-                events_iter.peek().is_some_and(|next_e| next_e.key == e.key);
+            let more_events_at_this_key = events_iter
+                .peek()
+                .map_or(false, |next_e| next_e.key == e.key);
             if !more_events_at_this_key {
                 let mut active_depth = 0;
                 for (_end_lsn, is_image, _idx) in active_set.iter().rev() {

pageserver/compaction/src/interface.rs

@@ -148,7 +148,7 @@ pub trait CompactionDeltaLayer<E: CompactionJobExecutor + ?Sized>: CompactionLay
         Self: 'a;
 
     /// Return all keys in this delta layer.
-    fn load_keys(
+    fn load_keys<'a>(
         &self,
         ctx: &E::RequestContext,
     ) -> impl Future<Output = anyhow::Result<Vec<Self::DeltaEntry<'_>>>> + Send;

pageserver/compaction/src/simulator.rs

@@ -143,7 +143,7 @@ impl interface::CompactionLayer<Key> for Arc<MockDeltaLayer> {
 impl interface::CompactionDeltaLayer<MockTimeline> for Arc<MockDeltaLayer> {
     type DeltaEntry<'a> = MockRecord;
 
-    async fn load_keys(&self, _ctx: &MockRequestContext) -> anyhow::Result<Vec<MockRecord>> {
+    async fn load_keys<'a>(&self, _ctx: &MockRequestContext) -> anyhow::Result<Vec<MockRecord>> {
         Ok(self.records.clone())
     }
 }

pageserver/pagebench/src/cmd/aux_files.rs

@@ -64,7 +64,7 @@ async fn main_impl(args: Args) -> anyhow::Result<()> {
     println!("operating on timeline {}", timeline);
 
     mgmt_api_client
-        .set_tenant_config(&TenantConfigRequest {
+        .tenant_config(&TenantConfigRequest {
             tenant_id: timeline.tenant_id,
             config: TenantConfig::default(),
         })

pageserver/src/basebackup.rs

@@ -248,7 +248,7 @@ where
     }
 }
 
-impl<W> Basebackup<'_, W>
+impl<'a, W> Basebackup<'a, W>
 where
     W: AsyncWrite + Send + Sync + Unpin,
 {

pageserver/src/deletion_queue/deleter.rs

@@ -9,6 +9,7 @@
 use remote_storage::GenericRemoteStorage;
 use remote_storage::RemotePath;
 use remote_storage::TimeoutOrCancel;
+use remote_storage::MAX_KEYS_PER_DELETE;
 use std::time::Duration;
 use tokio_util::sync::CancellationToken;
 use tracing::info;
@@ -130,8 +131,7 @@ impl Deleter {
     }
 
     pub(super) async fn background(&mut self) -> Result<(), DeletionQueueError> {
-        let max_keys_per_delete = self.remote_storage.max_keys_per_delete();
-        self.accumulator.reserve(max_keys_per_delete);
+        self.accumulator.reserve(MAX_KEYS_PER_DELETE);
 
         loop {
             if self.cancel.is_cancelled() {
@@ -156,14 +156,14 @@ impl Deleter {
 
             match msg {
                 DeleterMessage::Delete(mut list) => {
-                    while !list.is_empty() || self.accumulator.len() == max_keys_per_delete {
-                        if self.accumulator.len() == max_keys_per_delete {
+                    while !list.is_empty() || self.accumulator.len() == MAX_KEYS_PER_DELETE {
+                        if self.accumulator.len() == MAX_KEYS_PER_DELETE {
                             self.flush().await?;
                             // If we have received this number of keys, proceed with attempting to execute
                             assert_eq!(self.accumulator.len(), 0);
                         }
 
-                        let available_slots = max_keys_per_delete - self.accumulator.len();
+                        let available_slots = MAX_KEYS_PER_DELETE - self.accumulator.len();
                         let take_count = std::cmp::min(available_slots, list.len());
                         for path in list.drain(list.len() - take_count..) {
                             self.accumulator.push(path);

pageserver/src/http/openapi_spec.yml

@@ -767,27 +767,7 @@ paths:
   /v1/tenant/config:
     put:
       description: |
-        Update tenant's config by setting it to the provided value
-
-        Invalid fields in the tenant config will cause the request to be rejected with status 400.
-      requestBody:
-        content:
-          application/json:
-            schema:
-              $ref: "#/components/schemas/TenantConfigRequest"
-      responses:
-        "200":
-          description: OK
-          content:
-            application/json:
-              schema:
-                type: array
-                items:
-                  $ref: "#/components/schemas/TenantInfo"
-    patch:
-      description: |
-        Update tenant's config additively by patching the updated fields provided.
-        Null values unset the field and non-null values upsert it.
+        Update tenant's config.
 
         Invalid fields in the tenant config will cause the request to be rejected with status 400.
       requestBody:

pageserver/src/http/routes.rs

@@ -28,7 +28,6 @@ use pageserver_api::models::LsnLease;
 use pageserver_api::models::LsnLeaseRequest;
 use pageserver_api::models::OffloadedTimelineInfo;
 use pageserver_api::models::ShardParameters;
-use pageserver_api::models::TenantConfigPatchRequest;
 use pageserver_api::models::TenantDetails;
 use pageserver_api::models::TenantLocationConfigRequest;
 use pageserver_api::models::TenantLocationConfigResponse;
@@ -1696,47 +1695,7 @@ async fn update_tenant_config_handler(
     crate::tenant::Tenant::persist_tenant_config(state.conf, &tenant_shard_id, &location_conf)
         .await
         .map_err(|e| ApiError::InternalServerError(anyhow::anyhow!(e)))?;
-
-    let _ = tenant
-        .update_tenant_config(|_crnt| Ok(new_tenant_conf.clone()))
-        .expect("Closure returns Ok()");
-
-    json_response(StatusCode::OK, ())
-}
-
-async fn patch_tenant_config_handler(
-    mut request: Request<Body>,
-    _cancel: CancellationToken,
-) -> Result<Response<Body>, ApiError> {
-    let request_data: TenantConfigPatchRequest = json_request(&mut request).await?;
-    let tenant_id = request_data.tenant_id;
-    check_permission(&request, Some(tenant_id))?;
-
-    let state = get_state(&request);
-
-    let tenant_shard_id = TenantShardId::unsharded(tenant_id);
-
-    let tenant = state
-        .tenant_manager
-        .get_attached_tenant_shard(tenant_shard_id)?;
-    tenant.wait_to_become_active(ACTIVE_TENANT_TIMEOUT).await?;
-
-    let updated = tenant
-        .update_tenant_config(|crnt| crnt.apply_patch(request_data.config.clone()))
-        .map_err(ApiError::BadRequest)?;
-
-    // This is a legacy API that only operates on attached tenants: the preferred
-    // API to use is the location_config/ endpoint, which lets the caller provide
-    // the full LocationConf.
-    let location_conf = LocationConf::attached_single(
-        updated,
-        tenant.get_generation(),
-        &ShardParameters::default(),
-    );
-
-    crate::tenant::Tenant::persist_tenant_config(state.conf, &tenant_shard_id, &location_conf)
-        .await
-        .map_err(|e| ApiError::InternalServerError(anyhow::anyhow!(e)))?;
+    tenant.set_new_tenant_config(new_tenant_conf);
 
     json_response(StatusCode::OK, ())
 }
@@ -2081,20 +2040,13 @@ async fn timeline_compact_handler(
         .as_ref()
         .map(|r| r.sub_compaction)
         .unwrap_or(false);
-    let sub_compaction_max_job_size_mb = compact_request
-        .as_ref()
-        .and_then(|r| r.sub_compaction_max_job_size_mb);
-
     let options = CompactOptions {
-        compact_key_range: compact_request
+        compact_range: compact_request
             .as_ref()
-            .and_then(|r| r.compact_key_range.clone()),
-        compact_lsn_range: compact_request
-            .as_ref()
-            .and_then(|r| r.compact_lsn_range.clone()),
+            .and_then(|r| r.compact_range.clone()),
+        compact_below_lsn: compact_request.as_ref().and_then(|r| r.compact_below_lsn),
         flags,
         sub_compaction,
-        sub_compaction_max_job_size_mb,
     };
 
     let scheduled = compact_request
@@ -3336,9 +3288,6 @@ pub fn make_router(
         .get("/v1/tenant/:tenant_shard_id/synthetic_size", |r| {
             api_handler(r, tenant_size_handler)
         })
-        .patch("/v1/tenant/config", |r| {
-            api_handler(r, patch_tenant_config_handler)
-        })
         .put("/v1/tenant/config", |r| {
             api_handler(r, update_tenant_config_handler)
         })

pageserver/src/metrics.rs

@@ -3,7 +3,7 @@ use metrics::{
     register_counter_vec, register_gauge_vec, register_histogram, register_histogram_vec,
     register_int_counter, register_int_counter_pair_vec, register_int_counter_vec,
     register_int_gauge, register_int_gauge_vec, register_uint_gauge, register_uint_gauge_vec,
-    Counter, CounterVec, GaugeVec, Histogram, HistogramVec, IntCounter, IntCounterPair,
+    Counter, CounterVec, Gauge, GaugeVec, Histogram, HistogramVec, IntCounter, IntCounterPair,
     IntCounterPairVec, IntCounterVec, IntGauge, IntGaugeVec, UIntGauge, UIntGaugeVec,
 };
 use once_cell::sync::Lazy;
@@ -16,6 +16,7 @@ use postgres_backend::{is_expected_io_error, QueryError};
 use pq_proto::framed::ConnectionError;
 use strum::{EnumCount, VariantNames};
 use strum_macros::{IntoStaticStr, VariantNames};
+use tracing::warn;
 use utils::id::TimelineId;
 
 /// Prometheus histogram buckets (in seconds) for operations in the critical
@@ -445,6 +446,15 @@ pub(crate) static WAIT_LSN_TIME: Lazy<Histogram> = Lazy::new(|| {
     .expect("failed to define a metric")
 });
 
+static FLUSH_WAIT_UPLOAD_TIME: Lazy<GaugeVec> = Lazy::new(|| {
+    register_gauge_vec!(
+        "pageserver_flush_wait_upload_seconds",
+        "Time spent waiting for preceding uploads during layer flush",
+        &["tenant_id", "shard_id", "timeline_id"]
+    )
+    .expect("failed to define a metric")
+});
+
 static LAST_RECORD_LSN: Lazy<IntGaugeVec> = Lazy::new(|| {
     register_int_gauge_vec!(
         "pageserver_last_record_lsn",
@@ -1215,58 +1225,32 @@ pub(crate) mod virtual_file_io_engine {
 
 pub(crate) struct SmgrOpTimer(Option<SmgrOpTimerInner>);
 pub(crate) struct SmgrOpTimerInner {
-    global_execution_latency_histo: Histogram,
-    per_timeline_execution_latency_histo: Option<Histogram>,
+    global_latency_histo: Histogram,
 
-    global_batch_wait_time: Histogram,
-    per_timeline_batch_wait_time: Histogram,
+    // Optional because not all op types are tracked per-timeline
+    per_timeline_latency_histo: Option<Histogram>,
 
     global_flush_in_progress_micros: IntCounter,
     per_timeline_flush_in_progress_micros: IntCounter,
 
-    timings: SmgrOpTimerState,
-}
-
-#[derive(Debug)]
-enum SmgrOpTimerState {
-    Received {
-        received_at: Instant,
-    },
-    ThrottleDoneExecutionStarting {
-        received_at: Instant,
-        throttle_started_at: Instant,
-        started_execution_at: Instant,
-    },
+    start: Instant,
+    throttled: Duration,
+    op: SmgrQueryType,
 }
 
 pub(crate) struct SmgrOpFlushInProgress {
-    flush_started_at: Instant,
+    base: Instant,
     global_micros: IntCounter,
     per_timeline_micros: IntCounter,
 }
 
 impl SmgrOpTimer {
-    pub(crate) fn observe_throttle_done_execution_starting(&mut self, throttle: &ThrottleResult) {
+    pub(crate) fn deduct_throttle(&mut self, throttle: &Option<Duration>) {
+        let Some(throttle) = throttle else {
+            return;
+        };
         let inner = self.0.as_mut().expect("other public methods consume self");
-        match (&mut inner.timings, throttle) {
-            (SmgrOpTimerState::Received { received_at }, throttle) => match throttle {
-                ThrottleResult::NotThrottled { start } => {
-                    inner.timings = SmgrOpTimerState::ThrottleDoneExecutionStarting {
-                        received_at: *received_at,
-                        throttle_started_at: *start,
-                        started_execution_at: *start,
-                    };
-                }
-                ThrottleResult::Throttled { start, end } => {
-                    inner.timings = SmgrOpTimerState::ThrottleDoneExecutionStarting {
-                        received_at: *start,
-                        throttle_started_at: *start,
-                        started_execution_at: *end,
-                    };
-                }
-            },
-            (x, _) => panic!("called in unexpected state: {x:?}"),
-        }
+        inner.throttled += *throttle;
     }
 
     pub(crate) fn observe_smgr_op_completion_and_start_flushing(mut self) -> SmgrOpFlushInProgress {
@@ -1279,7 +1263,7 @@ impl SmgrOpTimer {
             ..
         } = inner;
         SmgrOpFlushInProgress {
-            flush_started_at: flush_start,
+            base: flush_start,
             global_micros: global_flush_in_progress_micros,
             per_timeline_micros: per_timeline_flush_in_progress_micros,
         }
@@ -1290,42 +1274,32 @@ impl SmgrOpTimer {
         let inner = self.0.take()?;
 
         let now = Instant::now();
+        let elapsed = now - inner.start;
 
-        let batch;
-        let execution;
-        let throttle;
-        match inner.timings {
-            SmgrOpTimerState::Received { received_at } => {
-                batch = (now - received_at).as_secs_f64();
-                // TODO: use label for dropped requests.
-                // This is quite rare in practice, only during tenant/pageservers shutdown.
-                throttle = Duration::ZERO;
-                execution = Duration::ZERO.as_secs_f64();
+        let elapsed = match elapsed.checked_sub(inner.throttled) {
+            Some(elapsed) => elapsed,
+            None => {
+                use utils::rate_limit::RateLimit;
+                static LOGGED: Lazy<Mutex<enum_map::EnumMap<SmgrQueryType, RateLimit>>> =
+                    Lazy::new(|| {
+                        Mutex::new(enum_map::EnumMap::from_array(std::array::from_fn(|_| {
+                            RateLimit::new(Duration::from_secs(10))
+                        })))
+                    });
+                let mut guard = LOGGED.lock().unwrap();
+                let rate_limit = &mut guard[inner.op];
+                rate_limit.call(|| {
+                    warn!(op=?inner.op, ?elapsed, ?inner.throttled, "implementation error: time spent throttled exceeds total request wall clock time");
+                });
+                elapsed // un-throttled time, more info than just saturating to 0
             }
-            SmgrOpTimerState::ThrottleDoneExecutionStarting {
-                received_at,
-                throttle_started_at,
-                started_execution_at,
-            } => {
-                batch = (throttle_started_at - received_at).as_secs_f64();
-                throttle = started_execution_at - throttle_started_at;
-                execution = (now - started_execution_at).as_secs_f64();
-            }
-        }
+        };
 
-        // update time spent in batching
-        inner.global_batch_wait_time.observe(batch);
-        inner.per_timeline_batch_wait_time.observe(batch);
+        let elapsed = elapsed.as_secs_f64();
 
-        // time spent in throttle metric is updated by throttle impl
-        let _ = throttle;
-
-        // update metrics for execution latency
-        inner.global_execution_latency_histo.observe(execution);
-        if let Some(per_timeline_execution_latency_histo) =
-            &inner.per_timeline_execution_latency_histo
-        {
-            per_timeline_execution_latency_histo.observe(execution);
+        inner.global_latency_histo.observe(elapsed);
+        if let Some(per_timeline_getpage_histo) = &inner.per_timeline_latency_histo {
+            per_timeline_getpage_histo.observe(elapsed);
         }
 
         Some((now, inner))
@@ -1351,12 +1325,12 @@ impl SmgrOpFlushInProgress {
         // Last call is tracked in `now`.
         let mut observe_guard = scopeguard::guard(
             || {
-                let elapsed = now - self.flush_started_at;
+                let elapsed = now - self.base;
                 self.global_micros
                     .inc_by(u64::try_from(elapsed.as_micros()).unwrap());
                 self.per_timeline_micros
                     .inc_by(u64::try_from(elapsed.as_micros()).unwrap());
-                self.flush_started_at = now;
+                self.base = now;
             },
             |mut observe| {
                 observe();
@@ -1403,8 +1377,6 @@ pub(crate) struct SmgrQueryTimePerTimeline {
     per_timeline_batch_size: Histogram,
     global_flush_in_progress_micros: IntCounter,
     per_timeline_flush_in_progress_micros: IntCounter,
-    global_batch_wait_time: Histogram,
-    per_timeline_batch_wait_time: Histogram,
 }
 
 static SMGR_QUERY_STARTED_GLOBAL: Lazy<IntCounterVec> = Lazy::new(|| {
@@ -1427,15 +1399,12 @@ static SMGR_QUERY_STARTED_PER_TENANT_TIMELINE: Lazy<IntCounterVec> = Lazy::new(|
     .expect("failed to define a metric")
 });
 
-// Alias so all histograms recording per-timeline smgr timings use the same buckets.
-static SMGR_QUERY_TIME_PER_TENANT_TIMELINE_BUCKETS: &[f64] = CRITICAL_OP_BUCKETS;
-
 static SMGR_QUERY_TIME_PER_TENANT_TIMELINE: Lazy<HistogramVec> = Lazy::new(|| {
     register_histogram_vec!(
         "pageserver_smgr_query_seconds",
-        "Time spent _executing_ smgr query handling, excluding batch and throttle delays.",
+        "Time spent on smgr query handling, aggegated by query type and tenant/timeline.",
         &["smgr_query_type", "tenant_id", "shard_id", "timeline_id"],
-        SMGR_QUERY_TIME_PER_TENANT_TIMELINE_BUCKETS.into(),
+        CRITICAL_OP_BUCKETS.into(),
     )
     .expect("failed to define a metric")
 });
@@ -1493,7 +1462,7 @@ static SMGR_QUERY_TIME_GLOBAL_BUCKETS: Lazy<Vec<f64>> = Lazy::new(|| {
 static SMGR_QUERY_TIME_GLOBAL: Lazy<HistogramVec> = Lazy::new(|| {
     register_histogram_vec!(
         "pageserver_smgr_query_seconds_global",
-        "Like pageserver_smgr_query_seconds, but aggregated to instance level.",
+        "Time spent on smgr query handling, aggregated by query type.",
         &["smgr_query_type"],
         SMGR_QUERY_TIME_GLOBAL_BUCKETS.clone(),
     )
@@ -1590,25 +1559,6 @@ static PAGE_SERVICE_SMGR_FLUSH_INPROGRESS_MICROS_GLOBAL: Lazy<IntCounter> = Lazy
     .expect("failed to define a metric")
 });
 
-static PAGE_SERVICE_SMGR_BATCH_WAIT_TIME: Lazy<HistogramVec> = Lazy::new(|| {
-    register_histogram_vec!(
-        "pageserver_page_service_pagestream_batch_wait_time_seconds",
-        "Time a request spent waiting in its batch until the batch moved to throttle&execution.",
-        &["tenant_id", "shard_id", "timeline_id"],
-        SMGR_QUERY_TIME_PER_TENANT_TIMELINE_BUCKETS.into(),
-    )
-    .expect("failed to define a metric")
-});
-
-static PAGE_SERVICE_SMGR_BATCH_WAIT_TIME_GLOBAL: Lazy<Histogram> = Lazy::new(|| {
-    register_histogram!(
-        "pageserver_page_service_pagestream_batch_wait_time_seconds_global",
-        "Like pageserver_page_service_pagestream_batch_wait_time_seconds, but aggregated to instance level.",
-        SMGR_QUERY_TIME_GLOBAL_BUCKETS.to_vec(),
-    )
-    .expect("failed to define a metric")
-});
-
 impl SmgrQueryTimePerTimeline {
     pub(crate) fn new(tenant_shard_id: &TenantShardId, timeline_id: &TimelineId) -> Self {
         let tenant_id = tenant_shard_id.tenant_id.to_string();
@@ -1649,11 +1599,6 @@ impl SmgrQueryTimePerTimeline {
             .get_metric_with_label_values(&[&tenant_id, &shard_slug, &timeline_id])
             .unwrap();
 
-        let global_batch_wait_time = PAGE_SERVICE_SMGR_BATCH_WAIT_TIME_GLOBAL.clone();
-        let per_timeline_batch_wait_time = PAGE_SERVICE_SMGR_BATCH_WAIT_TIME
-            .get_metric_with_label_values(&[&tenant_id, &shard_slug, &timeline_id])
-            .unwrap();
-
         let global_flush_in_progress_micros =
             PAGE_SERVICE_SMGR_FLUSH_INPROGRESS_MICROS_GLOBAL.clone();
         let per_timeline_flush_in_progress_micros = PAGE_SERVICE_SMGR_FLUSH_INPROGRESS_MICROS
@@ -1669,11 +1614,9 @@ impl SmgrQueryTimePerTimeline {
             per_timeline_batch_size,
             global_flush_in_progress_micros,
             per_timeline_flush_in_progress_micros,
-            global_batch_wait_time,
-            per_timeline_batch_wait_time,
         }
     }
-    pub(crate) fn start_smgr_op(&self, op: SmgrQueryType, received_at: Instant) -> SmgrOpTimer {
+    pub(crate) fn start_smgr_op(&self, op: SmgrQueryType, started_at: Instant) -> SmgrOpTimer {
         self.global_started[op as usize].inc();
 
         let per_timeline_latency_histo = if matches!(op, SmgrQueryType::GetPageAtLsn) {
@@ -1684,15 +1627,15 @@ impl SmgrQueryTimePerTimeline {
         };
 
         SmgrOpTimer(Some(SmgrOpTimerInner {
-            global_execution_latency_histo: self.global_latency[op as usize].clone(),
-            per_timeline_execution_latency_histo: per_timeline_latency_histo,
-            timings: SmgrOpTimerState::Received { received_at },
+            global_latency_histo: self.global_latency[op as usize].clone(),
+            per_timeline_latency_histo,
+            start: started_at,
+            op,
+            throttled: Duration::ZERO,
             global_flush_in_progress_micros: self.global_flush_in_progress_micros.clone(),
             per_timeline_flush_in_progress_micros: self
                 .per_timeline_flush_in_progress_micros
                 .clone(),
-            global_batch_wait_time: self.global_batch_wait_time.clone(),
-            per_timeline_batch_wait_time: self.per_timeline_batch_wait_time.clone(),
         }))
     }
 
@@ -2577,6 +2520,7 @@ pub(crate) struct TimelineMetrics {
     shard_id: String,
     timeline_id: String,
     pub flush_time_histo: StorageTimeMetrics,
+    pub flush_wait_upload_time_gauge: Gauge,
     pub compact_time_histo: StorageTimeMetrics,
     pub create_images_time_histo: StorageTimeMetrics,
     pub logical_size_histo: StorageTimeMetrics,
@@ -2622,6 +2566,9 @@ impl TimelineMetrics {
             &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();
         let compact_time_histo = StorageTimeMetrics::new(
             StorageTimeOperation::Compact,
             &tenant_id,
@@ -2767,6 +2714,7 @@ impl TimelineMetrics {
             shard_id,
             timeline_id,
             flush_time_histo,
+            flush_wait_upload_time_gauge,
             compact_time_histo,
             create_images_time_histo,
             logical_size_histo,
@@ -2816,6 +2764,14 @@ impl TimelineMetrics {
         self.resident_physical_size_gauge.get()
     }
 
+    pub(crate) fn flush_wait_upload_time_gauge_add(&self, duration: f64) {
+        self.flush_wait_upload_time_gauge.add(duration);
+        crate::metrics::FLUSH_WAIT_UPLOAD_TIME
+            .get_metric_with_label_values(&[&self.tenant_id, &self.shard_id, &self.timeline_id])
+            .unwrap()
+            .add(duration);
+    }
+
     pub(crate) fn shutdown(&self) {
         let was_shutdown = self
             .shutdown
@@ -2833,6 +2789,7 @@ impl TimelineMetrics {
         let shard_id = &self.shard_id;
         let _ = LAST_RECORD_LSN.remove_label_values(&[tenant_id, shard_id, timeline_id]);
         let _ = DISK_CONSISTENT_LSN.remove_label_values(&[tenant_id, shard_id, timeline_id]);
+        let _ = FLUSH_WAIT_UPLOAD_TIME.remove_label_values(&[tenant_id, shard_id, timeline_id]);
         let _ = STANDBY_HORIZON.remove_label_values(&[tenant_id, shard_id, timeline_id]);
         {
             RESIDENT_PHYSICAL_SIZE_GLOBAL.sub(self.resident_physical_size_get());
@@ -2932,11 +2889,6 @@ impl TimelineMetrics {
             shard_id,
             timeline_id,
         ]);
-        let _ = PAGE_SERVICE_SMGR_BATCH_WAIT_TIME.remove_label_values(&[
-            tenant_id,
-            shard_id,
-            timeline_id,
-        ]);
     }
 }
 
@@ -2967,7 +2919,6 @@ use crate::context::{PageContentKind, RequestContext};
 use crate::task_mgr::TaskKind;
 use crate::tenant::mgr::TenantSlot;
 use crate::tenant::tasks::BackgroundLoopKind;
-use crate::tenant::throttle::ThrottleResult;
 use crate::tenant::Timeline;
 
 /// Maintain a per timeline gauge in addition to the global gauge.
@@ -3822,7 +3773,6 @@ pub fn preinitialize_metrics(conf: &'static PageServerConf) {
         &REMOTE_ONDEMAND_DOWNLOADED_BYTES,
         &CIRCUIT_BREAKERS_BROKEN,
         &CIRCUIT_BREAKERS_UNBROKEN,
-        &PAGE_SERVICE_SMGR_FLUSH_INPROGRESS_MICROS_GLOBAL,
     ]
     .into_iter()
     .for_each(|c| {
@@ -3870,7 +3820,6 @@ pub fn preinitialize_metrics(conf: &'static PageServerConf) {
         &WAL_REDO_BYTES_HISTOGRAM,
         &WAL_REDO_PROCESS_LAUNCH_DURATION_HISTOGRAM,
         &PAGE_SERVICE_BATCH_SIZE_GLOBAL,
-        &PAGE_SERVICE_SMGR_BATCH_WAIT_TIME_GLOBAL,
     ]
     .into_iter()
     .for_each(|h| {

pageserver/src/page_service.rs

@@ -575,10 +575,7 @@ enum BatchedFeMessage {
 }
 
 impl BatchedFeMessage {
-    async fn throttle_and_record_start_processing(
-        &mut self,
-        cancel: &CancellationToken,
-    ) -> Result<(), QueryError> {
+    async fn throttle(&mut self, cancel: &CancellationToken) -> Result<(), QueryError> {
         let (shard, tokens, timers) = match self {
             BatchedFeMessage::Exists { shard, timer, .. }
             | BatchedFeMessage::Nblocks { shard, timer, .. }
@@ -606,7 +603,7 @@ impl BatchedFeMessage {
             }
         };
         for timer in timers {
-            timer.observe_throttle_done_execution_starting(&throttled);
+            timer.deduct_throttle(&throttled);
         }
         Ok(())
     }
@@ -1233,7 +1230,7 @@ impl PageServerHandler {
                 }
             };
 
-            if let Err(cancelled) = msg.throttle_and_record_start_processing(&self.cancel).await {
+            if let Err(cancelled) = msg.throttle(&self.cancel).await {
                 break cancelled;
             }
 
@@ -1400,9 +1397,7 @@ impl PageServerHandler {
                             return Err(e);
                         }
                     };
-                    batch
-                        .throttle_and_record_start_processing(&self.cancel)
-                        .await?;
+                    batch.throttle(&self.cancel).await?;
                     self.pagesteam_handle_batched_message(pgb_writer, batch, &cancel, &ctx)
                         .await?;
                 }

pageserver/src/pgdatadir_mapping.rs

@@ -1242,7 +1242,7 @@ pub struct DatadirModification<'a> {
     pending_metadata_bytes: usize,
 }
 
-impl DatadirModification<'_> {
+impl<'a> DatadirModification<'a> {
     // When a DatadirModification is committed, we do a monolithic serialization of all its contents.  WAL records can
     // contain multiple pages, so the pageserver's record-based batch size isn't sufficient to bound this allocation: we
     // additionally specify a limit on how much payload a DatadirModification may contain before it should be committed.
@@ -1263,7 +1263,7 @@ impl DatadirModification<'_> {
     pub(crate) fn has_dirty_data(&self) -> bool {
         self.pending_data_batch
             .as_ref()
-            .is_some_and(|b| b.has_data())
+            .map_or(false, |b| b.has_data())
     }
 
     /// Set the current lsn
@@ -1319,23 +1319,18 @@ impl DatadirModification<'_> {
 
         let buf: Bytes = SlruSegmentDirectory::ser(&SlruSegmentDirectory::default())?.into();
         let empty_dir = Value::Image(buf);
-
-        // Initialize SLRUs on shard 0 only: creating these on other shards would be
-        // harmless but they'd just be dropped on later compaction.
-        if self.tline.tenant_shard_id.is_shard_zero() {
-            self.put(slru_dir_to_key(SlruKind::Clog), empty_dir.clone());
-            self.pending_directory_entries
-                .push((DirectoryKind::SlruSegment(SlruKind::Clog), 0));
-            self.put(
-                slru_dir_to_key(SlruKind::MultiXactMembers),
-                empty_dir.clone(),
-            );
-            self.pending_directory_entries
-                .push((DirectoryKind::SlruSegment(SlruKind::Clog), 0));
-            self.put(slru_dir_to_key(SlruKind::MultiXactOffsets), empty_dir);
-            self.pending_directory_entries
-                .push((DirectoryKind::SlruSegment(SlruKind::MultiXactOffsets), 0));
-        }
+        self.put(slru_dir_to_key(SlruKind::Clog), empty_dir.clone());
+        self.pending_directory_entries
+            .push((DirectoryKind::SlruSegment(SlruKind::Clog), 0));
+        self.put(
+            slru_dir_to_key(SlruKind::MultiXactMembers),
+            empty_dir.clone(),
+        );
+        self.pending_directory_entries
+            .push((DirectoryKind::SlruSegment(SlruKind::Clog), 0));
+        self.put(slru_dir_to_key(SlruKind::MultiXactOffsets), empty_dir);
+        self.pending_directory_entries
+            .push((DirectoryKind::SlruSegment(SlruKind::MultiXactOffsets), 0));
 
         Ok(())
     }
@@ -2230,7 +2225,7 @@ impl DatadirModification<'_> {
                 assert!(!self
                     .pending_data_batch
                     .as_ref()
-                    .is_some_and(|b| b.updates_key(&key)));
+                    .map_or(false, |b| b.updates_key(&key)));
             }
         }
 
@@ -2299,7 +2294,7 @@ pub enum Version<'a> {
     Modified(&'a DatadirModification<'a>),
 }
 
-impl Version<'_> {
+impl<'a> Version<'a> {
     async fn get(
         &self,
         timeline: &Timeline,

pageserver/src/tenant.rs

@@ -44,7 +44,6 @@ use std::sync::atomic::AtomicBool;
 use std::sync::Weak;
 use std::time::SystemTime;
 use storage_broker::BrokerClientChannel;
-use timeline::compaction::GcCompactJob;
 use timeline::compaction::ScheduledCompactionTask;
 use timeline::import_pgdata;
 use timeline::offload::offload_timeline;
@@ -69,7 +68,6 @@ use utils::sync::gate::Gate;
 use utils::sync::gate::GateGuard;
 use utils::timeout::timeout_cancellable;
 use utils::timeout::TimeoutCancellableError;
-use utils::try_rcu::ArcSwapExt;
 use utils::zstd::create_zst_tarball;
 use utils::zstd::extract_zst_tarball;
 
@@ -3018,15 +3016,8 @@ impl Tenant {
                             warn!("ignoring scheduled compaction task: scheduled task must be gc compaction: {:?}", next_scheduled_compaction_task.options);
                         } else if next_scheduled_compaction_task.options.sub_compaction {
                             info!("running scheduled enhanced gc bottom-most compaction with sub-compaction, splitting compaction jobs");
-                            let jobs: Vec<GcCompactJob> = timeline
-                                .gc_compaction_split_jobs(
-                                    GcCompactJob::from_compact_options(
-                                        next_scheduled_compaction_task.options.clone(),
-                                    ),
-                                    next_scheduled_compaction_task
-                                        .options
-                                        .sub_compaction_max_job_size_mb,
-                                )
+                            let jobs = timeline
+                                .gc_compaction_split_jobs(next_scheduled_compaction_task.options)
                                 .await
                                 .map_err(CompactionError::Other)?;
                             if jobs.is_empty() {
@@ -3037,23 +3028,9 @@ impl Tenant {
                                 let mut guard = self.scheduled_compaction_tasks.lock().unwrap();
                                 let tline_pending_tasks = guard.entry(*timeline_id).or_default();
                                 for (idx, job) in jobs.into_iter().enumerate() {
-                                    // Unfortunately we need to convert the `GcCompactJob` back to `CompactionOptions`
-                                    // until we do further refactors to allow directly call `compact_with_gc`.
-                                    let mut flags: EnumSet<CompactFlags> = EnumSet::default();
-                                    flags |= CompactFlags::EnhancedGcBottomMostCompaction;
-                                    if job.dry_run {
-                                        flags |= CompactFlags::DryRun;
-                                    }
-                                    let options = CompactOptions {
-                                        flags,
-                                        sub_compaction: false,
-                                        compact_key_range: Some(job.compact_key_range.into()),
-                                        compact_lsn_range: Some(job.compact_lsn_range.into()),
-                                        sub_compaction_max_job_size_mb: None,
-                                    };
                                     tline_pending_tasks.push_back(if idx == jobs_len - 1 {
                                         ScheduledCompactionTask {
-                                            options,
+                                            options: job,
                                             // The last job in the queue sends the signal and releases the gc guard
                                             result_tx: next_scheduled_compaction_task
                                                 .result_tx
@@ -3064,7 +3041,7 @@ impl Tenant {
                                         }
                                     } else {
                                         ScheduledCompactionTask {
-                                            options,
+                                            options: job,
                                             result_tx: None,
                                             gc_block: None,
                                         }
@@ -3944,28 +3921,25 @@ impl Tenant {
         }
     }
 
-    pub fn update_tenant_config<F: Fn(TenantConfOpt) -> anyhow::Result<TenantConfOpt>>(
-        &self,
-        update: F,
-    ) -> anyhow::Result<TenantConfOpt> {
+    pub fn set_new_tenant_config(&self, new_tenant_conf: TenantConfOpt) {
         // Use read-copy-update in order to avoid overwriting the location config
         // state if this races with [`Tenant::set_new_location_config`]. Note that
         // this race is not possible if both request types come from the storage
         // controller (as they should!) because an exclusive op lock is required
         // on the storage controller side.
 
-        self.tenant_conf
-            .try_rcu(|attached_conf| -> Result<_, anyhow::Error> {
-                Ok(Arc::new(AttachedTenantConf {
-                    tenant_conf: update(attached_conf.tenant_conf.clone())?,
-                    location: attached_conf.location,
-                    lsn_lease_deadline: attached_conf.lsn_lease_deadline,
-                }))
-            })?;
+        self.tenant_conf.rcu(|inner| {
+            Arc::new(AttachedTenantConf {
+                tenant_conf: new_tenant_conf.clone(),
+                location: inner.location,
+                // Attached location is not changed, no need to update lsn lease deadline.
+                lsn_lease_deadline: inner.lsn_lease_deadline,
+            })
+        });
 
-        let updated = self.tenant_conf.load();
+        let updated = self.tenant_conf.load().clone();
 
-        self.tenant_conf_updated(&updated.tenant_conf);
+        self.tenant_conf_updated(&new_tenant_conf);
         // Don't hold self.timelines.lock() during the notifies.
         // There's no risk of deadlock right now, but there could be if we consolidate
         // mutexes in struct Timeline in the future.
@@ -3973,8 +3947,6 @@ impl Tenant {
         for timeline in timelines {
             timeline.tenant_conf_updated(&updated);
         }
-
-        Ok(updated.tenant_conf.clone())
     }
 
     pub(crate) fn set_new_location_config(&self, new_conf: AttachedTenantConf) {
@@ -4534,12 +4506,7 @@ impl Tenant {
                 // - this timeline was created while we were finding cutoffs
                 // - lsn for timestamp search fails for this timeline repeatedly
                 if let Some(cutoffs) = gc_cutoffs.get(&timeline.timeline_id) {
-                    let original_cutoffs = target.cutoffs.clone();
-                    // GC cutoffs should never go back
-                    target.cutoffs = GcCutoffs {
-                        space: Lsn(cutoffs.space.0.max(original_cutoffs.space.0)),
-                        time: Lsn(cutoffs.time.0.max(original_cutoffs.time.0)),
-                    }
+                    target.cutoffs = cutoffs.clone();
                 }
             }
 
@@ -5764,8 +5731,6 @@ mod tests {
     #[cfg(feature = "testing")]
     use timeline::compaction::{KeyHistoryRetention, KeyLogAtLsn};
     #[cfg(feature = "testing")]
-    use timeline::CompactLsnRange;
-    #[cfg(feature = "testing")]
     use timeline::GcInfo;
 
     static TEST_KEY: Lazy<Key> =
@@ -9357,6 +9322,7 @@ mod tests {
                 &cancel,
                 CompactOptions {
                     flags: dryrun_flags,
+                    compact_range: None,
                     ..Default::default()
                 },
                 &ctx,
@@ -9605,6 +9571,7 @@ mod tests {
                 &cancel,
                 CompactOptions {
                     flags: dryrun_flags,
+                    compact_range: None,
                     ..Default::default()
                 },
                 &ctx,
@@ -9634,8 +9601,6 @@ mod tests {
     #[cfg(feature = "testing")]
     #[tokio::test]
     async fn test_simple_bottom_most_compaction_on_branch() -> anyhow::Result<()> {
-        use timeline::CompactLsnRange;
-
         let harness = TenantHarness::create("test_simple_bottom_most_compaction_on_branch").await?;
         let (tenant, ctx) = harness.load().await;
 
@@ -9828,22 +9793,6 @@ mod tests {
 
         verify_result().await;
 
-        // Piggyback a compaction with above_lsn. Ensure it works correctly when the specified LSN intersects with the layer files.
-        // Now we already have a single large delta layer, so the compaction min_layer_lsn should be the same as ancestor LSN (0x18).
-        branch_tline
-            .compact_with_gc(
-                &cancel,
-                CompactOptions {
-                    compact_lsn_range: Some(CompactLsnRange::above(Lsn(0x40))),
-                    ..Default::default()
-                },
-                &ctx,
-            )
-            .await
-            .unwrap();
-
-        verify_result().await;
-
         Ok(())
     }
 
@@ -10132,7 +10081,7 @@ mod tests {
                 &cancel,
                 CompactOptions {
                     flags: EnumSet::new(),
-                    compact_key_range: Some((get_key(0)..get_key(2)).into()),
+                    compact_range: Some((get_key(0)..get_key(2)).into()),
                     ..Default::default()
                 },
                 &ctx,
@@ -10179,7 +10128,7 @@ mod tests {
                 &cancel,
                 CompactOptions {
                     flags: EnumSet::new(),
-                    compact_key_range: Some((get_key(2)..get_key(4)).into()),
+                    compact_range: Some((get_key(2)..get_key(4)).into()),
                     ..Default::default()
                 },
                 &ctx,
@@ -10231,7 +10180,7 @@ mod tests {
                 &cancel,
                 CompactOptions {
                     flags: EnumSet::new(),
-                    compact_key_range: Some((get_key(4)..get_key(9)).into()),
+                    compact_range: Some((get_key(4)..get_key(9)).into()),
                     ..Default::default()
                 },
                 &ctx,
@@ -10282,7 +10231,7 @@ mod tests {
                 &cancel,
                 CompactOptions {
                     flags: EnumSet::new(),
-                    compact_key_range: Some((get_key(9)..get_key(10)).into()),
+                    compact_range: Some((get_key(9)..get_key(10)).into()),
                     ..Default::default()
                 },
                 &ctx,
@@ -10338,7 +10287,7 @@ mod tests {
                 &cancel,
                 CompactOptions {
                     flags: EnumSet::new(),
-                    compact_key_range: Some((get_key(0)..get_key(10)).into()),
+                    compact_range: Some((get_key(0)..get_key(10)).into()),
                     ..Default::default()
                 },
                 &ctx,
@@ -10367,6 +10316,7 @@ mod tests {
                 },
             ],
         );
+
         Ok(())
     }
 
@@ -10419,602 +10369,4 @@ mod tests {
 
         Ok(())
     }
-
-    #[cfg(feature = "testing")]
-    #[tokio::test]
-    async fn test_simple_bottom_most_compaction_above_lsn() -> anyhow::Result<()> {
-        let harness = TenantHarness::create("test_simple_bottom_most_compaction_above_lsn").await?;
-        let (tenant, ctx) = harness.load().await;
-
-        fn get_key(id: u32) -> Key {
-            // using aux key here b/c they are guaranteed to be inside `collect_keyspace`.
-            let mut key = Key::from_hex("620000000033333333444444445500000000").unwrap();
-            key.field6 = id;
-            key
-        }
-
-        let img_layer = (0..10)
-            .map(|id| (get_key(id), Bytes::from(format!("value {id}@0x10"))))
-            .collect_vec();
-
-        let delta1 = vec![(
-            get_key(1),
-            Lsn(0x20),
-            Value::WalRecord(NeonWalRecord::wal_append("@0x20")),
-        )];
-        let delta4 = vec![(
-            get_key(1),
-            Lsn(0x28),
-            Value::WalRecord(NeonWalRecord::wal_append("@0x28")),
-        )];
-        let delta2 = vec![
-            (
-                get_key(1),
-                Lsn(0x30),
-                Value::WalRecord(NeonWalRecord::wal_append("@0x30")),
-            ),
-            (
-                get_key(1),
-                Lsn(0x38),
-                Value::WalRecord(NeonWalRecord::wal_append("@0x38")),
-            ),
-        ];
-        let delta3 = vec![
-            (
-                get_key(8),
-                Lsn(0x48),
-                Value::WalRecord(NeonWalRecord::wal_append("@0x48")),
-            ),
-            (
-                get_key(9),
-                Lsn(0x48),
-                Value::WalRecord(NeonWalRecord::wal_append("@0x48")),
-            ),
-        ];
-
-        let tline = tenant
-            .create_test_timeline_with_layers(
-                TIMELINE_ID,
-                Lsn(0x10),
-                DEFAULT_PG_VERSION,
-                &ctx,
-                vec![
-                    // delta1/2/4 only contain a single key but multiple updates
-                    DeltaLayerTestDesc::new_with_inferred_key_range(Lsn(0x20)..Lsn(0x28), delta1),
-                    DeltaLayerTestDesc::new_with_inferred_key_range(Lsn(0x30)..Lsn(0x50), delta2),
-                    DeltaLayerTestDesc::new_with_inferred_key_range(Lsn(0x28)..Lsn(0x30), delta4),
-                    DeltaLayerTestDesc::new_with_inferred_key_range(Lsn(0x30)..Lsn(0x50), delta3),
-                ], // delta layers
-                vec![(Lsn(0x10), img_layer)], // image layers
-                Lsn(0x50),
-            )
-            .await?;
-        {
-            tline
-                .latest_gc_cutoff_lsn
-                .lock_for_write()
-                .store_and_unlock(Lsn(0x30))
-                .wait()
-                .await;
-            // Update GC info
-            let mut guard = tline.gc_info.write().unwrap();
-            *guard = GcInfo {
-                retain_lsns: vec![
-                    (Lsn(0x10), tline.timeline_id, MaybeOffloaded::No),
-                    (Lsn(0x20), tline.timeline_id, MaybeOffloaded::No),
-                ],
-                cutoffs: GcCutoffs {
-                    time: Lsn(0x30),
-                    space: Lsn(0x30),
-                },
-                leases: Default::default(),
-                within_ancestor_pitr: false,
-            };
-        }
-
-        let expected_result = [
-            Bytes::from_static(b"value 0@0x10"),
-            Bytes::from_static(b"value 1@0x10@0x20@0x28@0x30@0x38"),
-            Bytes::from_static(b"value 2@0x10"),
-            Bytes::from_static(b"value 3@0x10"),
-            Bytes::from_static(b"value 4@0x10"),
-            Bytes::from_static(b"value 5@0x10"),
-            Bytes::from_static(b"value 6@0x10"),
-            Bytes::from_static(b"value 7@0x10"),
-            Bytes::from_static(b"value 8@0x10@0x48"),
-            Bytes::from_static(b"value 9@0x10@0x48"),
-        ];
-
-        let expected_result_at_gc_horizon = [
-            Bytes::from_static(b"value 0@0x10"),
-            Bytes::from_static(b"value 1@0x10@0x20@0x28@0x30"),
-            Bytes::from_static(b"value 2@0x10"),
-            Bytes::from_static(b"value 3@0x10"),
-            Bytes::from_static(b"value 4@0x10"),
-            Bytes::from_static(b"value 5@0x10"),
-            Bytes::from_static(b"value 6@0x10"),
-            Bytes::from_static(b"value 7@0x10"),
-            Bytes::from_static(b"value 8@0x10"),
-            Bytes::from_static(b"value 9@0x10"),
-        ];
-
-        let expected_result_at_lsn_20 = [
-            Bytes::from_static(b"value 0@0x10"),
-            Bytes::from_static(b"value 1@0x10@0x20"),
-            Bytes::from_static(b"value 2@0x10"),
-            Bytes::from_static(b"value 3@0x10"),
-            Bytes::from_static(b"value 4@0x10"),
-            Bytes::from_static(b"value 5@0x10"),
-            Bytes::from_static(b"value 6@0x10"),
-            Bytes::from_static(b"value 7@0x10"),
-            Bytes::from_static(b"value 8@0x10"),
-            Bytes::from_static(b"value 9@0x10"),
-        ];
-
-        let expected_result_at_lsn_10 = [
-            Bytes::from_static(b"value 0@0x10"),
-            Bytes::from_static(b"value 1@0x10"),
-            Bytes::from_static(b"value 2@0x10"),
-            Bytes::from_static(b"value 3@0x10"),
-            Bytes::from_static(b"value 4@0x10"),
-            Bytes::from_static(b"value 5@0x10"),
-            Bytes::from_static(b"value 6@0x10"),
-            Bytes::from_static(b"value 7@0x10"),
-            Bytes::from_static(b"value 8@0x10"),
-            Bytes::from_static(b"value 9@0x10"),
-        ];
-
-        let verify_result = || async {
-            let gc_horizon = {
-                let gc_info = tline.gc_info.read().unwrap();
-                gc_info.cutoffs.time
-            };
-            for idx in 0..10 {
-                assert_eq!(
-                    tline
-                        .get(get_key(idx as u32), Lsn(0x50), &ctx)
-                        .await
-                        .unwrap(),
-                    &expected_result[idx]
-                );
-                assert_eq!(
-                    tline
-                        .get(get_key(idx as u32), gc_horizon, &ctx)
-                        .await
-                        .unwrap(),
-                    &expected_result_at_gc_horizon[idx]
-                );
-                assert_eq!(
-                    tline
-                        .get(get_key(idx as u32), Lsn(0x20), &ctx)
-                        .await
-                        .unwrap(),
-                    &expected_result_at_lsn_20[idx]
-                );
-                assert_eq!(
-                    tline
-                        .get(get_key(idx as u32), Lsn(0x10), &ctx)
-                        .await
-                        .unwrap(),
-                    &expected_result_at_lsn_10[idx]
-                );
-            }
-        };
-
-        verify_result().await;
-
-        let cancel = CancellationToken::new();
-        tline
-            .compact_with_gc(
-                &cancel,
-                CompactOptions {
-                    compact_lsn_range: Some(CompactLsnRange::above(Lsn(0x28))),
-                    ..Default::default()
-                },
-                &ctx,
-            )
-            .await
-            .unwrap();
-        verify_result().await;
-
-        let all_layers = inspect_and_sort(&tline, Some(get_key(0)..get_key(10))).await;
-        check_layer_map_key_eq(
-            all_layers,
-            vec![
-                // The original image layer, not compacted
-                PersistentLayerKey {
-                    key_range: get_key(0)..get_key(10),
-                    lsn_range: Lsn(0x10)..Lsn(0x11),
-                    is_delta: false,
-                },
-                // Delta layer below the specified above_lsn not compacted
-                PersistentLayerKey {
-                    key_range: get_key(1)..get_key(2),
-                    lsn_range: Lsn(0x20)..Lsn(0x28),
-                    is_delta: true,
-                },
-                // Delta layer compacted above the LSN
-                PersistentLayerKey {
-                    key_range: get_key(1)..get_key(10),
-                    lsn_range: Lsn(0x28)..Lsn(0x50),
-                    is_delta: true,
-                },
-            ],
-        );
-
-        // compact again
-        tline
-            .compact_with_gc(&cancel, CompactOptions::default(), &ctx)
-            .await
-            .unwrap();
-        verify_result().await;
-
-        let all_layers = inspect_and_sort(&tline, Some(get_key(0)..get_key(10))).await;
-        check_layer_map_key_eq(
-            all_layers,
-            vec![
-                // The compacted image layer (full key range)
-                PersistentLayerKey {
-                    key_range: Key::MIN..Key::MAX,
-                    lsn_range: Lsn(0x10)..Lsn(0x11),
-                    is_delta: false,
-                },
-                // All other data in the delta layer
-                PersistentLayerKey {
-                    key_range: get_key(1)..get_key(10),
-                    lsn_range: Lsn(0x10)..Lsn(0x50),
-                    is_delta: true,
-                },
-            ],
-        );
-
-        Ok(())
-    }
-
-    #[cfg(feature = "testing")]
-    #[tokio::test]
-    async fn test_simple_bottom_most_compaction_rectangle() -> anyhow::Result<()> {
-        let harness = TenantHarness::create("test_simple_bottom_most_compaction_rectangle").await?;
-        let (tenant, ctx) = harness.load().await;
-
-        fn get_key(id: u32) -> Key {
-            // using aux key here b/c they are guaranteed to be inside `collect_keyspace`.
-            let mut key = Key::from_hex("620000000033333333444444445500000000").unwrap();
-            key.field6 = id;
-            key
-        }
-
-        let img_layer = (0..10)
-            .map(|id| (get_key(id), Bytes::from(format!("value {id}@0x10"))))
-            .collect_vec();
-
-        let delta1 = vec![(
-            get_key(1),
-            Lsn(0x20),
-            Value::WalRecord(NeonWalRecord::wal_append("@0x20")),
-        )];
-        let delta4 = vec![(
-            get_key(1),
-            Lsn(0x28),
-            Value::WalRecord(NeonWalRecord::wal_append("@0x28")),
-        )];
-        let delta2 = vec![
-            (
-                get_key(1),
-                Lsn(0x30),
-                Value::WalRecord(NeonWalRecord::wal_append("@0x30")),
-            ),
-            (
-                get_key(1),
-                Lsn(0x38),
-                Value::WalRecord(NeonWalRecord::wal_append("@0x38")),
-            ),
-        ];
-        let delta3 = vec![
-            (
-                get_key(8),
-                Lsn(0x48),
-                Value::WalRecord(NeonWalRecord::wal_append("@0x48")),
-            ),
-            (
-                get_key(9),
-                Lsn(0x48),
-                Value::WalRecord(NeonWalRecord::wal_append("@0x48")),
-            ),
-        ];
-
-        let tline = tenant
-            .create_test_timeline_with_layers(
-                TIMELINE_ID,
-                Lsn(0x10),
-                DEFAULT_PG_VERSION,
-                &ctx,
-                vec![
-                    // delta1/2/4 only contain a single key but multiple updates
-                    DeltaLayerTestDesc::new_with_inferred_key_range(Lsn(0x20)..Lsn(0x28), delta1),
-                    DeltaLayerTestDesc::new_with_inferred_key_range(Lsn(0x30)..Lsn(0x50), delta2),
-                    DeltaLayerTestDesc::new_with_inferred_key_range(Lsn(0x28)..Lsn(0x30), delta4),
-                    DeltaLayerTestDesc::new_with_inferred_key_range(Lsn(0x30)..Lsn(0x50), delta3),
-                ], // delta layers
-                vec![(Lsn(0x10), img_layer)], // image layers
-                Lsn(0x50),
-            )
-            .await?;
-        {
-            tline
-                .latest_gc_cutoff_lsn
-                .lock_for_write()
-                .store_and_unlock(Lsn(0x30))
-                .wait()
-                .await;
-            // Update GC info
-            let mut guard = tline.gc_info.write().unwrap();
-            *guard = GcInfo {
-                retain_lsns: vec![
-                    (Lsn(0x10), tline.timeline_id, MaybeOffloaded::No),
-                    (Lsn(0x20), tline.timeline_id, MaybeOffloaded::No),
-                ],
-                cutoffs: GcCutoffs {
-                    time: Lsn(0x30),
-                    space: Lsn(0x30),
-                },
-                leases: Default::default(),
-                within_ancestor_pitr: false,
-            };
-        }
-
-        let expected_result = [
-            Bytes::from_static(b"value 0@0x10"),
-            Bytes::from_static(b"value 1@0x10@0x20@0x28@0x30@0x38"),
-            Bytes::from_static(b"value 2@0x10"),
-            Bytes::from_static(b"value 3@0x10"),
-            Bytes::from_static(b"value 4@0x10"),
-            Bytes::from_static(b"value 5@0x10"),
-            Bytes::from_static(b"value 6@0x10"),
-            Bytes::from_static(b"value 7@0x10"),
-            Bytes::from_static(b"value 8@0x10@0x48"),
-            Bytes::from_static(b"value 9@0x10@0x48"),
-        ];
-
-        let expected_result_at_gc_horizon = [
-            Bytes::from_static(b"value 0@0x10"),
-            Bytes::from_static(b"value 1@0x10@0x20@0x28@0x30"),
-            Bytes::from_static(b"value 2@0x10"),
-            Bytes::from_static(b"value 3@0x10"),
-            Bytes::from_static(b"value 4@0x10"),
-            Bytes::from_static(b"value 5@0x10"),
-            Bytes::from_static(b"value 6@0x10"),
-            Bytes::from_static(b"value 7@0x10"),
-            Bytes::from_static(b"value 8@0x10"),
-            Bytes::from_static(b"value 9@0x10"),
-        ];
-
-        let expected_result_at_lsn_20 = [
-            Bytes::from_static(b"value 0@0x10"),
-            Bytes::from_static(b"value 1@0x10@0x20"),
-            Bytes::from_static(b"value 2@0x10"),
-            Bytes::from_static(b"value 3@0x10"),
-            Bytes::from_static(b"value 4@0x10"),
-            Bytes::from_static(b"value 5@0x10"),
-            Bytes::from_static(b"value 6@0x10"),
-            Bytes::from_static(b"value 7@0x10"),
-            Bytes::from_static(b"value 8@0x10"),
-            Bytes::from_static(b"value 9@0x10"),
-        ];
-
-        let expected_result_at_lsn_10 = [
-            Bytes::from_static(b"value 0@0x10"),
-            Bytes::from_static(b"value 1@0x10"),
-            Bytes::from_static(b"value 2@0x10"),
-            Bytes::from_static(b"value 3@0x10"),
-            Bytes::from_static(b"value 4@0x10"),
-            Bytes::from_static(b"value 5@0x10"),
-            Bytes::from_static(b"value 6@0x10"),
-            Bytes::from_static(b"value 7@0x10"),
-            Bytes::from_static(b"value 8@0x10"),
-            Bytes::from_static(b"value 9@0x10"),
-        ];
-
-        let verify_result = || async {
-            let gc_horizon = {
-                let gc_info = tline.gc_info.read().unwrap();
-                gc_info.cutoffs.time
-            };
-            for idx in 0..10 {
-                assert_eq!(
-                    tline
-                        .get(get_key(idx as u32), Lsn(0x50), &ctx)
-                        .await
-                        .unwrap(),
-                    &expected_result[idx]
-                );
-                assert_eq!(
-                    tline
-                        .get(get_key(idx as u32), gc_horizon, &ctx)
-                        .await
-                        .unwrap(),
-                    &expected_result_at_gc_horizon[idx]
-                );
-                assert_eq!(
-                    tline
-                        .get(get_key(idx as u32), Lsn(0x20), &ctx)
-                        .await
-                        .unwrap(),
-                    &expected_result_at_lsn_20[idx]
-                );
-                assert_eq!(
-                    tline
-                        .get(get_key(idx as u32), Lsn(0x10), &ctx)
-                        .await
-                        .unwrap(),
-                    &expected_result_at_lsn_10[idx]
-                );
-            }
-        };
-
-        verify_result().await;
-
-        let cancel = CancellationToken::new();
-
-        tline
-            .compact_with_gc(
-                &cancel,
-                CompactOptions {
-                    compact_key_range: Some((get_key(0)..get_key(2)).into()),
-                    compact_lsn_range: Some((Lsn(0x20)..Lsn(0x28)).into()),
-                    ..Default::default()
-                },
-                &ctx,
-            )
-            .await
-            .unwrap();
-        verify_result().await;
-
-        let all_layers = inspect_and_sort(&tline, Some(get_key(0)..get_key(10))).await;
-        check_layer_map_key_eq(
-            all_layers,
-            vec![
-                // The original image layer, not compacted
-                PersistentLayerKey {
-                    key_range: get_key(0)..get_key(10),
-                    lsn_range: Lsn(0x10)..Lsn(0x11),
-                    is_delta: false,
-                },
-                // According the selection logic, we select all layers with start key <= 0x28, so we would merge the layer 0x20-0x28 and
-                // the layer 0x28-0x30 into one.
-                PersistentLayerKey {
-                    key_range: get_key(1)..get_key(2),
-                    lsn_range: Lsn(0x20)..Lsn(0x30),
-                    is_delta: true,
-                },
-                // Above the upper bound and untouched
-                PersistentLayerKey {
-                    key_range: get_key(1)..get_key(2),
-                    lsn_range: Lsn(0x30)..Lsn(0x50),
-                    is_delta: true,
-                },
-                // This layer is untouched
-                PersistentLayerKey {
-                    key_range: get_key(8)..get_key(10),
-                    lsn_range: Lsn(0x30)..Lsn(0x50),
-                    is_delta: true,
-                },
-            ],
-        );
-
-        tline
-            .compact_with_gc(
-                &cancel,
-                CompactOptions {
-                    compact_key_range: Some((get_key(3)..get_key(8)).into()),
-                    compact_lsn_range: Some((Lsn(0x28)..Lsn(0x40)).into()),
-                    ..Default::default()
-                },
-                &ctx,
-            )
-            .await
-            .unwrap();
-        verify_result().await;
-
-        let all_layers = inspect_and_sort(&tline, Some(get_key(0)..get_key(10))).await;
-        check_layer_map_key_eq(
-            all_layers,
-            vec![
-                // The original image layer, not compacted
-                PersistentLayerKey {
-                    key_range: get_key(0)..get_key(10),
-                    lsn_range: Lsn(0x10)..Lsn(0x11),
-                    is_delta: false,
-                },
-                // Not in the compaction key range, uncompacted
-                PersistentLayerKey {
-                    key_range: get_key(1)..get_key(2),
-                    lsn_range: Lsn(0x20)..Lsn(0x30),
-                    is_delta: true,
-                },
-                // Not in the compaction key range, uncompacted but need rewrite because the delta layer overlaps with the range
-                PersistentLayerKey {
-                    key_range: get_key(1)..get_key(2),
-                    lsn_range: Lsn(0x30)..Lsn(0x50),
-                    is_delta: true,
-                },
-                // Note that when we specify the LSN upper bound to be 0x40, the compaction algorithm will not try to cut the layer
-                // horizontally in half. Instead, it will include all LSNs that overlap with 0x40. So the real max_lsn of the compaction
-                // becomes 0x50.
-                PersistentLayerKey {
-                    key_range: get_key(8)..get_key(10),
-                    lsn_range: Lsn(0x30)..Lsn(0x50),
-                    is_delta: true,
-                },
-            ],
-        );
-
-        // compact again
-        tline
-            .compact_with_gc(
-                &cancel,
-                CompactOptions {
-                    compact_key_range: Some((get_key(0)..get_key(5)).into()),
-                    compact_lsn_range: Some((Lsn(0x20)..Lsn(0x50)).into()),
-                    ..Default::default()
-                },
-                &ctx,
-            )
-            .await
-            .unwrap();
-        verify_result().await;
-
-        let all_layers = inspect_and_sort(&tline, Some(get_key(0)..get_key(10))).await;
-        check_layer_map_key_eq(
-            all_layers,
-            vec![
-                // The original image layer, not compacted
-                PersistentLayerKey {
-                    key_range: get_key(0)..get_key(10),
-                    lsn_range: Lsn(0x10)..Lsn(0x11),
-                    is_delta: false,
-                },
-                // The range gets compacted
-                PersistentLayerKey {
-                    key_range: get_key(1)..get_key(2),
-                    lsn_range: Lsn(0x20)..Lsn(0x50),
-                    is_delta: true,
-                },
-                // Not touched during this iteration of compaction
-                PersistentLayerKey {
-                    key_range: get_key(8)..get_key(10),
-                    lsn_range: Lsn(0x30)..Lsn(0x50),
-                    is_delta: true,
-                },
-            ],
-        );
-
-        // final full compaction
-        tline
-            .compact_with_gc(&cancel, CompactOptions::default(), &ctx)
-            .await
-            .unwrap();
-        verify_result().await;
-
-        let all_layers = inspect_and_sort(&tline, Some(get_key(0)..get_key(10))).await;
-        check_layer_map_key_eq(
-            all_layers,
-            vec![
-                // The compacted image layer (full key range)
-                PersistentLayerKey {
-                    key_range: Key::MIN..Key::MAX,
-                    lsn_range: Lsn(0x10)..Lsn(0x11),
-                    is_delta: false,
-                },
-                // All other data in the delta layer
-                PersistentLayerKey {
-                    key_range: get_key(1)..get_key(10),
-                    lsn_range: Lsn(0x10)..Lsn(0x50),
-                    is_delta: true,
-                },
-            ],
-        );
-
-        Ok(())
-    }
 }

pageserver/src/tenant/blob_io.rs

@@ -35,7 +35,7 @@ pub struct CompressionInfo {
     pub compressed_size: Option<usize>,
 }
 
-impl BlockCursor<'_> {
+impl<'a> BlockCursor<'a> {
     /// Read a blob into a new buffer.
     pub async fn read_blob(
         &self,

pageserver/src/tenant/block_io.rs

@@ -89,7 +89,7 @@ pub(crate) enum BlockReaderRef<'a> {
     VirtualFile(&'a VirtualFile),
 }
 
-impl BlockReaderRef<'_> {
+impl<'a> BlockReaderRef<'a> {
     #[inline(always)]
     async fn read_blk(
         &self,

pageserver/src/tenant/config.rs

@@ -11,7 +11,7 @@
 pub(crate) use pageserver_api::config::TenantConfigToml as TenantConf;
 use pageserver_api::models::CompactionAlgorithmSettings;
 use pageserver_api::models::EvictionPolicy;
-use pageserver_api::models::{self, TenantConfigPatch, ThrottleConfig};
+use pageserver_api::models::{self, ThrottleConfig};
 use pageserver_api::shard::{ShardCount, ShardIdentity, ShardNumber, ShardStripeSize};
 use serde::de::IntoDeserializer;
 use serde::{Deserialize, Serialize};
@@ -427,129 +427,6 @@ impl TenantConfOpt {
                 .or(global_conf.wal_receiver_protocol_override),
         }
     }
-
-    pub fn apply_patch(self, patch: TenantConfigPatch) -> anyhow::Result<TenantConfOpt> {
-        let Self {
-            mut checkpoint_distance,
-            mut checkpoint_timeout,
-            mut compaction_target_size,
-            mut compaction_period,
-            mut compaction_threshold,
-            mut compaction_algorithm,
-            mut gc_horizon,
-            mut gc_period,
-            mut image_creation_threshold,
-            mut pitr_interval,
-            mut walreceiver_connect_timeout,
-            mut lagging_wal_timeout,
-            mut max_lsn_wal_lag,
-            mut eviction_policy,
-            mut min_resident_size_override,
-            mut evictions_low_residence_duration_metric_threshold,
-            mut heatmap_period,
-            mut lazy_slru_download,
-            mut timeline_get_throttle,
-            mut image_layer_creation_check_threshold,
-            mut lsn_lease_length,
-            mut lsn_lease_length_for_ts,
-            mut timeline_offloading,
-            mut wal_receiver_protocol_override,
-        } = self;
-
-        patch.checkpoint_distance.apply(&mut checkpoint_distance);
-        patch
-            .checkpoint_timeout
-            .map(|v| humantime::parse_duration(&v))?
-            .apply(&mut checkpoint_timeout);
-        patch
-            .compaction_target_size
-            .apply(&mut compaction_target_size);
-        patch
-            .compaction_period
-            .map(|v| humantime::parse_duration(&v))?
-            .apply(&mut compaction_period);
-        patch.compaction_threshold.apply(&mut compaction_threshold);
-        patch.compaction_algorithm.apply(&mut compaction_algorithm);
-        patch.gc_horizon.apply(&mut gc_horizon);
-        patch
-            .gc_period
-            .map(|v| humantime::parse_duration(&v))?
-            .apply(&mut gc_period);
-        patch
-            .image_creation_threshold
-            .apply(&mut image_creation_threshold);
-        patch
-            .pitr_interval
-            .map(|v| humantime::parse_duration(&v))?
-            .apply(&mut pitr_interval);
-        patch
-            .walreceiver_connect_timeout
-            .map(|v| humantime::parse_duration(&v))?
-            .apply(&mut walreceiver_connect_timeout);
-        patch
-            .lagging_wal_timeout
-            .map(|v| humantime::parse_duration(&v))?
-            .apply(&mut lagging_wal_timeout);
-        patch.max_lsn_wal_lag.apply(&mut max_lsn_wal_lag);
-        patch.eviction_policy.apply(&mut eviction_policy);
-        patch
-            .min_resident_size_override
-            .apply(&mut min_resident_size_override);
-        patch
-            .evictions_low_residence_duration_metric_threshold
-            .map(|v| humantime::parse_duration(&v))?
-            .apply(&mut evictions_low_residence_duration_metric_threshold);
-        patch
-            .heatmap_period
-            .map(|v| humantime::parse_duration(&v))?
-            .apply(&mut heatmap_period);
-        patch.lazy_slru_download.apply(&mut lazy_slru_download);
-        patch
-            .timeline_get_throttle
-            .apply(&mut timeline_get_throttle);
-        patch
-            .image_layer_creation_check_threshold
-            .apply(&mut image_layer_creation_check_threshold);
-        patch
-            .lsn_lease_length
-            .map(|v| humantime::parse_duration(&v))?
-            .apply(&mut lsn_lease_length);
-        patch
-            .lsn_lease_length_for_ts
-            .map(|v| humantime::parse_duration(&v))?
-            .apply(&mut lsn_lease_length_for_ts);
-        patch.timeline_offloading.apply(&mut timeline_offloading);
-        patch
-            .wal_receiver_protocol_override
-            .apply(&mut wal_receiver_protocol_override);
-
-        Ok(Self {
-            checkpoint_distance,
-            checkpoint_timeout,
-            compaction_target_size,
-            compaction_period,
-            compaction_threshold,
-            compaction_algorithm,
-            gc_horizon,
-            gc_period,
-            image_creation_threshold,
-            pitr_interval,
-            walreceiver_connect_timeout,
-            lagging_wal_timeout,
-            max_lsn_wal_lag,
-            eviction_policy,
-            min_resident_size_override,
-            evictions_low_residence_duration_metric_threshold,
-            heatmap_period,
-            lazy_slru_download,
-            timeline_get_throttle,
-            image_layer_creation_check_threshold,
-            lsn_lease_length,
-            lsn_lease_length_for_ts,
-            timeline_offloading,
-            wal_receiver_protocol_override,
-        })
-    }
 }
 
 impl TryFrom<&'_ models::TenantConfig> for TenantConfOpt {

pageserver/src/tenant/disk_btree.rs

@@ -532,7 +532,7 @@ pub struct DiskBtreeIterator<'a> {
     >,
 }
 
-impl DiskBtreeIterator<'_> {
+impl<'a> DiskBtreeIterator<'a> {
     pub async fn next(&mut self) -> Option<std::result::Result<(Vec<u8>, u64), DiskBtreeError>> {
         self.stream.next().await
     }

pageserver/src/tenant/ephemeral_file.rs

@@ -174,11 +174,11 @@ impl EphemeralFile {
 }
 
 impl super::storage_layer::inmemory_layer::vectored_dio_read::File for EphemeralFile {
-    async fn read_exact_at_eof_ok<B: IoBufAlignedMut + Send>(
-        &self,
+    async fn read_exact_at_eof_ok<'a, 'b, B: IoBufAlignedMut + Send>(
+        &'b self,
         start: u64,
         dst: tokio_epoll_uring::Slice<B>,
-        ctx: &RequestContext,
+        ctx: &'a RequestContext,
     ) -> std::io::Result<(tokio_epoll_uring::Slice<B>, usize)> {
         let submitted_offset = self.buffered_writer.bytes_submitted();
 

pageserver/src/tenant/layer_map.rs

@@ -392,8 +392,8 @@ impl LayerMap {
         image_layer: Option<Arc<PersistentLayerDesc>>,
         end_lsn: Lsn,
     ) -> Option<SearchResult> {
-        assert!(delta_layer.as_ref().is_none_or(|l| l.is_delta()));
-        assert!(image_layer.as_ref().is_none_or(|l| !l.is_delta()));
+        assert!(delta_layer.as_ref().map_or(true, |l| l.is_delta()));
+        assert!(image_layer.as_ref().map_or(true, |l| !l.is_delta()));
 
         match (delta_layer, image_layer) {
             (None, None) => None,

pageserver/src/tenant/remote_timeline_client.rs

@@ -749,7 +749,7 @@ impl RemoteTimelineClient {
         // ahead of what's _actually_ on the remote during index upload.
         upload_queue.dirty.metadata = metadata.clone();
 
-        self.schedule_index_upload(upload_queue);
+        self.schedule_index_upload(upload_queue)?;
 
         Ok(())
     }
@@ -770,7 +770,7 @@ impl RemoteTimelineClient {
 
         upload_queue.dirty.metadata.apply(update);
 
-        self.schedule_index_upload(upload_queue);
+        self.schedule_index_upload(upload_queue)?;
 
         Ok(())
     }
@@ -809,7 +809,7 @@ impl RemoteTimelineClient {
         if let Some(archived_at_set) = need_upload_scheduled {
             let intended_archived_at = archived_at_set.then(|| Utc::now().naive_utc());
             upload_queue.dirty.archived_at = intended_archived_at;
-            self.schedule_index_upload(upload_queue);
+            self.schedule_index_upload(upload_queue)?;
         }
 
         let need_wait = need_change(&upload_queue.clean.0.archived_at, state).is_some();
@@ -824,7 +824,7 @@ impl RemoteTimelineClient {
         let mut guard = self.upload_queue.lock().unwrap();
         let upload_queue = guard.initialized_mut()?;
         upload_queue.dirty.import_pgdata = state;
-        self.schedule_index_upload(upload_queue);
+        self.schedule_index_upload(upload_queue)?;
         Ok(())
     }
 
@@ -843,14 +843,17 @@ impl RemoteTimelineClient {
         let upload_queue = guard.initialized_mut()?;
 
         if upload_queue.latest_files_changes_since_metadata_upload_scheduled > 0 {
-            self.schedule_index_upload(upload_queue);
+            self.schedule_index_upload(upload_queue)?;
         }
 
         Ok(())
     }
 
     /// Launch an index-file upload operation in the background (internal function)
-    fn schedule_index_upload(self: &Arc<Self>, upload_queue: &mut UploadQueueInitialized) {
+    fn schedule_index_upload(
+        self: &Arc<Self>,
+        upload_queue: &mut UploadQueueInitialized,
+    ) -> Result<(), NotInitialized> {
         let disk_consistent_lsn = upload_queue.dirty.metadata.disk_consistent_lsn();
         // fix up the duplicated field
         upload_queue.dirty.disk_consistent_lsn = disk_consistent_lsn;
@@ -877,6 +880,7 @@ impl RemoteTimelineClient {
 
         // Launch the task immediately, if possible
         self.launch_queued_tasks(upload_queue);
+        Ok(())
     }
 
     /// Reparent this timeline to a new parent.
@@ -905,7 +909,7 @@ impl RemoteTimelineClient {
                 upload_queue.dirty.metadata.reparent(new_parent);
                 upload_queue.dirty.lineage.record_previous_ancestor(&prev);
 
-                self.schedule_index_upload(upload_queue);
+                self.schedule_index_upload(upload_queue)?;
 
                 Some(self.schedule_barrier0(upload_queue))
             }
@@ -944,7 +948,7 @@ impl RemoteTimelineClient {
                     assert!(prev.is_none(), "copied layer existed already {layer}");
                 }
 
-                self.schedule_index_upload(upload_queue);
+                self.schedule_index_upload(upload_queue)?;
 
                 Some(self.schedule_barrier0(upload_queue))
             }
@@ -1000,7 +1004,7 @@ impl RemoteTimelineClient {
                     upload_queue.dirty.gc_blocking = current
                         .map(|x| x.with_reason(reason))
                         .or_else(|| Some(index::GcBlocking::started_now_for(reason)));
-                    self.schedule_index_upload(upload_queue);
+                    self.schedule_index_upload(upload_queue)?;
                     Some(self.schedule_barrier0(upload_queue))
                 }
             }
@@ -1053,7 +1057,8 @@ impl RemoteTimelineClient {
                     upload_queue.dirty.gc_blocking =
                         current.as_ref().and_then(|x| x.without_reason(reason));
                     assert!(wanted(upload_queue.dirty.gc_blocking.as_ref()));
-                    self.schedule_index_upload(upload_queue);
+                    // FIXME: bogus ?
+                    self.schedule_index_upload(upload_queue)?;
                     Some(self.schedule_barrier0(upload_queue))
                 }
             }
@@ -1120,8 +1125,8 @@ impl RemoteTimelineClient {
         let mut guard = self.upload_queue.lock().unwrap();
         let upload_queue = guard.initialized_mut()?;
 
-        let with_metadata =
-            self.schedule_unlinking_of_layers_from_index_part0(upload_queue, names.iter().cloned());
+        let with_metadata = self
+            .schedule_unlinking_of_layers_from_index_part0(upload_queue, names.iter().cloned())?;
 
         self.schedule_deletion_of_unlinked0(upload_queue, with_metadata);
 
@@ -1148,7 +1153,7 @@ impl RemoteTimelineClient {
 
         let names = gc_layers.iter().map(|x| x.layer_desc().layer_name());
 
-        self.schedule_unlinking_of_layers_from_index_part0(upload_queue, names);
+        self.schedule_unlinking_of_layers_from_index_part0(upload_queue, names)?;
 
         self.launch_queued_tasks(upload_queue);
 
@@ -1161,7 +1166,7 @@ impl RemoteTimelineClient {
         self: &Arc<Self>,
         upload_queue: &mut UploadQueueInitialized,
         names: I,
-    ) -> Vec<(LayerName, LayerFileMetadata)>
+    ) -> Result<Vec<(LayerName, LayerFileMetadata)>, NotInitialized>
     where
         I: IntoIterator<Item = LayerName>,
     {
@@ -1203,10 +1208,10 @@ impl RemoteTimelineClient {
         // index_part update, because that needs to be uploaded before we can actually delete the
         // files.
         if upload_queue.latest_files_changes_since_metadata_upload_scheduled > 0 {
-            self.schedule_index_upload(upload_queue);
+            self.schedule_index_upload(upload_queue)?;
         }
 
-        with_metadata
+        Ok(with_metadata)
     }
 
     /// Schedules deletion for layer files which have previously been unlinked from the
@@ -1297,7 +1302,7 @@ impl RemoteTimelineClient {
 
         let names = compacted_from.iter().map(|x| x.layer_desc().layer_name());
 
-        self.schedule_unlinking_of_layers_from_index_part0(upload_queue, names);
+        self.schedule_unlinking_of_layers_from_index_part0(upload_queue, names)?;
         self.launch_queued_tasks(upload_queue);
 
         Ok(())

pageserver/src/tenant/remote_timeline_client/download.rs

@@ -145,8 +145,8 @@ pub async fn download_layer_file<'a>(
 ///
 /// If Err() is returned, there was some error. The file at `dst_path` has been unlinked.
 /// The unlinking has _not_ been made durable.
-async fn download_object(
-    storage: &GenericRemoteStorage,
+async fn download_object<'a>(
+    storage: &'a GenericRemoteStorage,
     src_path: &RemotePath,
     dst_path: &Utf8PathBuf,
     #[cfg_attr(target_os = "macos", allow(unused_variables))] gate: &utils::sync::gate::Gate,

pageserver/src/tenant/remote_timeline_client/upload.rs

@@ -25,8 +25,8 @@ use utils::id::{TenantId, TimelineId};
 use tracing::info;
 
 /// Serializes and uploads the given index part data to the remote storage.
-pub(crate) async fn upload_index_part(
-    storage: &GenericRemoteStorage,
+pub(crate) async fn upload_index_part<'a>(
+    storage: &'a GenericRemoteStorage,
     tenant_shard_id: &TenantShardId,
     timeline_id: &TimelineId,
     generation: Generation,

pageserver/src/tenant/storage_layer.rs

@@ -345,7 +345,10 @@ impl LayerFringe {
     }
 
     pub(crate) fn next_layer(&mut self) -> Option<(ReadableLayer, KeySpace, Range<Lsn>)> {
-        let read_desc = self.planned_visits_by_lsn.pop()?;
+        let read_desc = match self.planned_visits_by_lsn.pop() {
+            Some(desc) => desc,
+            None => return None,
+        };
 
         let removed = self.visit_reads.remove_entry(&read_desc.layer_to_visit_id);
 

pageserver/src/tenant/storage_layer/delta_layer.rs

@@ -1486,7 +1486,7 @@ pub struct ValueRef<'a> {
     layer: &'a DeltaLayerInner,
 }
 
-impl ValueRef<'_> {
+impl<'a> ValueRef<'a> {
     /// Loads the value from disk
     pub async fn load(&self, ctx: &RequestContext) -> Result<Value> {
         let buf = self.load_raw(ctx).await?;
@@ -1543,7 +1543,7 @@ pub struct DeltaLayerIterator<'a> {
     is_end: bool,
 }
 
-impl DeltaLayerIterator<'_> {
+impl<'a> DeltaLayerIterator<'a> {
     pub(crate) fn layer_dbg_info(&self) -> String {
         self.delta_layer.layer_dbg_info()
     }

pageserver/src/tenant/storage_layer/image_layer.rs

@@ -1052,7 +1052,7 @@ pub struct ImageLayerIterator<'a> {
     is_end: bool,
 }
 
-impl ImageLayerIterator<'_> {
+impl<'a> ImageLayerIterator<'a> {
     pub(crate) fn layer_dbg_info(&self) -> String {
         self.image_layer.layer_dbg_info()
     }

pageserver/src/tenant/storage_layer/inmemory_layer/vectored_dio_read.rs

@@ -25,11 +25,11 @@ pub trait File: Send {
     /// [`std::io::ErrorKind::UnexpectedEof`] error if the file is shorter than `start+dst.len()`.
     ///
     /// No guarantees are made about the remaining bytes in `dst` in case of a short read.
-    async fn read_exact_at_eof_ok<B: IoBufAlignedMut + Send>(
-        &self,
+    async fn read_exact_at_eof_ok<'a, 'b, B: IoBufAlignedMut + Send>(
+        &'b self,
         start: u64,
         dst: Slice<B>,
-        ctx: &RequestContext,
+        ctx: &'a RequestContext,
     ) -> std::io::Result<(Slice<B>, usize)>;
 }
 
@@ -479,11 +479,11 @@ mod tests {
     }
 
     impl File for InMemoryFile {
-        async fn read_exact_at_eof_ok<B: IoBufMut + Send>(
-            &self,
+        async fn read_exact_at_eof_ok<'a, 'b, B: IoBufMut + Send>(
+            &'b self,
             start: u64,
             mut dst: Slice<B>,
-            _ctx: &RequestContext,
+            _ctx: &'a RequestContext,
         ) -> std::io::Result<(Slice<B>, usize)> {
             let dst_slice: &mut [u8] = dst.as_mut_rust_slice_full_zeroed();
             let nread = {
@@ -609,12 +609,12 @@ mod tests {
         }
     }
 
-    impl File for RecorderFile<'_> {
-        async fn read_exact_at_eof_ok<B: IoBufAlignedMut + Send>(
-            &self,
+    impl<'x> File for RecorderFile<'x> {
+        async fn read_exact_at_eof_ok<'a, 'b, B: IoBufAlignedMut + Send>(
+            &'b self,
             start: u64,
             dst: Slice<B>,
-            ctx: &RequestContext,
+            ctx: &'a RequestContext,
         ) -> std::io::Result<(Slice<B>, usize)> {
             let (dst, nread) = self.file.read_exact_at_eof_ok(start, dst, ctx).await?;
             self.recorded.borrow_mut().push(RecordedRead {
@@ -740,11 +740,11 @@ mod tests {
     }
 
     impl File for MockFile {
-        async fn read_exact_at_eof_ok<B: IoBufMut + Send>(
-            &self,
+        async fn read_exact_at_eof_ok<'a, 'b, B: IoBufMut + Send>(
+            &'b self,
             start: u64,
             mut dst: Slice<B>,
-            _ctx: &RequestContext,
+            _ctx: &'a RequestContext,
         ) -> std::io::Result<(Slice<B>, usize)> {
             let ExpectedRead {
                 expect_pos,

pageserver/src/tenant/throttle.rs

@@ -58,11 +58,6 @@ pub struct Stats {
     pub sum_throttled_usecs: u64,
 }
 
-pub enum ThrottleResult {
-    NotThrottled { start: Instant },
-    Throttled { start: Instant, end: Instant },
-}
-
 impl<M> Throttle<M>
 where
     M: Metric,
@@ -127,15 +122,15 @@ where
         self.inner.load().rate_limiter.steady_rps()
     }
 
-    pub async fn throttle(&self, key_count: usize) -> ThrottleResult {
+    pub async fn throttle(&self, key_count: usize) -> Option<Duration> {
         let inner = self.inner.load_full(); // clones the `Inner` Arc
 
-        let start = std::time::Instant::now();
-
         if !inner.enabled {
-            return ThrottleResult::NotThrottled { start };
+            return None;
         }
 
+        let start = std::time::Instant::now();
+
         self.metric.accounting_start();
         self.count_accounted_start.fetch_add(1, Ordering::Relaxed);
         let did_throttle = inner.rate_limiter.acquire(key_count).await;
@@ -150,9 +145,9 @@ where
                 .fetch_add(wait_time.as_micros() as u64, Ordering::Relaxed);
             let observation = Observation { wait_time };
             self.metric.observe_throttling(&observation);
-            ThrottleResult::Throttled { start, end: now }
+            Some(wait_time)
         } else {
-            ThrottleResult::NotThrottled { start }
+            None
         }
     }
 }

pageserver/src/tenant/timeline.rs

@@ -144,15 +144,19 @@ use self::layer_manager::LayerManager;
 use self::logical_size::LogicalSize;
 use self::walreceiver::{WalReceiver, WalReceiverConf};
 
-use super::config::TenantConf;
-use super::remote_timeline_client::index::IndexPart;
-use super::remote_timeline_client::RemoteTimelineClient;
-use super::secondary::heatmap::{HeatMapLayer, HeatMapTimeline};
-use super::storage_layer::{LayerFringe, LayerVisibilityHint, ReadableLayer};
-use super::upload_queue::NotInitialized;
-use super::GcError;
 use super::{
-    debug_assert_current_span_has_tenant_and_timeline_id, AttachedTenantConf, MaybeOffloaded,
+    config::TenantConf, storage_layer::LayerVisibilityHint, upload_queue::NotInitialized,
+    MaybeOffloaded,
+};
+use super::{debug_assert_current_span_has_tenant_and_timeline_id, AttachedTenantConf};
+use super::{remote_timeline_client::index::IndexPart, storage_layer::LayerFringe};
+use super::{
+    remote_timeline_client::RemoteTimelineClient, remote_timeline_client::WaitCompletionError,
+    storage_layer::ReadableLayer,
+};
+use super::{
+    secondary::heatmap::{HeatMapLayer, HeatMapTimeline},
+    GcError,
 };
 
 #[cfg(test)]
@@ -776,90 +780,46 @@ pub(crate) enum CompactFlags {
 #[serde_with::serde_as]
 #[derive(Debug, Clone, serde::Deserialize)]
 pub(crate) struct CompactRequest {
-    pub compact_key_range: Option<CompactKeyRange>,
-    pub compact_lsn_range: Option<CompactLsnRange>,
+    pub compact_range: Option<CompactRange>,
+    pub compact_below_lsn: Option<Lsn>,
     /// Whether the compaction job should be scheduled.
     #[serde(default)]
     pub scheduled: bool,
     /// Whether the compaction job should be split across key ranges.
     #[serde(default)]
     pub sub_compaction: bool,
-    /// Max job size for each subcompaction job.
-    pub sub_compaction_max_job_size_mb: Option<u64>,
 }
 
 #[serde_with::serde_as]
 #[derive(Debug, Clone, serde::Deserialize)]
-pub(crate) struct CompactLsnRange {
-    pub start: Lsn,
-    pub end: Lsn,
-}
-
-#[serde_with::serde_as]
-#[derive(Debug, Clone, serde::Deserialize)]
-pub(crate) struct CompactKeyRange {
+pub(crate) struct CompactRange {
     #[serde_as(as = "serde_with::DisplayFromStr")]
     pub start: Key,
     #[serde_as(as = "serde_with::DisplayFromStr")]
     pub end: Key,
 }
 
-impl From<Range<Lsn>> for CompactLsnRange {
-    fn from(range: Range<Lsn>) -> Self {
-        Self {
-            start: range.start,
-            end: range.end,
-        }
-    }
-}
-
-impl From<Range<Key>> for CompactKeyRange {
+impl From<Range<Key>> for CompactRange {
     fn from(range: Range<Key>) -> Self {
-        Self {
+        CompactRange {
             start: range.start,
             end: range.end,
         }
     }
 }
 
-impl From<CompactLsnRange> for Range<Lsn> {
-    fn from(range: CompactLsnRange) -> Self {
-        range.start..range.end
-    }
-}
-
-impl From<CompactKeyRange> for Range<Key> {
-    fn from(range: CompactKeyRange) -> Self {
-        range.start..range.end
-    }
-}
-
-impl CompactLsnRange {
-    #[cfg(test)]
-    #[cfg(feature = "testing")]
-    pub fn above(lsn: Lsn) -> Self {
-        Self {
-            start: lsn,
-            end: Lsn::MAX,
-        }
-    }
-}
-
 #[derive(Debug, Clone, Default)]
 pub(crate) struct CompactOptions {
     pub flags: EnumSet<CompactFlags>,
     /// If set, the compaction will only compact the key range specified by this option.
-    /// This option is only used by GC compaction. For the full explanation, see [`compaction::GcCompactJob`].
-    pub compact_key_range: Option<CompactKeyRange>,
-    /// If set, the compaction will only compact the LSN within this value.
-    /// This option is only used by GC compaction. For the full explanation, see [`compaction::GcCompactJob`].
-    pub compact_lsn_range: Option<CompactLsnRange>,
+    /// This option is only used by GC compaction.
+    pub compact_range: Option<CompactRange>,
+    /// If set, the compaction will only compact the LSN below this value.
+    /// This option is only used by GC compaction.
+    pub compact_below_lsn: Option<Lsn>,
     /// Enable sub-compaction (split compaction job across key ranges).
     /// This option is only used by GC compaction.
     pub sub_compaction: bool,
-    /// Set job size for the GC compaction.
-    /// This option is only used by GC compaction.
-    pub sub_compaction_max_job_size_mb: Option<u64>,
 }
 
 impl std::fmt::Debug for Timeline {
@@ -1681,10 +1641,9 @@ impl Timeline {
             cancel,
             CompactOptions {
                 flags,
-                compact_key_range: None,
-                compact_lsn_range: None,
+                compact_range: None,
+                compact_below_lsn: None,
                 sub_compaction: false,
-                sub_compaction_max_job_size_mb: None,
             },
             ctx,
         )
@@ -3893,6 +3852,24 @@ impl Timeline {
             // release lock on 'layers'
         };
 
+        // Backpressure mechanism: wait with continuation of the flush loop until we have uploaded all layer files.
+        // This makes us refuse ingest until the new layers have been persisted to the remote
+        let start = Instant::now();
+        self.remote_client
+            .wait_completion()
+            .await
+            .map_err(|e| match e {
+                WaitCompletionError::UploadQueueShutDownOrStopped
+                | WaitCompletionError::NotInitialized(
+                    NotInitialized::ShuttingDown | NotInitialized::Stopped,
+                ) => FlushLayerError::Cancelled,
+                WaitCompletionError::NotInitialized(NotInitialized::Uninitialized) => {
+                    FlushLayerError::Other(anyhow!(e).into())
+                }
+            })?;
+        let duration = start.elapsed().as_secs_f64();
+        self.metrics.flush_wait_upload_time_gauge_add(duration);
+
         // FIXME: between create_delta_layer and the scheduling of the upload in `update_metadata_file`,
         // a compaction can delete the file and then it won't be available for uploads any more.
         // We still schedule the upload, resulting in an error, but ideally we'd somehow avoid this
@@ -4042,11 +4019,8 @@ impl Timeline {
             // NB: there are two callers, one is the compaction task, of which there is only one per struct Tenant and hence Timeline.
             // The other is the initdb optimization in flush_frozen_layer, used by `boostrap_timeline`, which runs before `.activate()`
             // and hence before the compaction task starts.
-            // Note that there are a third "caller" that will take the `partitioning` lock. It is `gc_compaction_split_jobs` for
-            // gc-compaction where it uses the repartition data to determine the split jobs. In the future, it might use its own
-            // heuristics, but for now, we should allow concurrent access to it and let the caller retry compaction.
             return Err(CompactionError::Other(anyhow!(
-                "repartition() called concurrently, this is rare and a retry should be fine"
+                "repartition() called concurrently, this should not happen"
             )));
         };
         let ((dense_partition, sparse_partition), partition_lsn) = &*partitioning_guard;
@@ -5842,7 +5816,7 @@ enum OpenLayerAction {
     None,
 }
 
-impl TimelineWriter<'_> {
+impl<'a> TimelineWriter<'a> {
     async fn handle_open_layer_action(
         &mut self,
         at: Lsn,

pageserver/src/tenant/timeline/compaction.rs

@@ -10,8 +10,8 @@ use std::sync::Arc;
 
 use super::layer_manager::LayerManager;
 use super::{
-    CompactFlags, CompactOptions, CreateImageLayersError, DurationRecorder, ImageLayerCreationMode,
-    RecordedDuration, Timeline,
+    CompactFlags, CompactOptions, CompactRange, CreateImageLayersError, DurationRecorder,
+    ImageLayerCreationMode, RecordedDuration, Timeline,
 };
 
 use anyhow::{anyhow, bail, Context};
@@ -64,9 +64,6 @@ const COMPACTION_DELTA_THRESHOLD: usize = 5;
 
 /// A scheduled compaction task.
 pub(crate) struct ScheduledCompactionTask {
-    /// It's unfortunate that we need to store a compact options struct here because the only outer
-    /// API we can call here is `compact_with_options` which does a few setup calls before starting the
-    /// actual compaction job... We should refactor this to store `GcCompactionJob` in the future.
     pub options: CompactOptions,
     /// The channel to send the compaction result. If this is a subcompaction, the last compaction job holds the sender.
     pub result_tx: Option<tokio::sync::oneshot::Sender<()>>,
@@ -74,57 +71,16 @@ pub(crate) struct ScheduledCompactionTask {
     pub gc_block: Option<gc_block::Guard>,
 }
 
-/// A job description for the gc-compaction job. This structure describes the rectangle range that the job will
-/// process. The exact layers that need to be compacted/rewritten will be generated when `compact_with_gc` gets
-/// called.
-#[derive(Debug, Clone)]
-pub(crate) struct GcCompactJob {
-    pub dry_run: bool,
-    /// The key range to be compacted. The compaction algorithm will only regenerate key-value pairs within this range
-    /// [left inclusive, right exclusive), and other pairs will be rewritten into new files if necessary.
-    pub compact_key_range: Range<Key>,
-    /// The LSN range to be compacted. The compaction algorithm will use this range to determine the layers to be
-    /// selected for the compaction, and it does not guarantee the generated layers will have exactly the same LSN range
-    /// as specified here. The true range being compacted is `min_lsn/max_lsn` in [`GcCompactionJobDescription`].
-    /// min_lsn will always <= the lower bound specified here, and max_lsn will always >= the upper bound specified here.
-    pub compact_lsn_range: Range<Lsn>,
-}
-
-impl GcCompactJob {
-    pub fn from_compact_options(options: CompactOptions) -> Self {
-        GcCompactJob {
-            dry_run: options.flags.contains(CompactFlags::DryRun),
-            compact_key_range: options
-                .compact_key_range
-                .map(|x| x.into())
-                .unwrap_or(Key::MIN..Key::MAX),
-            compact_lsn_range: options
-                .compact_lsn_range
-                .map(|x| x.into())
-                .unwrap_or(Lsn::INVALID..Lsn::MAX),
-        }
-    }
-}
-
-/// A job description for the gc-compaction job. This structure is generated when `compact_with_gc` is called
-/// and contains the exact layers we want to compact.
 pub struct GcCompactionJobDescription {
     /// All layers to read in the compaction job
     selected_layers: Vec<Layer>,
-    /// GC cutoff of the job. This is the lowest LSN that will be accessed by the read/GC path and we need to
-    /// keep all deltas <= this LSN or generate an image == this LSN.
+    /// GC cutoff of the job
     gc_cutoff: Lsn,
-    /// LSNs to retain for the job. Read path will use this LSN so we need to keep deltas <= this LSN or
-    /// generate an image == this LSN.
+    /// LSNs to retain for the job
     retain_lsns_below_horizon: Vec<Lsn>,
-    /// Maximum layer LSN processed in this compaction, that is max(end_lsn of layers). Exclusive. All data
-    /// \>= this LSN will be kept and will not be rewritten.
+    /// Maximum layer LSN processed in this compaction
     max_layer_lsn: Lsn,
-    /// Minimum layer LSN processed in this compaction, that is min(start_lsn of layers). Inclusive.
-    /// All access below (strict lower than `<`) this LSN will be routed through the normal read path instead of
-    /// k-merge within gc-compaction.
-    min_layer_lsn: Lsn,
-    /// Only compact layers overlapping with this range.
+    /// Only compact layers overlapping with this range
     compaction_key_range: Range<Key>,
     /// When partial compaction is enabled, these layers need to be rewritten to ensure no overlap.
     /// This field is here solely for debugging. The field will not be read once the compaction
@@ -343,7 +299,7 @@ impl Timeline {
             )));
         }
 
-        if options.compact_key_range.is_some() || options.compact_lsn_range.is_some() {
+        if options.compact_range.is_some() {
             // maybe useful in the future? could implement this at some point
             return Err(CompactionError::Other(anyhow!(
                 "compaction range is not supported for legacy compaction for now"
@@ -1110,7 +1066,7 @@ impl Timeline {
                 return Err(CompactionError::ShuttingDown);
             }
 
-            let same_key = prev_key == Some(key);
+            let same_key = prev_key.map_or(false, |prev_key| prev_key == key);
             // We need to check key boundaries once we reach next key or end of layer with the same key
             if !same_key || lsn == dup_end_lsn {
                 let mut next_key_size = 0u64;
@@ -1798,35 +1754,32 @@ impl Timeline {
         Ok(())
     }
 
-    /// Split a gc-compaction job into multiple compaction jobs. The split is based on the key range and the estimated size of the compaction job.
-    /// The function returns a list of compaction jobs that can be executed separately. If the upper bound of the compact LSN
-    /// range is not specified, we will use the latest gc_cutoff as the upper bound, so that all jobs in the jobset acts
-    /// like a full compaction of the specified keyspace.
+    /// Split a gc-compaction job into multiple compaction jobs. Optimally, this function should return a vector of
+    /// `GcCompactionJobDesc`. But we want to keep it simple on the tenant scheduling side without exposing too much
+    /// ad-hoc information about gc compaction itself.
     pub(crate) async fn gc_compaction_split_jobs(
         self: &Arc<Self>,
-        job: GcCompactJob,
-        sub_compaction_max_job_size_mb: Option<u64>,
-    ) -> anyhow::Result<Vec<GcCompactJob>> {
-        let compact_below_lsn = if job.compact_lsn_range.end != Lsn::MAX {
-            job.compact_lsn_range.end
+        options: CompactOptions,
+    ) -> anyhow::Result<Vec<CompactOptions>> {
+        if !options.sub_compaction {
+            return Ok(vec![options]);
+        }
+        let compact_range = options.compact_range.clone().unwrap_or(CompactRange {
+            start: Key::MIN,
+            end: Key::MAX,
+        });
+        let compact_below_lsn = if let Some(compact_below_lsn) = options.compact_below_lsn {
+            compact_below_lsn
         } else {
             *self.get_latest_gc_cutoff_lsn() // use the real gc cutoff
         };
-
-        // Split compaction job to about 4GB each
-        const GC_COMPACT_MAX_SIZE_MB: u64 = 4 * 1024;
-        let sub_compaction_max_job_size_mb =
-            sub_compaction_max_job_size_mb.unwrap_or(GC_COMPACT_MAX_SIZE_MB);
-
         let mut compact_jobs = Vec::new();
         // For now, we simply use the key partitioning information; we should do a more fine-grained partitioning
         // by estimating the amount of files read for a compaction job. We should also partition on LSN.
-        let ((dense_ks, sparse_ks), _) = {
-            let Ok(partition) = self.partitioning.try_lock() else {
-                bail!("failed to acquire partition lock");
-            };
-            partition.clone()
+        let Ok(partition) = self.partitioning.try_lock() else {
+            bail!("failed to acquire partition lock");
         };
+        let ((dense_ks, sparse_ks), _) = &*partition;
         // Truncate the key range to be within user specified compaction range.
         fn truncate_to(
             source_start: &Key,
@@ -1855,8 +1808,8 @@ impl Timeline {
             let Some((start, end)) = truncate_to(
                 &range.start,
                 &range.end,
-                &job.compact_key_range.start,
-                &job.compact_key_range.end,
+                &compact_range.start,
+                &compact_range.end,
             ) else {
                 continue;
             };
@@ -1866,6 +1819,8 @@ impl Timeline {
         let guard = self.layers.read().await;
         let layer_map = guard.layer_map()?;
         let mut current_start = None;
+        // Split compaction job to about 2GB each
+        const GC_COMPACT_MAX_SIZE_MB: u64 = 4 * 1024; // 4GB, TODO: should be configuration in the future
         let ranges_num = split_key_ranges.len();
         for (idx, (start, end)) in split_key_ranges.into_iter().enumerate() {
             if current_start.is_none() {
@@ -1878,7 +1833,8 @@ impl Timeline {
             }
             let res = layer_map.range_search(start..end, compact_below_lsn);
             let total_size = res.found.keys().map(|x| x.layer.file_size()).sum::<u64>();
-            if total_size > sub_compaction_max_job_size_mb * 1024 * 1024 || ranges_num == idx + 1 {
+            if total_size > GC_COMPACT_MAX_SIZE_MB * 1024 * 1024 || ranges_num == idx + 1 {
+                let mut compact_options = options.clone();
                 // Try to extend the compaction range so that we include at least one full layer file.
                 let extended_end = res
                     .found
@@ -1896,11 +1852,10 @@ impl Timeline {
                     "splitting compaction job: {}..{}, estimated_size={}",
                     start, end, total_size
                 );
-                compact_jobs.push(GcCompactJob {
-                    dry_run: job.dry_run,
-                    compact_key_range: start..end,
-                    compact_lsn_range: job.compact_lsn_range.start..compact_below_lsn,
-                });
+                compact_options.compact_range = Some(CompactRange { start, end });
+                compact_options.compact_below_lsn = Some(compact_below_lsn);
+                compact_options.sub_compaction = false;
+                compact_jobs.push(compact_options);
                 current_start = Some(end);
             }
         }
@@ -1922,7 +1877,7 @@ impl Timeline {
     /// Key::MIN..Key..MAX to the function indicates a full compaction, though technically, `Key::MAX` is not
     /// part of the range.
     ///
-    /// If `options.compact_lsn_range.end` is provided, the compaction will only compact layers below or intersect with
+    /// If `options.compact_below_lsn` is provided, the compaction will only compact layers below or intersect with
     /// the LSN. Otherwise, it will use the gc cutoff by default.
     pub(crate) async fn compact_with_gc(
         self: &Arc<Self>,
@@ -1930,13 +1885,9 @@ impl Timeline {
         options: CompactOptions,
         ctx: &RequestContext,
     ) -> anyhow::Result<()> {
-        let sub_compaction = options.sub_compaction;
-        let job = GcCompactJob::from_compact_options(options.clone());
-        if sub_compaction {
+        if options.sub_compaction {
             info!("running enhanced gc bottom-most compaction with sub-compaction, splitting compaction jobs");
-            let jobs = self
-                .gc_compaction_split_jobs(job, options.sub_compaction_max_job_size_mb)
-                .await?;
+            let jobs = self.gc_compaction_split_jobs(options).await?;
             let jobs_len = jobs.len();
             for (idx, job) in jobs.into_iter().enumerate() {
                 info!(
@@ -1951,15 +1902,19 @@ impl Timeline {
             }
             return Ok(());
         }
-        self.compact_with_gc_inner(cancel, job, ctx).await
+        self.compact_with_gc_inner(cancel, options, ctx).await
     }
 
     async fn compact_with_gc_inner(
         self: &Arc<Self>,
         cancel: &CancellationToken,
-        job: GcCompactJob,
+        options: CompactOptions,
         ctx: &RequestContext,
     ) -> anyhow::Result<()> {
+        assert!(
+            !options.sub_compaction,
+            "sub-compaction should be handled by the outer function"
+        );
         // Block other compaction/GC tasks from running for now. GC-compaction could run along
         // with legacy compaction tasks in the future. Always ensure the lock order is compaction -> gc.
         // Note that we already acquired the compaction lock when the outer `compact` function gets called.
@@ -1979,11 +1934,19 @@ impl Timeline {
         )
         .await?;
 
-        let dry_run = job.dry_run;
-        let compact_key_range = job.compact_key_range;
-        let compact_lsn_range = job.compact_lsn_range;
+        let flags = options.flags;
+        let compaction_key_range = options
+            .compact_range
+            .map(|range| range.start..range.end)
+            .unwrap_or_else(|| Key::MIN..Key::MAX);
 
-        info!("running enhanced gc bottom-most compaction, dry_run={dry_run}, compact_key_range={}..{}, compact_lsn_range={}..{}", compact_key_range.start, compact_key_range.end, compact_lsn_range.start, compact_lsn_range.end);
+        let dry_run = flags.contains(CompactFlags::DryRun);
+
+        if compaction_key_range == (Key::MIN..Key::MAX) {
+            info!("running enhanced gc bottom-most compaction, dry_run={dry_run}, compaction_key_range={}..{}", compaction_key_range.start, compaction_key_range.end);
+        } else {
+            info!("running enhanced gc bottom-most compaction, dry_run={dry_run}");
+        }
 
         scopeguard::defer! {
             info!("done enhanced gc bottom-most compaction");
@@ -2007,15 +1970,11 @@ impl Timeline {
                 // to get the truth data.
                 let real_gc_cutoff = *self.get_latest_gc_cutoff_lsn();
                 // The compaction algorithm will keep all keys above the gc_cutoff while keeping only necessary keys below the gc_cutoff for
-                // each of the retain_lsn. Therefore, if the user-provided `compact_lsn_range.end` is larger than the real gc cutoff, we will use
+                // each of the retain_lsn. Therefore, if the user-provided `compact_below_lsn` is larger than the real gc cutoff, we will use
                 // the real cutoff.
-                let mut gc_cutoff = if compact_lsn_range.end == Lsn::MAX {
-                    real_gc_cutoff
-                } else {
-                    compact_lsn_range.end
-                };
+                let mut gc_cutoff = options.compact_below_lsn.unwrap_or(real_gc_cutoff);
                 if gc_cutoff > real_gc_cutoff {
-                    warn!("provided compact_lsn_range.end={} is larger than the real_gc_cutoff={}, using the real gc cutoff", gc_cutoff, real_gc_cutoff);
+                    warn!("provided compact_below_lsn={} is larger than the real_gc_cutoff={}, using the real gc cutoff", gc_cutoff, real_gc_cutoff);
                     gc_cutoff = real_gc_cutoff;
                 }
                 gc_cutoff
@@ -2032,7 +1991,7 @@ impl Timeline {
             }
             let mut selected_layers: Vec<Layer> = Vec::new();
             drop(gc_info);
-            // Firstly, pick all the layers intersect or below the gc_cutoff, get the largest LSN in the selected layers.
+            // Pick all the layers intersect or below the gc_cutoff, get the largest LSN in the selected layers.
             let Some(max_layer_lsn) = layers
                 .iter_historic_layers()
                 .filter(|desc| desc.get_lsn_range().start <= gc_cutoff)
@@ -2042,45 +2001,27 @@ impl Timeline {
                 info!("no layers to compact with gc: no historic layers below gc_cutoff, gc_cutoff={}", gc_cutoff);
                 return Ok(());
             };
-            // Next, if the user specifies compact_lsn_range.start, we need to filter some layers out. All the layers (strictly) below
-            // the min_layer_lsn computed as below will be filtered out and the data will be accessed using the normal read path, as if
-            // it is a branch.
-            let Some(min_layer_lsn) = layers
-                .iter_historic_layers()
-                .filter(|desc| {
-                    if compact_lsn_range.start == Lsn::INVALID {
-                        true // select all layers below if start == Lsn(0)
-                    } else {
-                        desc.get_lsn_range().end > compact_lsn_range.start // strictly larger than compact_above_lsn
-                    }
-                })
-                .map(|desc| desc.get_lsn_range().start)
-                .min()
-            else {
-                info!("no layers to compact with gc: no historic layers above compact_above_lsn, compact_above_lsn={}", compact_lsn_range.end);
-                return Ok(());
-            };
             // Then, pick all the layers that are below the max_layer_lsn. This is to ensure we can pick all single-key
             // layers to compact.
             let mut rewrite_layers = Vec::new();
             for desc in layers.iter_historic_layers() {
                 if desc.get_lsn_range().end <= max_layer_lsn
-                    && desc.get_lsn_range().start >= min_layer_lsn
-                    && overlaps_with(&desc.get_key_range(), &compact_key_range)
+                    && overlaps_with(&desc.get_key_range(), &compaction_key_range)
                 {
                     // If the layer overlaps with the compaction key range, we need to read it to obtain all keys within the range,
                     // even if it might contain extra keys
                     selected_layers.push(guard.get_from_desc(&desc));
                     // If the layer is not fully contained within the key range, we need to rewrite it if it's a delta layer (it's fine
                     // to overlap image layers)
-                    if desc.is_delta() && !fully_contains(&compact_key_range, &desc.get_key_range())
+                    if desc.is_delta()
+                        && !fully_contains(&compaction_key_range, &desc.get_key_range())
                     {
                         rewrite_layers.push(desc);
                     }
                 }
             }
             if selected_layers.is_empty() {
-                info!("no layers to compact with gc: no layers within the key range, gc_cutoff={}, key_range={}..{}", gc_cutoff, compact_key_range.start, compact_key_range.end);
+                info!("no layers to compact with gc: no layers within the key range, gc_cutoff={}, key_range={}..{}", gc_cutoff, compaction_key_range.start, compaction_key_range.end);
                 return Ok(());
             }
             retain_lsns_below_horizon.sort();
@@ -2088,20 +2029,13 @@ impl Timeline {
                 selected_layers,
                 gc_cutoff,
                 retain_lsns_below_horizon,
-                min_layer_lsn,
                 max_layer_lsn,
-                compaction_key_range: compact_key_range,
+                compaction_key_range,
                 rewrite_layers,
             }
         };
-        let (has_data_below, lowest_retain_lsn) = if compact_lsn_range.start != Lsn::INVALID {
-            // If we only compact above some LSN, we should get the history from the current branch below the specified LSN.
-            // We use job_desc.min_layer_lsn as if it's the lowest branch point.
-            (true, job_desc.min_layer_lsn)
-        } else if self.ancestor_timeline.is_some() {
-            // In theory, we can also use min_layer_lsn here, but using ancestor LSN makes sure the delta layers cover the
-            // LSN ranges all the way to the ancestor timeline.
-            (true, self.ancestor_lsn)
+        let lowest_retain_lsn = if self.ancestor_timeline.is_some() {
+            Lsn(self.ancestor_lsn.0 + 1)
         } else {
             let res = job_desc
                 .retain_lsns_below_horizon
@@ -2119,19 +2053,17 @@ impl Timeline {
                         .unwrap_or(job_desc.gc_cutoff)
                 );
             }
-            (false, res)
+            res
         };
         info!(
-            "picked {} layers for compaction ({} layers need rewriting) with max_layer_lsn={} min_layer_lsn={} gc_cutoff={} lowest_retain_lsn={}, key_range={}..{}, has_data_below={}",
+            "picked {} layers for compaction ({} layers need rewriting) with max_layer_lsn={} gc_cutoff={} lowest_retain_lsn={}, key_range={}..{}",
             job_desc.selected_layers.len(),
             job_desc.rewrite_layers.len(),
             job_desc.max_layer_lsn,
-            job_desc.min_layer_lsn,
             job_desc.gc_cutoff,
             lowest_retain_lsn,
             job_desc.compaction_key_range.start,
-            job_desc.compaction_key_range.end,
-            has_data_below,
+            job_desc.compaction_key_range.end
         );
 
         for layer in &job_desc.selected_layers {
@@ -2175,22 +2107,10 @@ impl Timeline {
         let mut delta_layers = Vec::new();
         let mut image_layers = Vec::new();
         let mut downloaded_layers = Vec::new();
-        let mut total_downloaded_size = 0;
-        let mut total_layer_size = 0;
         for layer in &job_desc.selected_layers {
-            if layer.needs_download().await?.is_some() {
-                total_downloaded_size += layer.layer_desc().file_size;
-            }
-            total_layer_size += layer.layer_desc().file_size;
             let resident_layer = layer.download_and_keep_resident().await?;
             downloaded_layers.push(resident_layer);
         }
-        info!(
-            "finish downloading layers, downloaded={}, total={}, ratio={:.2}",
-            total_downloaded_size,
-            total_layer_size,
-            total_downloaded_size as f64 / total_layer_size as f64
-        );
         for resident_layer in &downloaded_layers {
             if resident_layer.layer_desc().is_delta() {
                 let layer = resident_layer.get_as_delta(ctx).await?;
@@ -2213,7 +2133,7 @@ impl Timeline {
 
         // Only create image layers when there is no ancestor branches. TODO: create covering image layer
         // when some condition meet.
-        let mut image_layer_writer = if !has_data_below {
+        let mut image_layer_writer = if self.ancestor_timeline.is_none() {
             Some(
                 SplitImageLayerWriter::new(
                     self.conf,
@@ -2246,11 +2166,7 @@ impl Timeline {
         }
         let mut delta_layer_rewriters = HashMap::<Arc<PersistentLayerKey>, RewritingLayers>::new();
 
-        /// When compacting not at a bottom range (=`[0,X)`) of the root branch, we "have data below" (`has_data_below=true`).
-        /// The two cases are compaction in ancestor branches and when `compact_lsn_range.start` is set.
-        /// In those cases, we need to pull up data from below the LSN range we're compaction.
-        ///
-        /// This function unifies the cases so that later code doesn't have to think about it.
+        /// Returns None if there is no ancestor branch. Throw an error when the key is not found.
         ///
         /// Currently, we always get the ancestor image for each key in the child branch no matter whether the image
         /// is needed for reconstruction. This should be fixed in the future.
@@ -2258,19 +2174,17 @@ impl Timeline {
         /// Furthermore, we should do vectored get instead of a single get, or better, use k-merge for ancestor
         /// images.
         async fn get_ancestor_image(
-            this_tline: &Arc<Timeline>,
+            tline: &Arc<Timeline>,
             key: Key,
             ctx: &RequestContext,
-            has_data_below: bool,
-            history_lsn_point: Lsn,
         ) -> anyhow::Result<Option<(Key, Lsn, Bytes)>> {
-            if !has_data_below {
+            if tline.ancestor_timeline.is_none() {
                 return Ok(None);
             };
             // This function is implemented as a get of the current timeline at ancestor LSN, therefore reusing
             // as much existing code as possible.
-            let img = this_tline.get(key, history_lsn_point, ctx).await?;
-            Ok(Some((key, history_lsn_point, img)))
+            let img = tline.get(key, tline.ancestor_lsn, ctx).await?;
+            Ok(Some((key, tline.ancestor_lsn, img)))
         }
 
         // Actually, we can decide not to write to the image layer at all at this point because
@@ -2354,8 +2268,7 @@ impl Timeline {
                         job_desc.gc_cutoff,
                         &job_desc.retain_lsns_below_horizon,
                         COMPACTION_DELTA_THRESHOLD,
-                        get_ancestor_image(self, *last_key, ctx, has_data_below, lowest_retain_lsn)
-                            .await?,
+                        get_ancestor_image(self, *last_key, ctx).await?,
                     )
                     .await?;
                 retention
@@ -2384,7 +2297,7 @@ impl Timeline {
                 job_desc.gc_cutoff,
                 &job_desc.retain_lsns_below_horizon,
                 COMPACTION_DELTA_THRESHOLD,
-                get_ancestor_image(self, last_key, ctx, has_data_below, lowest_retain_lsn).await?,
+                get_ancestor_image(self, last_key, ctx).await?,
             )
             .await?;
         retention
@@ -2904,7 +2817,7 @@ impl CompactionLayer<Key> for ResidentDeltaLayer {
 impl CompactionDeltaLayer<TimelineAdaptor> for ResidentDeltaLayer {
     type DeltaEntry<'a> = DeltaEntry<'a>;
 
-    async fn load_keys(&self, ctx: &RequestContext) -> anyhow::Result<Vec<DeltaEntry<'_>>> {
+    async fn load_keys<'a>(&self, ctx: &RequestContext) -> anyhow::Result<Vec<DeltaEntry<'_>>> {
         self.0.get_as_delta(ctx).await?.index_entries(ctx).await
     }
 }

pageserver/src/walingest.rs

@@ -877,24 +877,22 @@ impl WalIngest {
         // will block waiting for the last valid LSN to advance up to
         // it. So we use the previous record's LSN in the get calls
         // instead.
-        if modification.tline.get_shard_identity().is_shard_zero() {
-            for segno in modification
-                .tline
-                .list_slru_segments(SlruKind::Clog, Version::Modified(modification), ctx)
-                .await?
-            {
-                let segpage = segno * pg_constants::SLRU_PAGES_PER_SEGMENT;
+        for segno in modification
+            .tline
+            .list_slru_segments(SlruKind::Clog, Version::Modified(modification), ctx)
+            .await?
+        {
+            let segpage = segno * pg_constants::SLRU_PAGES_PER_SEGMENT;
 
-                let may_delete = dispatch_pgversion!(modification.tline.pg_version, {
-                    pgv::nonrelfile_utils::slru_may_delete_clogsegment(segpage, pageno)
-                });
+            let may_delete = dispatch_pgversion!(modification.tline.pg_version, {
+                pgv::nonrelfile_utils::slru_may_delete_clogsegment(segpage, pageno)
+            });
 
-                if may_delete {
-                    modification
-                        .drop_slru_segment(SlruKind::Clog, segno, ctx)
-                        .await?;
-                    trace!("Drop CLOG segment {:>04X}", segno);
-                }
+            if may_delete {
+                modification
+                    .drop_slru_segment(SlruKind::Clog, segno, ctx)
+                    .await?;
+                trace!("Drop CLOG segment {:>04X}", segno);
             }
         }
 
@@ -1049,18 +1047,16 @@ impl WalIngest {
 
         // Delete all the segments except the last one. The last segment can still
         // contain, possibly partially, valid data.
-        if modification.tline.get_shard_identity().is_shard_zero() {
-            while segment != endsegment {
-                modification
-                    .drop_slru_segment(SlruKind::MultiXactMembers, segment as u32, ctx)
-                    .await?;
+        while segment != endsegment {
+            modification
+                .drop_slru_segment(SlruKind::MultiXactMembers, segment as u32, ctx)
+                .await?;
 
-                /* move to next segment, handling wraparound correctly */
-                if segment == maxsegment {
-                    segment = 0;
-                } else {
-                    segment += 1;
-                }
+            /* move to next segment, handling wraparound correctly */
+            if segment == maxsegment {
+                segment = 0;
+            } else {
+                segment += 1;
             }
         }
 

pgxn/neon/control_plane_connector.c

@@ -428,8 +428,6 @@ MergeTable()
 		hash_seq_init(&status, old_table->role_table);
 		while ((entry = hash_seq_search(&status)) != NULL)
 		{
-			RoleEntry * old;
-			bool found_old = false;
 			RoleEntry  *to_write = hash_search(
 											   CurrentDdlTable->role_table,
 											   entry->name,
@@ -437,23 +435,30 @@ MergeTable()
 											   NULL);
 
 			to_write->type = entry->type;
-			to_write->password = entry->password;
+			if (entry->password)
+				to_write->password = entry->password;
 			strlcpy(to_write->old_name, entry->old_name, NAMEDATALEN);
-			if (entry->old_name[0] == '\0')
-				continue;
+			if (entry->old_name[0] != '\0')
+			{
+				bool		found_old = false;
+				RoleEntry  *old = hash_search(
+											  CurrentDdlTable->role_table,
+											  entry->old_name,
+											  HASH_FIND,
+											  &found_old);
 
-			old = hash_search(
-							  CurrentDdlTable->role_table,
-							  entry->old_name,
-							  HASH_FIND,
-							  &found_old);
-			if (!found_old)
-				continue;
-			strlcpy(to_write->old_name, old->old_name, NAMEDATALEN);
-			hash_search(CurrentDdlTable->role_table,
-						entry->old_name,
-						HASH_REMOVE,
-						NULL);
+				if (found_old)
+				{
+					if (old->old_name[0] != '\0')
+						strlcpy(to_write->old_name, old->old_name, NAMEDATALEN);
+					else
+						strlcpy(to_write->old_name, entry->old_name, NAMEDATALEN);
+					hash_search(CurrentDdlTable->role_table,
+								entry->old_name,
+								HASH_REMOVE,
+								NULL);
+				}
+			}
 		}
 		hash_destroy(old_table->role_table);
 	}