fix accidental recursion

.github/actionlint.yml

@@ -21,7 +21,3 @@ config-variables:
   - SLACK_UPCOMING_RELEASE_CHANNEL_ID
   - DEV_AWS_OIDC_ROLE_ARN
   - 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:
+  aws_oicd_role_arn:
-    description: 'OIDC role arn to interract with S3'
+    description: 'the OIDC role arn to (re-)acquire for allure report upload - if not set call must acquire OIDC role'
-    required: true
+    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
+    - name: (Re-)configure AWS credentials # necessary to upload reports to S3 after a long-running test
-      if: ${{ !cancelled() }}
+      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:
+  aws_oicd_role_arn:
-    description: 'OIDC role arn to interract with S3'
+    description: 'the OIDC role arn to (re-)acquire for allure report upload - if not set call must acquire OIDC role'
-    required: true
+    required: false
+    default: ''
 
 runs:
   using: "composite"
@@ -35,11 +36,12 @@ runs:
       env:
         REPORT_DIR: ${{ inputs.report-dir }}
 
-    - uses: aws-actions/configure-aws-credentials@v4
+    - name: (Re-)configure AWS credentials # necessary to upload reports to S3 after a long-running test
-      if: ${{ !cancelled() }}
+      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:
+  aws_oicd_role_arn:
-    description: 'OIDC role arn to interract with S3'
+    description: 'the OIDC role arn to (re-)acquire for allure report upload - if not set call must acquire OIDC role'
-    required: true
+    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'
@@ -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
+    - name: (Re-)configure AWS credentials # necessary to upload reports to S3 after a long-running test
-      if: ${{ !cancelled() }}
+      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:
@@ -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 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 without LFC on v17 release only
       test-cfg: |
-        ${{ matrix.build-type == 'release' && '[{"pg_version":"v14", "lfc_state": "with-lfc"},
+        ${{ matrix.build-type == 'release' && '[{"pg_version":"v14", "lfc_state": "without-lfc"},
-                                                {"pg_version":"v15", "lfc_state": "with-lfc"},
+                                                {"pg_version":"v15", "lfc_state": "without-lfc"},
-                                                {"pg_version":"v16", "lfc_state": "with-lfc"},
+                                                {"pg_version":"v16", "lfc_state": "without-lfc"},
-                                                {"pg_version":"v17", "lfc_state": "with-lfc"},
+                                                {"pg_version":"v17", "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" }]' }}
+                                           || '[{"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: docker/login-action@v3
-        uses: aws-actions/configure-aws-credentials@v4
         with:
-          aws-region: eu-central-1
+          registry: 369495373322.dkr.ecr.eu-central-1.amazonaws.com
-          role-to-assume: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
+          username: ${{ secrets.AWS_ACCESS_KEY_DEV }}
-          role-duration-seconds: 3600
+          password: ${{ secrets.AWS_SECRET_KEY_DEV }}
-
-      - name: Login to Amazon Dev ECR
-        uses: aws-actions/amazon-ecr-login@v2
 
       - 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: docker/login-action@v3
-        uses: aws-actions/configure-aws-credentials@v4
         with:
-          aws-region: eu-central-1
+          registry: 369495373322.dkr.ecr.eu-central-1.amazonaws.com
-          role-to-assume: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
+          username: ${{ secrets.AWS_ACCESS_KEY_DEV }}
-          role-duration-seconds: 3600
+          password: ${{ secrets.AWS_SECRET_KEY_DEV }}
-
-      - name: Login to Amazon Dev ECR
-        uses: aws-actions/amazon-ecr-login@v2
 
       - 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: docker/login-action@v3
-        uses: aws-actions/configure-aws-credentials@v4
         with:
-          aws-region: eu-central-1
+          registry: 369495373322.dkr.ecr.eu-central-1.amazonaws.com
-          role-to-assume: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
+          username: ${{ secrets.AWS_ACCESS_KEY_DEV }}
-          role-duration-seconds: 3600
+          password: ${{ secrets.AWS_SECRET_KEY_DEV }}
-
-      - name: Login to Amazon Dev ECR
-        uses: aws-actions/amazon-ecr-login@v2
 
       - 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
+      id-token: write # for `aws-actions/configure-aws-credentials`
-      statuses: write
-      contents: read
 
     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
+      - name: Login to dev ECR
-        uses: aws-actions/configure-aws-credentials@v4
+        uses: docker/login-action@v3
         with:
-          aws-region: eu-central-1
+          registry: 369495373322.dkr.ecr.eu-central-1.amazonaws.com
-          role-to-assume: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
+          username: ${{ secrets.AWS_ACCESS_KEY_DEV }}
-          role-duration-seconds: 3600
+          password: ${{ secrets.AWS_SECRET_KEY_DEV }}
-
-      - name: Login to Amazon Dev ECR
-        uses: aws-actions/amazon-ecr-login@v2
 
       - 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

@@ -19,21 +19,15 @@ concurrency:
   group: ${{ github.workflow }}
   cancel-in-progress: true
 
-permissions:
-  id-token: write # aws-actions/configure-aws-credentials
-  statuses: write
-  contents: write
-
 jobs:
   regress:
     env:
       POSTGRES_DISTRIB_DIR: /tmp/neon/pg_install
+      DEFAULT_PG_VERSION: 16
       TEST_OUTPUT: /tmp/test_output
       BUILD_TYPE: remote
-    strategy:
+      AWS_ACCESS_KEY_ID: ${{ secrets.AWS_ACCESS_KEY_DEV }}
-      fail-fast: false
+      AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_KEY_DEV }}
-      matrix:
-        pg-version: [16, 17]
 
     runs-on: us-east-2
     container:
@@ -46,11 +40,9 @@ jobs:
           submodules: true
 
       - name: Patch the test
-        env:
-          PG_VERSION: ${{matrix.pg-version}}
         run: |
-          cd "vendor/postgres-v${PG_VERSION}"
+          cd "vendor/postgres-v${DEFAULT_PG_VERSION}"
-          patch -p1 < "../../compute/patches/cloud_regress_pg${PG_VERSION}.patch"
+          patch -p1 < "../../compute/patches/cloud_regress_pg${DEFAULT_PG_VERSION}.patch"
 
       - name: Generate a random password
         id: pwgen
@@ -63,9 +55,8 @@ jobs:
       - name: Change tests according to the generated password
         env:
           DBPASS: ${{ steps.pwgen.outputs.DBPASS }}
-          PG_VERSION: ${{matrix.pg-version}}
         run: |
-          cd vendor/postgres-v"${PG_VERSION}"/src/test/regress
+          cd vendor/postgres-v"${DEFAULT_PG_VERSION}"/src/test/regress
           for fname in sql/*.sql expected/*.out; do
             sed -i.bak s/NEON_PASSWORD_PLACEHOLDER/"'${DBPASS}'"/ "${fname}"
           done
@@ -81,46 +72,27 @@ jobs:
           name: neon-${{ 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
-        uses: ./.github/actions/neon-branch-create
-        with:
-          api_key: ${{ secrets.NEON_STAGING_API_KEY }}
-          project_id: ${{ vars[format('PGREGRESS_PG{0}_PROJECT_ID', matrix.pg-version)] }}
 
       - name: Run the regression tests
         uses: ./.github/actions/run-python-test-set
         with:
           build_type: ${{ env.BUILD_TYPE }}
           test_selection: cloud_regress
-          pg_version: ${{matrix.pg-version}}
+          pg_version: ${{ env.DEFAULT_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}}
+          BENCHMARK_CONNSTR: ${{ secrets.PG_REGRESS_CONNSTR }}
-
-      - name: Delete branch
-        if: always()
-        uses: ./.github/actions/neon-branch-delete
-        with:
-          api_key: ${{ secrets.NEON_STAGING_API_KEY }}
-          project_id: ${{ vars[format('PGREGRESS_PG{0}_PROJECT_ID', matrix.pg-version)] }}
-          branch_id: ${{steps.create-branch.outputs.branch_id}}
 
       - name: Create Allure report
         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: docker/login-action@v3
-        uses: aws-actions/configure-aws-credentials@v4
         with:
-          aws-region: eu-central-1
+          registry: 369495373322.dkr.ecr.eu-central-1.amazonaws.com
-          role-to-assume: ${{ vars.DEV_AWS_OIDC_ROLE_ARN }}
+          username: ${{ secrets.AWS_ACCESS_KEY_DEV }}
-          role-duration-seconds: 3600
+          password: ${{ secrets.AWS_SECRET_KEY_DEV }}
-
-      - name: Login to Amazon Dev ECR
-        uses: aws-actions/amazon-ecr-login@v2
 
       - 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

CODEOWNERS

@@ -1,29 +1,16 @@
-# Autoscaling
-/libs/vm_monitor/ @neondatabase/autoscaling
-
-# DevProd
 /.github/ @neondatabase/developer-productivity
-
+/compute_tools/ @neondatabase/control-plane @neondatabase/compute
-# Compute
+/libs/pageserver_api/ @neondatabase/storage
-/pgxn/ @neondatabase/compute
+/libs/postgres_ffi/ @neondatabase/compute @neondatabase/storage
-/vendor/ @neondatabase/compute
-/compute/ @neondatabase/compute
-/compute_tools/ @neondatabase/compute
-
-# Proxy
 /libs/proxy/ @neondatabase/proxy
-/proxy/ @neondatabase/proxy
+/libs/remote_storage/ @neondatabase/storage
-
+/libs/safekeeper_api/ @neondatabase/storage
-# Storage
+/libs/vm_monitor/ @neondatabase/autoscaling
 /pageserver/ @neondatabase/storage
+/pgxn/ @neondatabase/compute
+/pgxn/neon/ @neondatabase/compute @neondatabase/storage
+/proxy/ @neondatabase/proxy
 /safekeeper/ @neondatabase/storage
 /storage_controller @neondatabase/storage
 /storage_scrubber @neondatabase/storage
-/libs/pageserver_api/ @neondatabase/storage
+/vendor/ @neondatabase/compute
-/libs/remote_storage/ @neondatabase/storage
-/libs/safekeeper_api/ @neondatabase/storage
-
-# Shared
-/pgxn/neon/ @neondatabase/compute @neondatabase/storage
-/libs/compute_api/ @neondatabase/compute @neondatabase/control-plane
-/libs/postgres_ffi/ @neondatabase/compute @neondatabase/storage

Cargo.toml

@@ -11,7 +11,6 @@ members = [
     "pageserver/pagebench",
     "proxy",
     "safekeeper",
-    "safekeeper/client",
     "storage_broker",
     "storage_controller",
     "storage_controller/client",
@@ -52,7 +51,10 @@ 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"
+azure_core = { version = "0.19", default-features = false, features = ["enable_reqwest_rustls", "hmac_rust"] }
+azure_identity = { version = "0.19", default-features = false, features = ["enable_reqwest_rustls"] }
+azure_storage = { version = "0.19", default-features = false, features = ["enable_reqwest_rustls"] }
+azure_storage_blobs = { version = "0.19", default-features = false, features = ["enable_reqwest_rustls"] }
 flate2 = "1.0.26"
 async-stream = "0.3"
 async-trait = "0.1"
@@ -214,12 +216,6 @@ postgres-protocol = { git = "https://github.com/neondatabase/rust-postgres.git",
 postgres-types = { git = "https://github.com/neondatabase/rust-postgres.git", branch = "neon" }
 tokio-postgres = { git = "https://github.com/neondatabase/rust-postgres.git", branch = "neon" }
 
-## 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"] }
-
 ## Local libraries
 compute_api = { version = "0.1", path = "./libs/compute_api/" }
 consumption_metrics = { version = "0.1", path = "./libs/consumption_metrics/" }
@@ -235,7 +231,6 @@ 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" }

build-tools.Dockerfile

@@ -115,7 +115,7 @@ RUN set -e \
 
 # Keep the version the same as in compute/compute-node.Dockerfile and
 # test_runner/regress/test_compute_metrics.py.
-ENV SQL_EXPORTER_VERSION=0.16.0
+ENV SQL_EXPORTER_VERSION=0.13.1
 RUN curl -fsSL \
     "https://github.com/burningalchemist/sql_exporter/releases/download/${SQL_EXPORTER_VERSION}/sql_exporter-${SQL_EXPORTER_VERSION}.linux-$(case "$(uname -m)" in x86_64) echo amd64;; aarch64) echo arm64;; esac).tar.gz" \
     --output sql_exporter.tar.gz \

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 \
+    apt install --no-install-recommends -y git autoconf automake libtool build-essential bison flex libreadline-dev \
-    ninja-build 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 \
+    $VERSION_INSTALLS
-    && apt clean && rm -rf /var/lib/apt/lists/*
 
 #########################################################################################
 #
@@ -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 \
+    apt install --no-install-recommends -y gdal-bin libboost-dev libboost-thread-dev libboost-filesystem-dev \
-    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 \
+    protobuf-c-compiler xsltproc
-    && apt clean && rm -rf /var/lib/apt/lists/*
 
 
 # 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) && \
+    cmake -DCMAKE_BUILD_TYPE=Release . && make -j $(getconf _NPROCESSORS_ONLN) && \
-    DESTDIR=/sfcgal ninja install -j $(getconf _NPROCESSORS_ONLN) && \
+    DESTDIR=/sfcgal make install -j $(getconf _NPROCESSORS_ONLN) && \
-    ninja clean && cp -R /sfcgal/* /
+    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 .. && \
+    cmake -DCMAKE_BUILD_TYPE=Release .. && \
-    ninja -j $(getconf _NPROCESSORS_ONLN) && \
+    make -j $(getconf _NPROCESSORS_ONLN) && \
-    ninja -j $(getconf _NPROCESSORS_ONLN) install && \
+    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 \
+    apt install --no-install-recommends -y ninja-build python3-dev libncurses5 binutils clang
-    ninja-build python3-dev libncurses5 binutils clang \
-    && apt clean && rm -rf /var/lib/apt/lists/*
 
 # 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 \
+    cmake .. -DCMAKE_BUILD_TYPE=Release && \
-        -DBUILD_FUZZERS=0 -DBUILD_FILTERS=0 -DBUILD_GENERATORS=0 -DBUILD_TESTING=0 \
+    make -j $(getconf _NPROCESSORS_ONLN) && \
-    && ninja -j $(getconf _NPROCESSORS_ONLN) && \
+    DESTDIR=/h3 make install && \
-    DESTDIR=/h3 ninja 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 && \
+RUN apt-get update && \
-    apt install --no-install-recommends --no-install-suggests -y \
+    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 \
+        libboost-all-dev
-    && apt clean && rm -rf /var/lib/apt/lists/*
 
 # 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) && \
+    make -j $(getconf _NPROCESSORS_ONLN) && \
-    ninja -j $(getconf _NPROCESSORS_ONLN) install && \
+    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 && \
+RUN apt-get update && \
-    apt install --no-install-recommends --no-install-suggests -y curl libclang-dev && \
+    apt-get install --no-install-recommends -y curl libclang-dev && \
-    apt clean && rm -rf /var/lib/apt/lists/* && \
     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 && \
+RUN apt-get update && \
-    apt install --no-install-recommends --no-install-suggests -y curl libclang-dev && \
+    apt-get install --no-install-recommends -y curl libclang-dev && \
-    apt clean && rm -rf /var/lib/apt/lists/* && \
     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 \
+RUN apt-get update && apt-get install -y python3 python3-pip python3-venv && \
-    python3 python3-pip python3-venv && \
-    apt clean && rm -rf /var/lib/apt/lists/* && \
     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 \
+    ./build.sh --config Release --parallel --skip_submodule_sync --skip_tests --allow_running_as_root
-    --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 \
+RUN apt-get install -y protobuf-compiler && \
-    && apt clean && rm -rf /var/lib/apt/lists/* && \
     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-get update \
-    && apt install --no-install-suggests --no-install-recommends -y \
+    && 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 \
+        pkg-config
-    && apt clean && rm -rf /var/lib/apt/lists/*
 
 # Use `dist_man_MANS=` to skip manpage generation (which requires python3/pandoc)
 ENV PGBOUNCER_TAG=pgbouncer_1_22_1
@@ -1362,7 +1324,7 @@ FROM quay.io/prometheuscommunity/postgres-exporter:v0.12.1 AS postgres-exporter
 
 # Keep the version the same as in build-tools.Dockerfile and
 # test_runner/regress/test_compute_metrics.py.
-FROM burningalchemist/sql_exporter:0.16.0 AS sql-exporter
+FROM burningalchemist/sql_exporter:0.13.1 AS sql-exporter
 
 #########################################################################################
 #
@@ -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',
+  metric_name: 'compute_backpressure_throttling_seconds',
-  type: 'counter',
+  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_passwd1 | NEON_MD5_PLACEHOLDER_regress_passwd1
-+ regress_passwd2 | NEON_MD5_PLACEHOLDER:regress_passwd2
++ 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
+ -- already encrypted with MD5, use as it is
-+CREATE ROLE regress_passwd5 PASSWORD NEON_PASSWORD_PLACEHOLDER;
+ CREATE ROLE regress_passwd5 PASSWORD 'md5e73a4b11df52a6068f8b39f90be36023';
-+-- Neon does not support encrypted passwords, use unencrypted instead
++ERROR:  Received HTTP code 400 from control plane: {"error":"Neon only supports being given plaintext passwords"}
-+CREATE ROLE regress_passwd6 PASSWORD NEON_PASSWORD_PLACEHOLDER;
+ -- This looks like a valid SCRAM-SHA-256 secret, but it is not
-+CREATE ROLE regress_passwd7 PASSWORD NEON_PASSWORD_PLACEHOLDER;
+ -- so it should be hashed with SCRAM-SHA-256.
-+CREATE ROLE regress_passwd8 PASSWORD NEON_PASSWORD_PLACEHOLDER;
+ 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_passwd1 | NEON_MD5_PLACEHOLDER_regress_passwd1
-+ regress_passwd2 | NEON_MD5_PLACEHOLDER:regress_passwd2
++ 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_passwd6 | SCRAM-SHA-256$4096:<salt>$<storedkey>:<serverkey>
+- regress_passwd7 | 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_passwd8 | SCRAM-SHA-256$4096:<salt>$<storedkey>:<serverkey>
+  regress_passwd9 | SCRAM-SHA-256$1024:<salt>$<storedkey>:<serverkey>
-@@ -95,23 +87,20 @@ SELECT rolname, regexp_replace(rolpassword, '(SCRAM-SHA-256)\$(\d+):([a-zA-Z0-9+
+-(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.
+ -- Test with invalid stored and server keys.
----
+ --
---- The first is valid, to act as a control. The others have too long
+ -- The first is valid, to act as a control. The others have too long
---- stored/server keys. They will be re-hashed.
+ -- 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_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=';
++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=';
+ CREATE ROLE regress_passwd_sha_len1 PASSWORD 'SCRAM-SHA-256$4096:A6xHKoH/494E941doaPOYg==$Ky+A30sewHIH3VHQLRN9vYsuzlgNyGNKCh37dy96RqwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=:COPdlNiIkrsacU5QoxydEuOH6e/KfiipeETb/bPw8ZI=';
-+-- Neon does not support encrypted passwords, use unencrypted instead
++ERROR:  Received HTTP code 400 from control plane: {"error":"Neon only supports being given plaintext passwords"}
-+CREATE ROLE regress_passwd_sha_len0 PASSWORD NEON_PASSWORD_PLACEHOLDER;
+ CREATE ROLE regress_passwd_sha_len2 PASSWORD 'SCRAM-SHA-256$4096:A6xHKoH/494E941doaPOYg==$Ky+A30sewHIH3VHQLRN9vYsuzlgNyGNKCh37dy96Rqw=:COPdlNiIkrsacU5QoxydEuOH6e/KfiipeETb/bPw8ZIAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA=';
-+CREATE ROLE regress_passwd_sha_len1 PASSWORD NEON_PASSWORD_PLACEHOLDER;
++ERROR:  Received HTTP code 400 from control plane: {"error":"Neon only supports being given plaintext passwords"}
-+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.
  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_len1 | t
+- regress_passwd_sha_len2 | t
-  regress_passwd_sha_len2 | t
+-(3 rows)
- (3 rows)
++ rolname | is_rolpassword_rehashed 
-@@ -135,6 +124,7 @@ DROP ROLE regress_passwd7;
++---------+-------------------------
++(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
+ -- already encrypted with MD5, use as it is
-+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;
-@@ -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_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
+    let spec;
-    if let Some(spec_json) = spec_json {
+    let mut live_config_allowed = false;
-        info!("got spec from cli argument {}", spec_json);
+    match spec_json {
-        return Ok(CliSpecParams {
+        // First, try to get cluster spec from the cli argument
-            spec: Some(serde_json::from_str(spec_json)?),
+        Some(json) => {
-            live_config_allowed: false,
+            info!("got spec from cli argument {}", json);
-        });
+            spec = Some(serde_json::from_str(json)?);
-    }
-
-    // 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)
         }
-    }
+        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:
+              versions:
-                type: string
+                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)
+/// we only keep the highest and lowest version across all databases.
-/// for each extension version.
 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| {
+            .map(|row| (row.get("extname"), row.get("extversion")))
-                (
-                    row.get("extname"),
-                    row.get("extversion"),
-                    row.get("extowner"),
-                )
-            })
             .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
+            // increment the number of databases where the version of extension is installed
-            // 10 is the oid of superuser
+            INSTALLED_EXTENSIONS
-            let owned_by_superuser = if *extowner == 10 { "1" } else { "0" };
+                .with_label_values(&[extname, &version])
+                .inc();
 
             extensions_map
-                .entry((
+                .entry(extname.to_string())
-                    extname.to_string(),
-                    version.clone(),
-                    owned_by_superuser.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 res = InstalledExtensions {
-        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 {
         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,
+        ShardParameters, TenantConfig, TenantConfigRequest, TenantShardSplitRequest,
-        TenantShardSplitRequest, TenantShardSplitResponse,
+        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.
+    TenantConfig {
-    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 {
         #[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_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_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(

deny.toml

@@ -42,7 +42,6 @@ allow = [
     "MPL-2.0",
     "OpenSSL",
     "Unicode-DFS-2016",
-    "Unicode-3.0",
 ]
 confidence-threshold = 0.8
 exceptions = [

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
+    # The support of pg_anon not yet added to PG17, so we have to remove the corresponding option
-    if [ "${pg_version}" -ne 17 ]; then
+    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 ${TEST_VERSION_ONLY-14 15 16 17}; do
         fi
     fi
     cleanup
-    # Restore the original spec.json
+    # The support of pg_anon not yet added to PG17, so we have to remove the corresponding option
-    if [ "$pg_version" -ne 17 ]; then
+    if [ $pg_version -eq 17 ]; then
-      mv "$SPEC_PATH/spec.bak" "$SPEC_PATH/spec.json"
+      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 {
@@ -245,17 +245,6 @@ impl From<NodeAvailability> for NodeAvailabilityWrapper {
     }
 }
 
-/// Scheduling policy enables us to selectively disable some automatic actions that the
-/// controller performs on a tenant shard. This is only set to a non-default value by
-/// human intervention, and it is reset to the default value (Active) when the tenant's
-/// placement policy is modified away from Attached.
-///
-/// The typical use of a non-Active scheduling policy is one of:
-/// - Pinnning a shard to a node (i.e. migrating it there & setting a non-Active scheduling policy)
-/// - Working around a bug (e.g. if something is flapping and we need to stop it until the bug is fixed)
-///
-/// If you're not sure which policy to use to pin a shard to its current location, you probably
-/// want Pause.
 #[derive(Serialize, Deserialize, Clone, Copy, Eq, PartialEq, Debug)]
 pub enum ShardSchedulingPolicy {
     // Normal mode: the tenant's scheduled locations may be updated at will, including

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()
+        if key.is_slru_block_key() || key.is_slru_segment_size_key() || key.is_aux_file_key() {
-            || key.is_slru_segment_size_key()
-            || key.is_aux_file_key()
-            || key.is_slru_dir_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/postgres-protocol2/Cargo.toml

@@ -1,7 +1,7 @@
 [package]
 name = "postgres-protocol2"
 version = "0.1.0"
-edition = "2021"
+edition = "2018"
 license = "MIT/Apache-2.0"
 
 [dependencies]

libs/proxy/postgres-protocol2/src/lib.rs

@@ -9,7 +9,8 @@
 //!
 //! This library assumes that the `client_encoding` backend parameter has been
 //! set to `UTF8`. It will most likely not behave properly if that is not the case.
-#![warn(missing_docs, clippy::all)]
+#![doc(html_root_url = "https://docs.rs/postgres-protocol/0.6")]
+#![warn(missing_docs, rust_2018_idioms, clippy::all)]
 
 use byteorder::{BigEndian, ByteOrder};
 use bytes::{BufMut, BytesMut};

libs/proxy/postgres-protocol2/src/message/frontend.rs

@@ -3,6 +3,7 @@
 
 use byteorder::{BigEndian, ByteOrder};
 use bytes::{Buf, BufMut, BytesMut};
+use std::convert::TryFrom;
 use std::error::Error;
 use std::io;
 use std::marker;

libs/proxy/postgres-types2/Cargo.toml

@@ -1,7 +1,7 @@
 [package]
 name = "postgres-types2"
 version = "0.1.0"
-edition = "2021"
+edition = "2018"
 license = "MIT/Apache-2.0"
 
 [dependencies]

libs/proxy/postgres-types2/src/lib.rs

@@ -2,7 +2,8 @@
 //!
 //! This crate is used by the `tokio-postgres` and `postgres` crates. You normally don't need to depend directly on it
 //! unless you want to define your own `ToSql` or `FromSql` definitions.
-#![warn(clippy::all, missing_docs)]
+#![doc(html_root_url = "https://docs.rs/postgres-types/0.2")]
+#![warn(clippy::all, rust_2018_idioms, missing_docs)]
 
 use fallible_iterator::FallibleIterator;
 use postgres_protocol2::types;

libs/proxy/tokio-postgres2/Cargo.toml

@@ -1,7 +1,7 @@
 [package]
 name = "tokio-postgres2"
 version = "0.1.0"
-edition = "2021"
+edition = "2018"
 license = "MIT/Apache-2.0"
 
 [dependencies]

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

@@ -4,23 +4,18 @@ use crate::config::Host;
 use crate::config::SslMode;
 use crate::connection::{Request, RequestMessages};
 
-use crate::query::RowStream;
+use crate::types::{Oid, Type};
-use crate::simple_query::SimpleQueryStream;
-
-use crate::types::{Oid, ToSql, Type};
 
 use crate::{
-    prepare, query, simple_query, slice_iter, CancelToken, Error, ReadyForQueryStatus, Row,
+    simple_query, CancelToken, Error, ReadyForQueryStatus, Statement, Transaction,
-    SimpleQueryMessage, Statement, ToStatement, Transaction, TransactionBuilder,
+    TransactionBuilder,
 };
 use bytes::BytesMut;
 use fallible_iterator::FallibleIterator;
-use futures_util::{future, ready, TryStreamExt};
+use futures_util::{future, ready};
-use parking_lot::Mutex;
 use postgres_protocol2::message::{backend::Message, frontend};
 use std::collections::HashMap;
 use std::fmt;
-use std::sync::Arc;
 use std::task::{Context, Poll};
 use tokio::sync::mpsc;
 
@@ -55,7 +50,7 @@ impl Responses {
 /// A cache of type info and prepared statements for fetching type info
 /// (corresponding to the queries in the [prepare] module).
 #[derive(Default)]
-struct CachedTypeInfo {
+pub(crate) struct CachedTypeInfo {
     /// A statement for basic information for a type from its
     /// OID. Corresponds to [TYPEINFO_QUERY](prepare::TYPEINFO_QUERY) (or its
     /// fallback).
@@ -71,13 +66,45 @@ struct CachedTypeInfo {
     /// Cache of types already looked up.
     types: HashMap<Oid, Type>,
 }
+impl CachedTypeInfo {
+    pub(crate) fn typeinfo(&mut self) -> Option<&Statement> {
+        self.typeinfo.as_ref()
+    }
+
+    pub(crate) fn set_typeinfo(&mut self, statement: Statement) -> &Statement {
+        self.typeinfo.insert(statement)
+    }
+
+    pub(crate) fn typeinfo_composite(&mut self) -> Option<&Statement> {
+        self.typeinfo_composite.as_ref()
+    }
+
+    pub(crate) fn set_typeinfo_composite(&mut self, statement: Statement) -> &Statement {
+        self.typeinfo_composite.insert(statement)
+    }
+
+    pub(crate) fn typeinfo_enum(&mut self) -> Option<&Statement> {
+        self.typeinfo_enum.as_ref()
+    }
+
+    pub(crate) fn set_typeinfo_enum(&mut self, statement: Statement) -> &Statement {
+        self.typeinfo_enum.insert(statement)
+    }
+
+    pub(crate) fn type_(&mut self, oid: Oid) -> Option<Type> {
+        self.types.get(&oid).cloned()
+    }
+
+    pub(crate) fn set_type(&mut self, oid: Oid, type_: &Type) {
+        self.types.insert(oid, type_.clone());
+    }
+}
 
 pub struct InnerClient {
     sender: mpsc::UnboundedSender<Request>,
-    cached_typeinfo: Mutex<CachedTypeInfo>,
 
     /// A buffer to use when writing out postgres commands.
-    buffer: Mutex<BytesMut>,
+    buffer: BytesMut,
 }
 
 impl InnerClient {
@@ -92,47 +119,14 @@ impl InnerClient {
         })
     }
 
-    pub fn typeinfo(&self) -> Option<Statement> {
-        self.cached_typeinfo.lock().typeinfo.clone()
-    }
-
-    pub fn set_typeinfo(&self, statement: &Statement) {
-        self.cached_typeinfo.lock().typeinfo = Some(statement.clone());
-    }
-
-    pub fn typeinfo_composite(&self) -> Option<Statement> {
-        self.cached_typeinfo.lock().typeinfo_composite.clone()
-    }
-
-    pub fn set_typeinfo_composite(&self, statement: &Statement) {
-        self.cached_typeinfo.lock().typeinfo_composite = Some(statement.clone());
-    }
-
-    pub fn typeinfo_enum(&self) -> Option<Statement> {
-        self.cached_typeinfo.lock().typeinfo_enum.clone()
-    }
-
-    pub fn set_typeinfo_enum(&self, statement: &Statement) {
-        self.cached_typeinfo.lock().typeinfo_enum = Some(statement.clone());
-    }
-
-    pub fn type_(&self, oid: Oid) -> Option<Type> {
-        self.cached_typeinfo.lock().types.get(&oid).cloned()
-    }
-
-    pub fn set_type(&self, oid: Oid, type_: &Type) {
-        self.cached_typeinfo.lock().types.insert(oid, type_.clone());
-    }
-
     /// Call the given function with a buffer to be used when writing out
     /// postgres commands.
-    pub fn with_buf<F, R>(&self, f: F) -> R
+    pub fn with_buf<F, R>(&mut self, f: F) -> R
     where
         F: FnOnce(&mut BytesMut) -> R,
     {
-        let mut buffer = self.buffer.lock();
+        let r = f(&mut self.buffer);
-        let r = f(&mut buffer);
+        self.buffer.clear();
-        buffer.clear();
         r
     }
 }
@@ -150,7 +144,8 @@ pub struct SocketConfig {
 /// The client is one half of what is returned when a connection is established. Users interact with the database
 /// through this client object.
 pub struct Client {
-    inner: Arc<InnerClient>,
+    pub(crate) inner: InnerClient,
+    pub(crate) cached_typeinfo: CachedTypeInfo,
 
     socket_config: SocketConfig,
     ssl_mode: SslMode,
@@ -167,11 +162,11 @@ impl Client {
         secret_key: i32,
     ) -> Client {
         Client {
-            inner: Arc::new(InnerClient {
+            inner: InnerClient {
                 sender,
-                cached_typeinfo: Default::default(),
                 buffer: Default::default(),
-            }),
+            },
+            cached_typeinfo: Default::default(),
 
             socket_config,
             ssl_mode,
@@ -185,161 +180,6 @@ impl Client {
         self.process_id
     }
 
-    pub(crate) fn inner(&self) -> &Arc<InnerClient> {
-        &self.inner
-    }
-
-    /// Creates a new prepared statement.
-    ///
-    /// Prepared statements can be executed repeatedly, and may contain query parameters (indicated by `$1`, `$2`, etc),
-    /// which are set when executed. Prepared statements can only be used with the connection that created them.
-    pub async fn prepare(&self, query: &str) -> Result<Statement, Error> {
-        self.prepare_typed(query, &[]).await
-    }
-
-    /// Like `prepare`, but allows the types of query parameters to be explicitly specified.
-    ///
-    /// The list of types may be smaller than the number of parameters - the types of the remaining parameters will be
-    /// inferred. For example, `client.prepare_typed(query, &[])` is equivalent to `client.prepare(query)`.
-    pub async fn prepare_typed(
-        &self,
-        query: &str,
-        parameter_types: &[Type],
-    ) -> Result<Statement, Error> {
-        prepare::prepare(&self.inner, query, parameter_types).await
-    }
-
-    /// Executes a statement, returning a vector of the resulting rows.
-    ///
-    /// A statement may contain parameters, specified by `$n`, where `n` is the index of the parameter of the list
-    /// provided, 1-indexed.
-    ///
-    /// The `statement` argument can either be a `Statement`, or a raw query string. If the same statement will be
-    /// repeatedly executed (perhaps with different query parameters), consider preparing the statement up front
-    /// with the `prepare` method.
-    ///
-    /// # Panics
-    ///
-    /// Panics if the number of parameters provided does not match the number expected.
-    pub async fn query<T>(
-        &self,
-        statement: &T,
-        params: &[&(dyn ToSql + Sync)],
-    ) -> Result<Vec<Row>, Error>
-    where
-        T: ?Sized + ToStatement,
-    {
-        self.query_raw(statement, slice_iter(params))
-            .await?
-            .try_collect()
-            .await
-    }
-
-    /// The maximally flexible version of [`query`].
-    ///
-    /// A statement may contain parameters, specified by `$n`, where `n` is the index of the parameter of the list
-    /// provided, 1-indexed.
-    ///
-    /// The `statement` argument can either be a `Statement`, or a raw query string. If the same statement will be
-    /// repeatedly executed (perhaps with different query parameters), consider preparing the statement up front
-    /// with the `prepare` method.
-    ///
-    /// # Panics
-    ///
-    /// Panics if the number of parameters provided does not match the number expected.
-    ///
-    /// [`query`]: #method.query
-    pub async fn query_raw<'a, T, I>(&self, statement: &T, params: I) -> Result<RowStream, Error>
-    where
-        T: ?Sized + ToStatement,
-        I: IntoIterator<Item = &'a (dyn ToSql + Sync)>,
-        I::IntoIter: ExactSizeIterator,
-    {
-        let statement = statement.__convert().into_statement(self).await?;
-        query::query(&self.inner, statement, params).await
-    }
-
-    /// Pass text directly to the Postgres backend to allow it to sort out typing itself and
-    /// to save a roundtrip
-    pub async fn query_raw_txt<S, I>(&self, statement: &str, params: I) -> Result<RowStream, Error>
-    where
-        S: AsRef<str>,
-        I: IntoIterator<Item = Option<S>>,
-        I::IntoIter: ExactSizeIterator,
-    {
-        query::query_txt(&self.inner, statement, params).await
-    }
-
-    /// Executes a statement, returning the number of rows modified.
-    ///
-    /// A statement may contain parameters, specified by `$n`, where `n` is the index of the parameter of the list
-    /// provided, 1-indexed.
-    ///
-    /// The `statement` argument can either be a `Statement`, or a raw query string. If the same statement will be
-    /// repeatedly executed (perhaps with different query parameters), consider preparing the statement up front
-    /// with the `prepare` method.
-    ///
-    /// If the statement does not modify any rows (e.g. `SELECT`), 0 is returned.
-    ///
-    /// # Panics
-    ///
-    /// Panics if the number of parameters provided does not match the number expected.
-    pub async fn execute<T>(
-        &self,
-        statement: &T,
-        params: &[&(dyn ToSql + Sync)],
-    ) -> Result<u64, Error>
-    where
-        T: ?Sized + ToStatement,
-    {
-        self.execute_raw(statement, slice_iter(params)).await
-    }
-
-    /// The maximally flexible version of [`execute`].
-    ///
-    /// A statement may contain parameters, specified by `$n`, where `n` is the index of the parameter of the list
-    /// provided, 1-indexed.
-    ///
-    /// The `statement` argument can either be a `Statement`, or a raw query string. If the same statement will be
-    /// repeatedly executed (perhaps with different query parameters), consider preparing the statement up front
-    /// with the `prepare` method.
-    ///
-    /// # Panics
-    ///
-    /// Panics if the number of parameters provided does not match the number expected.
-    ///
-    /// [`execute`]: #method.execute
-    pub async fn execute_raw<'a, T, I>(&self, statement: &T, params: I) -> Result<u64, Error>
-    where
-        T: ?Sized + ToStatement,
-        I: IntoIterator<Item = &'a (dyn ToSql + Sync)>,
-        I::IntoIter: ExactSizeIterator,
-    {
-        let statement = statement.__convert().into_statement(self).await?;
-        query::execute(self.inner(), statement, params).await
-    }
-
-    /// Executes a sequence of SQL statements using the simple query protocol, returning the resulting rows.
-    ///
-    /// Statements should be separated by semicolons. If an error occurs, execution of the sequence will stop at that
-    /// point. The simple query protocol returns the values in rows as strings rather than in their binary encodings,
-    /// so the associated row type doesn't work with the `FromSql` trait. Rather than simply returning a list of the
-    /// rows, this method returns a list of an enum which indicates either the completion of one of the commands,
-    /// or a row of data. This preserves the framing between the separate statements in the request.
-    ///
-    /// # Warning
-    ///
-    /// Prepared statements should be use for any query which contains user-specified data, as they provided the
-    /// functionality to safely embed that data in the request. Do not form statements via string concatenation and pass
-    /// them to this method!
-    pub async fn simple_query(&self, query: &str) -> Result<Vec<SimpleQueryMessage>, Error> {
-        self.simple_query_raw(query).await?.try_collect().await
-    }
-
-    pub(crate) async fn simple_query_raw(&self, query: &str) -> Result<SimpleQueryStream, Error> {
-        simple_query::simple_query(self.inner(), query).await
-    }
-
     /// Executes a sequence of SQL statements using the simple query protocol.
     ///
     /// Statements should be separated by semicolons. If an error occurs, execution of the sequence will stop at that
@@ -350,8 +190,8 @@ impl Client {
     /// Prepared statements should be use for any query which contains user-specified data, as they provided the
     /// functionality to safely embed that data in the request. Do not form statements via string concatenation and pass
     /// them to this method!
-    pub async fn batch_execute(&self, query: &str) -> Result<ReadyForQueryStatus, Error> {
+    pub async fn batch_execute(&mut self, query: &str) -> Result<ReadyForQueryStatus, Error> {
-        simple_query::batch_execute(self.inner(), query).await
+        simple_query::batch_execute(&mut self.inner, query).await
     }
 
     /// Begins a new database transaction.
@@ -359,7 +199,7 @@ impl Client {
     /// The transaction will roll back by default - use the `commit` method to commit it.
     pub async fn transaction(&mut self) -> Result<Transaction<'_>, Error> {
         struct RollbackIfNotDone<'me> {
-            client: &'me Client,
+            client: &'me mut Client,
             done: bool,
         }
 
@@ -369,13 +209,13 @@ impl Client {
                     return;
                 }
 
-                let buf = self.client.inner().with_buf(|buf| {
+                let buf = self.client.inner.with_buf(|buf| {
                     frontend::query("ROLLBACK", buf).unwrap();
                     buf.split().freeze()
                 });
                 let _ = self
                     .client
-                    .inner()
+                    .inner
                     .send(RequestMessages::Single(FrontendMessage::Raw(buf)));
             }
         }
@@ -390,7 +230,7 @@ impl Client {
                 client: self,
                 done: false,
             };
-            self.batch_execute("BEGIN").await?;
+            cleaner.client.batch_execute("BEGIN").await?;
             cleaner.done = true;
         }
 
@@ -416,11 +256,6 @@ impl Client {
         }
     }
 
-    /// Query for type information
-    pub async fn get_type(&self, oid: Oid) -> Result<Type, Error> {
-        crate::prepare::get_type(&self.inner, oid).await
-    }
-
     /// Determines if the connection to the server has already closed.
     ///
     /// In that case, all future queries will fail.

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

@@ -1,4 +1,4 @@
-use crate::query::RowStream;
+use crate::query::{self, RowStream};
 use crate::types::Type;
 use crate::{Client, Error, Transaction};
 use async_trait::async_trait;
@@ -13,33 +13,32 @@ mod private {
 /// This trait is "sealed", and cannot be implemented outside of this crate.
 #[async_trait]
 pub trait GenericClient: private::Sealed {
-    /// Like `Client::query_raw_txt`.
+    async fn query_raw_txt<S, I>(&mut self, statement: &str, params: I) -> Result<RowStream, Error>
-    async fn query_raw_txt<S, I>(&self, statement: &str, params: I) -> Result<RowStream, Error>
     where
         S: AsRef<str> + Sync + Send,
         I: IntoIterator<Item = Option<S>> + Sync + Send,
         I::IntoIter: ExactSizeIterator + Sync + Send;
 
     /// Query for type information
-    async fn get_type(&self, oid: Oid) -> Result<Type, Error>;
+    async fn get_type(&mut self, oid: Oid) -> Result<Type, Error>;
 }
 
 impl private::Sealed for Client {}
 
 #[async_trait]
 impl GenericClient for Client {
-    async fn query_raw_txt<S, I>(&self, statement: &str, params: I) -> Result<RowStream, Error>
+    async fn query_raw_txt<S, I>(&mut self, statement: &str, params: I) -> Result<RowStream, Error>
     where
         S: AsRef<str> + Sync + Send,
         I: IntoIterator<Item = Option<S>> + Sync + Send,
         I::IntoIter: ExactSizeIterator + Sync + Send,
     {
-        self.query_raw_txt(statement, params).await
+        query::query_txt(&mut self.inner, statement, params).await
     }
 
     /// Query for type information
-    async fn get_type(&self, oid: Oid) -> Result<Type, Error> {
+    async fn get_type(&mut self, oid: Oid) -> Result<Type, Error> {
-        self.get_type(oid).await
+        crate::prepare::get_type(&mut self.inner, &mut self.cached_typeinfo, oid).await
     }
 }
 
@@ -48,17 +47,18 @@ impl private::Sealed for Transaction<'_> {}
 #[async_trait]
 #[allow(clippy::needless_lifetimes)]
 impl GenericClient for Transaction<'_> {
-    async fn query_raw_txt<S, I>(&self, statement: &str, params: I) -> Result<RowStream, Error>
+    async fn query_raw_txt<S, I>(&mut self, statement: &str, params: I) -> Result<RowStream, Error>
     where
         S: AsRef<str> + Sync + Send,
         I: IntoIterator<Item = Option<S>> + Sync + Send,
         I::IntoIter: ExactSizeIterator + Sync + Send,
     {
-        self.query_raw_txt(statement, params).await
+        query::query_txt(&mut self.client().inner, statement, params).await
     }
 
     /// Query for type information
-    async fn get_type(&self, oid: Oid) -> Result<Type, Error> {
+    async fn get_type(&mut self, oid: Oid) -> Result<Type, Error> {
-        self.client().get_type(oid).await
+        let client = self.client();
+        crate::prepare::get_type(&mut client.inner, &mut client.cached_typeinfo, oid).await
     }
 }

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

@@ -1,5 +1,5 @@
 //! An asynchronous, pipelined, PostgreSQL client.
-#![warn(clippy::all)]
+#![warn(rust_2018_idioms, clippy::all)]
 
 pub use crate::cancel_token::CancelToken;
 pub use crate::client::{Client, SocketConfig};
@@ -10,11 +10,10 @@ use crate::error::DbError;
 pub use crate::error::Error;
 pub use crate::generic_client::GenericClient;
 pub use crate::query::RowStream;
-pub use crate::row::{Row, SimpleQueryRow};
+pub use crate::row::Row;
-pub use crate::simple_query::SimpleQueryStream;
 pub use crate::statement::{Column, Statement};
 pub use crate::tls::NoTls;
-pub use crate::to_statement::ToStatement;
+// pub use crate::to_statement::ToStatement;
 pub use crate::transaction::Transaction;
 pub use crate::transaction_builder::{IsolationLevel, TransactionBuilder};
 use crate::types::ToSql;
@@ -65,7 +64,7 @@ pub mod row;
 mod simple_query;
 mod statement;
 pub mod tls;
-mod to_statement;
+// mod to_statement;
 mod transaction;
 mod transaction_builder;
 pub mod types;
@@ -98,7 +97,6 @@ impl Notification {
 /// An asynchronous message from the server.
 #[allow(clippy::large_enum_variant)]
 #[derive(Debug, Clone)]
-#[non_exhaustive]
 pub enum AsyncMessage {
     /// A notice.
     ///
@@ -110,18 +108,6 @@ pub enum AsyncMessage {
     Notification(Notification),
 }
 
-/// Message returned by the `SimpleQuery` stream.
-#[derive(Debug)]
-#[non_exhaustive]
-pub enum SimpleQueryMessage {
-    /// A row of data.
-    Row(SimpleQueryRow),
-    /// A statement in the query has completed.
-    ///
-    /// The number of rows modified or selected is returned.
-    CommandComplete(u64),
-}
-
 fn slice_iter<'a>(
     s: &'a [&'a (dyn ToSql + Sync)],
 ) -> impl ExactSizeIterator<Item = &'a (dyn ToSql + Sync)> + 'a {

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

@@ -1,4 +1,4 @@
-use crate::client::InnerClient;
+use crate::client::{CachedTypeInfo, InnerClient};
 use crate::codec::FrontendMessage;
 use crate::connection::RequestMessages;
 use crate::error::SqlState;
@@ -7,14 +7,13 @@ use crate::{query, slice_iter};
 use crate::{Column, Error, Statement};
 use bytes::Bytes;
 use fallible_iterator::FallibleIterator;
-use futures_util::{pin_mut, TryStreamExt};
+use futures_util::{pin_mut, StreamExt, TryStreamExt};
 use log::debug;
 use postgres_protocol2::message::backend::Message;
 use postgres_protocol2::message::frontend;
 use std::future::Future;
-use std::pin::Pin;
+use std::pin::{pin, Pin};
 use std::sync::atomic::{AtomicUsize, Ordering};
-use std::sync::Arc;
 
 pub(crate) const TYPEINFO_QUERY: &str = "\
 SELECT t.typname, t.typtype, t.typelem, r.rngsubtype, t.typbasetype, n.nspname, t.typrelid
@@ -59,7 +58,8 @@ ORDER BY attnum
 static NEXT_ID: AtomicUsize = AtomicUsize::new(0);
 
 pub async fn prepare(
-    client: &Arc<InnerClient>,
+    client: &mut InnerClient,
+    cache: &mut CachedTypeInfo,
     query: &str,
     types: &[Type],
 ) -> Result<Statement, Error> {
@@ -86,7 +86,7 @@ pub async fn prepare(
     let mut parameters = vec![];
     let mut it = parameter_description.parameters();
     while let Some(oid) = it.next().map_err(Error::parse)? {
-        let type_ = get_type(client, oid).await?;
+        let type_ = get_type(client, cache, oid).await?;
         parameters.push(type_);
     }
 
@@ -94,24 +94,30 @@ pub async fn prepare(
     if let Some(row_description) = row_description {
         let mut it = row_description.fields();
         while let Some(field) = it.next().map_err(Error::parse)? {
-            let type_ = get_type(client, field.type_oid()).await?;
+            let type_ = get_type(client, cache, field.type_oid()).await?;
             let column = Column::new(field.name().to_string(), type_, field);
             columns.push(column);
         }
     }
 
-    Ok(Statement::new(client, name, parameters, columns))
+    Ok(Statement::new(name, parameters, columns))
 }
 
 fn prepare_rec<'a>(
-    client: &'a Arc<InnerClient>,
+    client: &'a mut InnerClient,
+    cache: &'a mut CachedTypeInfo,
     query: &'a str,
     types: &'a [Type],
 ) -> Pin<Box<dyn Future<Output = Result<Statement, Error>> + 'a + Send>> {
-    Box::pin(prepare(client, query, types))
+    Box::pin(prepare(client, cache, query, types))
 }
 
-fn encode(client: &InnerClient, name: &str, query: &str, types: &[Type]) -> Result<Bytes, Error> {
+fn encode(
+    client: &mut InnerClient,
+    name: &str,
+    query: &str,
+    types: &[Type],
+) -> Result<Bytes, Error> {
     if types.is_empty() {
         debug!("preparing query {}: {}", name, query);
     } else {
@@ -126,16 +132,20 @@ fn encode(client: &InnerClient, name: &str, query: &str, types: &[Type]) -> Resu
     })
 }
 
-pub async fn get_type(client: &Arc<InnerClient>, oid: Oid) -> Result<Type, Error> {
+pub async fn get_type(
+    client: &mut InnerClient,
+    cache: &mut CachedTypeInfo,
+    oid: Oid,
+) -> Result<Type, Error> {
     if let Some(type_) = Type::from_oid(oid) {
         return Ok(type_);
     }
 
-    if let Some(type_) = client.type_(oid) {
+    if let Some(type_) = cache.type_(oid) {
         return Ok(type_);
     }
 
-    let stmt = typeinfo_statement(client).await?;
+    let stmt = typeinfo_statement(client, cache).await?;
 
     let rows = query::query(client, stmt, slice_iter(&[&oid])).await?;
     pin_mut!(rows);
@@ -145,118 +155,141 @@ pub async fn get_type(client: &Arc<InnerClient>, oid: Oid) -> Result<Type, Error
         None => return Err(Error::unexpected_message()),
     };
 
-    let name: String = row.try_get(0)?;
+    let name: String = row.try_get(stmt.columns(), 0)?;
-    let type_: i8 = row.try_get(1)?;
+    let type_: i8 = row.try_get(stmt.columns(), 1)?;
-    let elem_oid: Oid = row.try_get(2)?;
+    let elem_oid: Oid = row.try_get(stmt.columns(), 2)?;
-    let rngsubtype: Option<Oid> = row.try_get(3)?;
+    let rngsubtype: Option<Oid> = row.try_get(stmt.columns(), 3)?;
-    let basetype: Oid = row.try_get(4)?;
+    let basetype: Oid = row.try_get(stmt.columns(), 4)?;
-    let schema: String = row.try_get(5)?;
+    let schema: String = row.try_get(stmt.columns(), 5)?;
-    let relid: Oid = row.try_get(6)?;
+    let relid: Oid = row.try_get(stmt.columns(), 6)?;
 
     let kind = if type_ == b'e' as i8 {
-        let variants = get_enum_variants(client, oid).await?;
+        let variants = get_enum_variants(client, cache, oid).await?;
         Kind::Enum(variants)
     } else if type_ == b'p' as i8 {
         Kind::Pseudo
     } else if basetype != 0 {
-        let type_ = get_type_rec(client, basetype).await?;
+        let type_ = get_type_rec(client, cache, basetype).await?;
         Kind::Domain(type_)
     } else if elem_oid != 0 {
-        let type_ = get_type_rec(client, elem_oid).await?;
+        let type_ = get_type_rec(client, cache, elem_oid).await?;
         Kind::Array(type_)
     } else if relid != 0 {
-        let fields = get_composite_fields(client, relid).await?;
+        let fields = get_composite_fields(client, cache, relid).await?;
         Kind::Composite(fields)
     } else if let Some(rngsubtype) = rngsubtype {
-        let type_ = get_type_rec(client, rngsubtype).await?;
+        let type_ = get_type_rec(client, cache, rngsubtype).await?;
         Kind::Range(type_)
     } else {
         Kind::Simple
     };
 
     let type_ = Type::new(name, oid, kind, schema);
-    client.set_type(oid, &type_);
+    cache.set_type(oid, &type_);
 
     Ok(type_)
 }
 
 fn get_type_rec<'a>(
-    client: &'a Arc<InnerClient>,
+    client: &'a mut InnerClient,
+    cache: &'a mut CachedTypeInfo,
     oid: Oid,
 ) -> Pin<Box<dyn Future<Output = Result<Type, Error>> + Send + 'a>> {
-    Box::pin(get_type(client, oid))
+    Box::pin(get_type(client, cache, oid))
 }
 
-async fn typeinfo_statement(client: &Arc<InnerClient>) -> Result<Statement, Error> {
+async fn typeinfo_statement<'c>(
-    if let Some(stmt) = client.typeinfo() {
+    client: &mut InnerClient,
-        return Ok(stmt);
+    cache: &'c mut CachedTypeInfo,
+) -> Result<&'c Statement, Error> {
+    if cache.typeinfo().is_some() {
+        // needed to get around a borrow checker limitation
+        return Ok(cache.typeinfo().unwrap());
     }
 
-    let stmt = match prepare_rec(client, TYPEINFO_QUERY, &[]).await {
+    let stmt = match prepare_rec(client, cache, TYPEINFO_QUERY, &[]).await {
         Ok(stmt) => stmt,
         Err(ref e) if e.code() == Some(&SqlState::UNDEFINED_TABLE) => {
-            prepare_rec(client, TYPEINFO_FALLBACK_QUERY, &[]).await?
+            prepare_rec(client, cache, TYPEINFO_FALLBACK_QUERY, &[]).await?
         }
         Err(e) => return Err(e),
     };
 
-    client.set_typeinfo(&stmt);
+    Ok(cache.set_typeinfo(stmt))
-    Ok(stmt)
 }
 
-async fn get_enum_variants(client: &Arc<InnerClient>, oid: Oid) -> Result<Vec<String>, Error> {
+async fn get_enum_variants(
-    let stmt = typeinfo_enum_statement(client).await?;
+    client: &mut InnerClient,
+    cache: &mut CachedTypeInfo,
+    oid: Oid,
+) -> Result<Vec<String>, Error> {
+    let stmt = typeinfo_enum_statement(client, cache).await?;
 
-    query::query(client, stmt, slice_iter(&[&oid]))
+    let mut out = vec![];
-        .await?
+
-        .and_then(|row| async move { row.try_get(0) })
+    let mut rows = pin!(query::query(client, stmt, slice_iter(&[&oid])).await?);
-        .try_collect()
+    while let Some(row) = rows.next().await {
-        .await
+        out.push(row?.try_get(stmt.columns(), 0)?)
+    }
+    Ok(out)
 }
 
-async fn typeinfo_enum_statement(client: &Arc<InnerClient>) -> Result<Statement, Error> {
+async fn typeinfo_enum_statement<'c>(
-    if let Some(stmt) = client.typeinfo_enum() {
+    client: &mut InnerClient,
-        return Ok(stmt);
+    cache: &'c mut CachedTypeInfo,
+) -> Result<&'c Statement, Error> {
+    if cache.typeinfo_enum().is_some() {
+        // needed to get around a borrow checker limitation
+        return Ok(cache.typeinfo_enum().unwrap());
     }
 
-    let stmt = match prepare_rec(client, TYPEINFO_ENUM_QUERY, &[]).await {
+    let stmt = match prepare_rec(client, cache, TYPEINFO_ENUM_QUERY, &[]).await {
         Ok(stmt) => stmt,
         Err(ref e) if e.code() == Some(&SqlState::UNDEFINED_COLUMN) => {
-            prepare_rec(client, TYPEINFO_ENUM_FALLBACK_QUERY, &[]).await?
+            prepare_rec(client, cache, TYPEINFO_ENUM_FALLBACK_QUERY, &[]).await?
         }
         Err(e) => return Err(e),
     };
 
-    client.set_typeinfo_enum(&stmt);
+    Ok(cache.set_typeinfo_enum(stmt))
-    Ok(stmt)
 }
 
-async fn get_composite_fields(client: &Arc<InnerClient>, oid: Oid) -> Result<Vec<Field>, Error> {
+async fn get_composite_fields(
-    let stmt = typeinfo_composite_statement(client).await?;
+    client: &mut InnerClient,
+    cache: &mut CachedTypeInfo,
+    oid: Oid,
+) -> Result<Vec<Field>, Error> {
+    let stmt = typeinfo_composite_statement(client, cache).await?;
 
-    let rows = query::query(client, stmt, slice_iter(&[&oid]))
+    let mut rows = pin!(query::query(client, stmt, slice_iter(&[&oid])).await?);
-        .await?
+
-        .try_collect::<Vec<_>>()
+    let mut oids = vec![];
-        .await?;
+    while let Some(row) = rows.next().await {
+        let row = row?;
+        let name = row.try_get(stmt.columns(), 0)?;
+        let oid = row.try_get(stmt.columns(), 1)?;
+        oids.push((name, oid));
+    }
 
     let mut fields = vec![];
-    for row in rows {
+    for (name, oid) in oids {
-        let name = row.try_get(0)?;
+        let type_ = get_type_rec(client, cache, oid).await?;
-        let oid = row.try_get(1)?;
-        let type_ = get_type_rec(client, oid).await?;
         fields.push(Field::new(name, type_));
     }
 
     Ok(fields)
 }
 
-async fn typeinfo_composite_statement(client: &Arc<InnerClient>) -> Result<Statement, Error> {
+async fn typeinfo_composite_statement<'c>(
-    if let Some(stmt) = client.typeinfo_composite() {
+    client: &mut InnerClient,
-        return Ok(stmt);
+    cache: &'c mut CachedTypeInfo,
+) -> Result<&'c Statement, Error> {
+    if cache.typeinfo_composite().is_some() {
+        // needed to get around a borrow checker limitation
+        return Ok(cache.typeinfo_composite().unwrap());
     }
 
-    let stmt = prepare_rec(client, TYPEINFO_COMPOSITE_QUERY, &[]).await?;
+    let stmt = prepare_rec(client, cache, TYPEINFO_COMPOSITE_QUERY, &[]).await?;
 
-    client.set_typeinfo_composite(&stmt);
+    Ok(cache.set_typeinfo_composite(stmt))
-    Ok(stmt)
 }

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

@@ -14,7 +14,6 @@ use postgres_types2::{Format, ToSql, Type};
 use std::fmt;
 use std::marker::PhantomPinned;
 use std::pin::Pin;
-use std::sync::Arc;
 use std::task::{Context, Poll};
 
 struct BorrowToSqlParamsDebug<'a>(&'a [&'a (dyn ToSql + Sync)]);
@@ -26,10 +25,10 @@ impl fmt::Debug for BorrowToSqlParamsDebug<'_> {
 }
 
 pub async fn query<'a, I>(
-    client: &InnerClient,
+    client: &mut InnerClient,
-    statement: Statement,
+    statement: &Statement,
     params: I,
-) -> Result<RowStream, Error>
+) -> Result<RawRowStream, Error>
 where
     I: IntoIterator<Item = &'a (dyn ToSql + Sync)>,
     I::IntoIter: ExactSizeIterator,
@@ -41,13 +40,12 @@ where
             statement.name(),
             BorrowToSqlParamsDebug(params.as_slice()),
         );
-        encode(client, &statement, params)?
+        encode(client, statement, params)?
     } else {
-        encode(client, &statement, params)?
+        encode(client, statement, params)?
     };
     let responses = start(client, buf).await?;
-    Ok(RowStream {
+    Ok(RawRowStream {
-        statement,
         responses,
         command_tag: None,
         status: ReadyForQueryStatus::Unknown,
@@ -57,7 +55,7 @@ where
 }
 
 pub async fn query_txt<S, I>(
-    client: &Arc<InnerClient>,
+    client: &mut InnerClient,
     query: &str,
     params: I,
 ) -> Result<RowStream, Error>
@@ -157,49 +155,6 @@ where
     })
 }
 
-pub async fn execute<'a, I>(
-    client: &InnerClient,
-    statement: Statement,
-    params: I,
-) -> Result<u64, Error>
-where
-    I: IntoIterator<Item = &'a (dyn ToSql + Sync)>,
-    I::IntoIter: ExactSizeIterator,
-{
-    let buf = if log_enabled!(Level::Debug) {
-        let params = params.into_iter().collect::<Vec<_>>();
-        debug!(
-            "executing statement {} with parameters: {:?}",
-            statement.name(),
-            BorrowToSqlParamsDebug(params.as_slice()),
-        );
-        encode(client, &statement, params)?
-    } else {
-        encode(client, &statement, params)?
-    };
-    let mut responses = start(client, buf).await?;
-
-    let mut rows = 0;
-    loop {
-        match responses.next().await? {
-            Message::DataRow(_) => {}
-            Message::CommandComplete(body) => {
-                rows = body
-                    .tag()
-                    .map_err(Error::parse)?
-                    .rsplit(' ')
-                    .next()
-                    .unwrap()
-                    .parse()
-                    .unwrap_or(0);
-            }
-            Message::EmptyQueryResponse => rows = 0,
-            Message::ReadyForQuery(_) => return Ok(rows),
-            _ => return Err(Error::unexpected_message()),
-        }
-    }
-}
-
 async fn start(client: &InnerClient, buf: Bytes) -> Result<Responses, Error> {
     let mut responses = client.send(RequestMessages::Single(FrontendMessage::Raw(buf)))?;
 
@@ -211,7 +166,11 @@ async fn start(client: &InnerClient, buf: Bytes) -> Result<Responses, Error> {
     Ok(responses)
 }
 
-pub fn encode<'a, I>(client: &InnerClient, statement: &Statement, params: I) -> Result<Bytes, Error>
+pub fn encode<'a, I>(
+    client: &mut InnerClient,
+    statement: &Statement,
+    params: I,
+) -> Result<Bytes, Error>
 where
     I: IntoIterator<Item = &'a (dyn ToSql + Sync)>,
     I::IntoIter: ExactSizeIterator,
@@ -296,11 +255,7 @@ impl Stream for RowStream {
         loop {
             match ready!(this.responses.poll_next(cx)?) {
                 Message::DataRow(body) => {
-                    return Poll::Ready(Some(Ok(Row::new(
+                    return Poll::Ready(Some(Ok(Row::new(body, *this.output_format)?)))
-                        this.statement.clone(),
-                        body,
-                        *this.output_format,
-                    )?)))
                 }
                 Message::EmptyQueryResponse | Message::PortalSuspended => {}
                 Message::CommandComplete(body) => {
@@ -338,3 +293,41 @@ impl RowStream {
         self.status
     }
 }
+
+pin_project! {
+    /// A stream of table rows.
+    pub struct RawRowStream {
+        responses: Responses,
+        command_tag: Option<String>,
+        output_format: Format,
+        status: ReadyForQueryStatus,
+        #[pin]
+        _p: PhantomPinned,
+    }
+}
+
+impl Stream for RawRowStream {
+    type Item = Result<Row, Error>;
+
+    fn poll_next(self: Pin<&mut Self>, cx: &mut Context<'_>) -> Poll<Option<Self::Item>> {
+        let this = self.project();
+        loop {
+            match ready!(this.responses.poll_next(cx)?) {
+                Message::DataRow(body) => {
+                    return Poll::Ready(Some(Ok(Row::new(body, *this.output_format)?)))
+                }
+                Message::EmptyQueryResponse | Message::PortalSuspended => {}
+                Message::CommandComplete(body) => {
+                    if let Ok(tag) = body.tag() {
+                        *this.command_tag = Some(tag.to_string());
+                    }
+                }
+                Message::ReadyForQuery(status) => {
+                    *this.status = status.into();
+                    return Poll::Ready(None);
+                }
+                _ => return Poll::Ready(Some(Err(Error::unexpected_message()))),
+            }
+        }
+    }
+}

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

@@ -1,103 +1,16 @@
 //! Rows.
-
-use crate::row::sealed::{AsName, Sealed};
-use crate::simple_query::SimpleColumn;
 use crate::statement::Column;
 use crate::types::{FromSql, Type, WrongType};
-use crate::{Error, Statement};
+use crate::Error;
 use fallible_iterator::FallibleIterator;
 use postgres_protocol2::message::backend::DataRowBody;
 use postgres_types2::{Format, WrongFormat};
 use std::fmt;
 use std::ops::Range;
 use std::str;
-use std::sync::Arc;
-
-mod sealed {
-    pub trait Sealed {}
-
-    pub trait AsName {
-        fn as_name(&self) -> &str;
-    }
-}
-
-impl AsName for Column {
-    fn as_name(&self) -> &str {
-        self.name()
-    }
-}
-
-impl AsName for String {
-    fn as_name(&self) -> &str {
-        self
-    }
-}
-
-/// A trait implemented by types that can index into columns of a row.
-///
-/// This cannot be implemented outside of this crate.
-pub trait RowIndex: Sealed {
-    #[doc(hidden)]
-    fn __idx<T>(&self, columns: &[T]) -> Option<usize>
-    where
-        T: AsName;
-}
-
-impl Sealed for usize {}
-
-impl RowIndex for usize {
-    #[inline]
-    fn __idx<T>(&self, columns: &[T]) -> Option<usize>
-    where
-        T: AsName,
-    {
-        if *self >= columns.len() {
-            None
-        } else {
-            Some(*self)
-        }
-    }
-}
-
-impl Sealed for str {}
-
-impl RowIndex for str {
-    #[inline]
-    fn __idx<T>(&self, columns: &[T]) -> Option<usize>
-    where
-        T: AsName,
-    {
-        if let Some(idx) = columns.iter().position(|d| d.as_name() == self) {
-            return Some(idx);
-        };
-
-        // FIXME ASCII-only case insensitivity isn't really the right thing to
-        // do. Postgres itself uses a dubious wrapper around tolower and JDBC
-        // uses the US locale.
-        columns
-            .iter()
-            .position(|d| d.as_name().eq_ignore_ascii_case(self))
-    }
-}
-
-impl<T> Sealed for &T where T: ?Sized + Sealed {}
-
-impl<T> RowIndex for &T
-where
-    T: ?Sized + RowIndex,
-{
-    #[inline]
-    fn __idx<U>(&self, columns: &[U]) -> Option<usize>
-    where
-        U: AsName,
-    {
-        T::__idx(*self, columns)
-    }
-}
 
 /// A row of data returned from the database by a query.
 pub struct Row {
-    statement: Statement,
     output_format: Format,
     body: DataRowBody,
     ranges: Vec<Option<Range<usize>>>,
@@ -105,80 +18,33 @@ pub struct Row {
 
 impl fmt::Debug for Row {
     fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
-        f.debug_struct("Row")
+        f.debug_struct("Row").finish()
-            .field("columns", &self.columns())
-            .finish()
     }
 }
 
 impl Row {
     pub(crate) fn new(
-        statement: Statement,
+        // statement: Statement,
         body: DataRowBody,
         output_format: Format,
     ) -> Result<Row, Error> {
         let ranges = body.ranges().collect().map_err(Error::parse)?;
         Ok(Row {
-            statement,
             body,
             ranges,
             output_format,
         })
     }
 
-    /// Returns information about the columns of data in the row.
+    pub(crate) fn try_get<'a, T>(&'a self, columns: &[Column], idx: usize) -> Result<T, Error>
-    pub fn columns(&self) -> &[Column] {
-        self.statement.columns()
-    }
-
-    /// Determines if the row contains no values.
-    pub fn is_empty(&self) -> bool {
-        self.len() == 0
-    }
-
-    /// Returns the number of values in the row.
-    pub fn len(&self) -> usize {
-        self.columns().len()
-    }
-
-    /// Deserializes a value from the row.
-    ///
-    /// The value can be specified either by its numeric index in the row, or by its column name.
-    ///
-    /// # Panics
-    ///
-    /// Panics if the index is out of bounds or if the value cannot be converted to the specified type.
-    pub fn get<'a, I, T>(&'a self, idx: I) -> T
     where
-        I: RowIndex + fmt::Display,
         T: FromSql<'a>,
     {
-        match self.get_inner(&idx) {
+        let Some(column) = columns.get(idx) else {
-            Ok(ok) => ok,
+            return Err(Error::column(idx.to_string()));
-            Err(err) => panic!("error retrieving column {}: {}", idx, err),
-        }
-    }
-
-    /// Like `Row::get`, but returns a `Result` rather than panicking.
-    pub fn try_get<'a, I, T>(&'a self, idx: I) -> Result<T, Error>
-    where
-        I: RowIndex + fmt::Display,
-        T: FromSql<'a>,
-    {
-        self.get_inner(&idx)
-    }
-
-    fn get_inner<'a, I, T>(&'a self, idx: &I) -> Result<T, Error>
-    where
-        I: RowIndex + fmt::Display,
-        T: FromSql<'a>,
-    {
-        let idx = match idx.__idx(self.columns()) {
-            Some(idx) => idx,
-            None => return Err(Error::column(idx.to_string())),
         };
 
-        let ty = self.columns()[idx].type_();
+        let ty = column.type_();
         if !T::accepts(ty) {
             return Err(Error::from_sql(
                 Box::new(WrongType::new::<T>(ty.clone())),
@@ -216,85 +82,3 @@ impl Row {
         self.body.buffer().len()
     }
 }
-
-impl AsName for SimpleColumn {
-    fn as_name(&self) -> &str {
-        self.name()
-    }
-}
-
-/// A row of data returned from the database by a simple query.
-#[derive(Debug)]
-pub struct SimpleQueryRow {
-    columns: Arc<[SimpleColumn]>,
-    body: DataRowBody,
-    ranges: Vec<Option<Range<usize>>>,
-}
-
-impl SimpleQueryRow {
-    #[allow(clippy::new_ret_no_self)]
-    pub(crate) fn new(
-        columns: Arc<[SimpleColumn]>,
-        body: DataRowBody,
-    ) -> Result<SimpleQueryRow, Error> {
-        let ranges = body.ranges().collect().map_err(Error::parse)?;
-        Ok(SimpleQueryRow {
-            columns,
-            body,
-            ranges,
-        })
-    }
-
-    /// Returns information about the columns of data in the row.
-    pub fn columns(&self) -> &[SimpleColumn] {
-        &self.columns
-    }
-
-    /// Determines if the row contains no values.
-    pub fn is_empty(&self) -> bool {
-        self.len() == 0
-    }
-
-    /// Returns the number of values in the row.
-    pub fn len(&self) -> usize {
-        self.columns.len()
-    }
-
-    /// Returns a value from the row.
-    ///
-    /// The value can be specified either by its numeric index in the row, or by its column name.
-    ///
-    /// # Panics
-    ///
-    /// Panics if the index is out of bounds or if the value cannot be converted to the specified type.
-    pub fn get<I>(&self, idx: I) -> Option<&str>
-    where
-        I: RowIndex + fmt::Display,
-    {
-        match self.get_inner(&idx) {
-            Ok(ok) => ok,
-            Err(err) => panic!("error retrieving column {}: {}", idx, err),
-        }
-    }
-
-    /// Like `SimpleQueryRow::get`, but returns a `Result` rather than panicking.
-    pub fn try_get<I>(&self, idx: I) -> Result<Option<&str>, Error>
-    where
-        I: RowIndex + fmt::Display,
-    {
-        self.get_inner(&idx)
-    }
-
-    fn get_inner<I>(&self, idx: &I) -> Result<Option<&str>, Error>
-    where
-        I: RowIndex + fmt::Display,
-    {
-        let idx = match idx.__idx(&self.columns) {
-            Some(idx) => idx,
-            None => return Err(Error::column(idx.to_string())),
-        };
-
-        let buf = self.ranges[idx].clone().map(|r| &self.body.buffer()[r]);
-        FromSql::from_sql_nullable(&Type::TEXT, buf).map_err(|e| Error::from_sql(e, idx))
-    }
-}

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

@@ -1,52 +1,14 @@
-use crate::client::{InnerClient, Responses};
+use crate::client::InnerClient;
 use crate::codec::FrontendMessage;
 use crate::connection::RequestMessages;
-use crate::{Error, ReadyForQueryStatus, SimpleQueryMessage, SimpleQueryRow};
+use crate::{Error, ReadyForQueryStatus};
 use bytes::Bytes;
-use fallible_iterator::FallibleIterator;
-use futures_util::{ready, Stream};
 use log::debug;
-use pin_project_lite::pin_project;
 use postgres_protocol2::message::backend::Message;
 use postgres_protocol2::message::frontend;
-use std::marker::PhantomPinned;
-use std::pin::Pin;
-use std::sync::Arc;
-use std::task::{Context, Poll};
-
-/// Information about a column of a single query row.
-#[derive(Debug)]
-pub struct SimpleColumn {
-    name: String,
-}
-
-impl SimpleColumn {
-    pub(crate) fn new(name: String) -> SimpleColumn {
-        SimpleColumn { name }
-    }
-
-    /// Returns the name of the column.
-    pub fn name(&self) -> &str {
-        &self.name
-    }
-}
-
-pub async fn simple_query(client: &InnerClient, query: &str) -> Result<SimpleQueryStream, Error> {
-    debug!("executing simple query: {}", query);
-
-    let buf = encode(client, query)?;
-    let responses = client.send(RequestMessages::Single(FrontendMessage::Raw(buf)))?;
-
-    Ok(SimpleQueryStream {
-        responses,
-        columns: None,
-        status: ReadyForQueryStatus::Unknown,
-        _p: PhantomPinned,
-    })
-}
 
 pub async fn batch_execute(
-    client: &InnerClient,
+    client: &mut InnerClient,
     query: &str,
 ) -> Result<ReadyForQueryStatus, Error> {
     debug!("executing statement batch: {}", query);
@@ -66,77 +28,9 @@ pub async fn batch_execute(
     }
 }
 
-pub(crate) fn encode(client: &InnerClient, query: &str) -> Result<Bytes, Error> {
+pub(crate) fn encode(client: &mut InnerClient, query: &str) -> Result<Bytes, Error> {
     client.with_buf(|buf| {
         frontend::query(query, buf).map_err(Error::encode)?;
         Ok(buf.split().freeze())
     })
 }
-
-pin_project! {
-    /// A stream of simple query results.
-    pub struct SimpleQueryStream {
-        responses: Responses,
-        columns: Option<Arc<[SimpleColumn]>>,
-        status: ReadyForQueryStatus,
-        #[pin]
-        _p: PhantomPinned,
-    }
-}
-
-impl SimpleQueryStream {
-    /// Returns if the connection is ready for querying, with the status of the connection.
-    ///
-    /// This might be available only after the stream has been exhausted.
-    pub fn ready_status(&self) -> ReadyForQueryStatus {
-        self.status
-    }
-}
-
-impl Stream for SimpleQueryStream {
-    type Item = Result<SimpleQueryMessage, Error>;
-
-    fn poll_next(self: Pin<&mut Self>, cx: &mut Context<'_>) -> Poll<Option<Self::Item>> {
-        let this = self.project();
-        loop {
-            match ready!(this.responses.poll_next(cx)?) {
-                Message::CommandComplete(body) => {
-                    let rows = body
-                        .tag()
-                        .map_err(Error::parse)?
-                        .rsplit(' ')
-                        .next()
-                        .unwrap()
-                        .parse()
-                        .unwrap_or(0);
-                    return Poll::Ready(Some(Ok(SimpleQueryMessage::CommandComplete(rows))));
-                }
-                Message::EmptyQueryResponse => {
-                    return Poll::Ready(Some(Ok(SimpleQueryMessage::CommandComplete(0))));
-                }
-                Message::RowDescription(body) => {
-                    let columns = body
-                        .fields()
-                        .map(|f| Ok(SimpleColumn::new(f.name().to_string())))
-                        .collect::<Vec<_>>()
-                        .map_err(Error::parse)?
-                        .into();
-
-                    *this.columns = Some(columns);
-                }
-                Message::DataRow(body) => {
-                    let row = match &this.columns {
-                        Some(columns) => SimpleQueryRow::new(columns.clone(), body)?,
-                        None => return Poll::Ready(Some(Err(Error::unexpected_message()))),
-                    };
-                    return Poll::Ready(Some(Ok(SimpleQueryMessage::Row(row))));
-                }
-                Message::ReadyForQuery(s) => {
-                    *this.status = s.into();
-                    return Poll::Ready(None);
-                }
-                _ => return Poll::Ready(Some(Err(Error::unexpected_message()))),
-            }
-        }
-    }
-}

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

@@ -1,64 +1,33 @@
-use crate::client::InnerClient;
-use crate::codec::FrontendMessage;
-use crate::connection::RequestMessages;
 use crate::types::Type;
-use postgres_protocol2::{
+use postgres_protocol2::{message::backend::Field, Oid};
-    message::{backend::Field, frontend},
+use std::fmt;
-    Oid,
-};
-use std::{
-    fmt,
-    sync::{Arc, Weak},
-};
 
 struct StatementInner {
-    client: Weak<InnerClient>,
     name: String,
     params: Vec<Type>,
     columns: Vec<Column>,
 }
 
-impl Drop for StatementInner {
-    fn drop(&mut self) {
-        if let Some(client) = self.client.upgrade() {
-            let buf = client.with_buf(|buf| {
-                frontend::close(b'S', &self.name, buf).unwrap();
-                frontend::sync(buf);
-                buf.split().freeze()
-            });
-            let _ = client.send(RequestMessages::Single(FrontendMessage::Raw(buf)));
-        }
-    }
-}
-
 /// A prepared statement.
 ///
 /// Prepared statements can only be used with the connection that created them.
-#[derive(Clone)]
+pub struct Statement(StatementInner);
-pub struct Statement(Arc<StatementInner>);
 
 impl Statement {
-    pub(crate) fn new(
+    pub(crate) fn new(name: String, params: Vec<Type>, columns: Vec<Column>) -> Statement {
-        inner: &Arc<InnerClient>,
+        Statement(StatementInner {
-        name: String,
-        params: Vec<Type>,
-        columns: Vec<Column>,
-    ) -> Statement {
-        Statement(Arc::new(StatementInner {
-            client: Arc::downgrade(inner),
             name,
             params,
             columns,
-        }))
+        })
     }
 
     pub(crate) fn new_anonymous(params: Vec<Type>, columns: Vec<Column>) -> Statement {
-        Statement(Arc::new(StatementInner {
+        Statement(StatementInner {
-            client: Weak::new(),
             name: String::new(),
             params,
             columns,
-        }))
+        })
     }
 
     pub(crate) fn name(&self) -> &str {

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

@@ -1,57 +0,0 @@
-use crate::to_statement::private::{Sealed, ToStatementType};
-use crate::Statement;
-
-mod private {
-    use crate::{Client, Error, Statement};
-
-    pub trait Sealed {}
-
-    pub enum ToStatementType<'a> {
-        Statement(&'a Statement),
-        Query(&'a str),
-    }
-
-    impl ToStatementType<'_> {
-        pub async fn into_statement(self, client: &Client) -> Result<Statement, Error> {
-            match self {
-                ToStatementType::Statement(s) => Ok(s.clone()),
-                ToStatementType::Query(s) => client.prepare(s).await,
-            }
-        }
-    }
-}
-
-/// A trait abstracting over prepared and unprepared statements.
-///
-/// Many methods are generic over this bound, so that they support both a raw query string as well as a statement which
-/// was prepared previously.
-///
-/// This trait is "sealed" and cannot be implemented by anything outside this crate.
-pub trait ToStatement: Sealed {
-    #[doc(hidden)]
-    fn __convert(&self) -> ToStatementType<'_>;
-}
-
-impl ToStatement for Statement {
-    fn __convert(&self) -> ToStatementType<'_> {
-        ToStatementType::Statement(self)
-    }
-}
-
-impl Sealed for Statement {}
-
-impl ToStatement for str {
-    fn __convert(&self) -> ToStatementType<'_> {
-        ToStatementType::Query(self)
-    }
-}
-
-impl Sealed for str {}
-
-impl ToStatement for String {
-    fn __convert(&self) -> ToStatementType<'_> {
-        ToStatementType::Query(self)
-    }
-}
-
-impl Sealed for String {}

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

@@ -1,6 +1,5 @@
 use crate::codec::FrontendMessage;
 use crate::connection::RequestMessages;
-use crate::query::RowStream;
 use crate::{CancelToken, Client, Error, ReadyForQueryStatus};
 use postgres_protocol2::message::frontend;
 
@@ -19,13 +18,13 @@ impl Drop for Transaction<'_> {
             return;
         }
 
-        let buf = self.client.inner().with_buf(|buf| {
+        let buf = self.client.inner.with_buf(|buf| {
             frontend::query("ROLLBACK", buf).unwrap();
             buf.split().freeze()
         });
         let _ = self
             .client
-            .inner()
+            .inner
             .send(RequestMessages::Single(FrontendMessage::Raw(buf)));
     }
 }
@@ -52,23 +51,13 @@ impl<'a> Transaction<'a> {
         self.client.batch_execute("ROLLBACK").await
     }
 
-    /// Like `Client::query_raw_txt`.
-    pub async fn query_raw_txt<S, I>(&self, statement: &str, params: I) -> Result<RowStream, Error>
-    where
-        S: AsRef<str>,
-        I: IntoIterator<Item = Option<S>>,
-        I::IntoIter: ExactSizeIterator,
-    {
-        self.client.query_raw_txt(statement, params).await
-    }
-
     /// Like `Client::cancel_token`.
     pub fn cancel_token(&self) -> CancelToken {
         self.client.cancel_token()
     }
 
     /// Returns a reference to the underlying `Client`.
-    pub fn client(&self) -> &Client {
+    pub fn client(&mut self) -> &mut Client {
         self.client
     }
 }

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

@@ -13,12 +13,10 @@ use std::time::Duration;
 use std::time::SystemTime;
 
 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_identity::DefaultAzureCredential;
 use azure_storage::StorageCredentials;
 use azure_storage_blobs::blob::CopyStatus;
 use azure_storage_blobs::prelude::ClientBuilder;
@@ -78,18 +76,12 @@ impl AzureBlobStorage {
         let credentials = if let Ok(access_key) = env::var("AZURE_STORAGE_ACCESS_KEY") {
             StorageCredentials::access_key(account.clone(), access_key)
         } else {
-            let token_credential = azure_identity::create_default_credential()
+            let token_credential = DefaultAzureCredential::default();
-                .context("trying to obtain Azure default credentials")?;
+            StorageCredentials::token_credential(Arc::new(token_credential))
-            StorageCredentials::token_credential(token_credential)
         };
 
-        let builder = ClientBuilder::new(account, credentials)
+        // we have an outer retry
-            // we have an outer retry
+        let builder = ClientBuilder::new(account, credentials).retry(RetryOptions::none());
-            .retry(RetryOptions::none())
-            // Customize transport to configure conneciton pooling
-            .transport(TransportOptions::new(Self::reqwest_client(
-                azure_config.conn_pool_size,
-            )));
 
         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;
@@ -640,10 +624,6 @@ 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
-///
+pub const MAX_KEYS_PER_DELETE: usize = 1000;
-/// <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;
 
 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
+const_format.workspace = true
-pq_proto.workspace = true
-tokio.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 data = tokio::task::spawn_blocking(move || prof_ctl.dump_pprof())
-                let bytes = prof_ctl.dump_pprof()?;
+                .await
-                // Symbolize the profile.
+                .map_err(|join_err| ApiError::InternalServerError(join_err.into()))?
-                // TODO: consider moving this upstream to jemalloc_pprof and avoiding the
+                .map_err(ApiError::InternalServerError)?;
-                // 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)?;
             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;
+  int64 high = 1;
-  uint64 low = 2;
+  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,
+            high: (value.raw() >> 64) as i64,
-            low: value.raw() as u64,
+            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 =
+            let more_events_at_this_key = events_iter
-                events_iter.peek().is_some_and(|next_e| next_e.key == e.key);
+                .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 {
+                    while !list.is_empty() || self.accumulator.len() == MAX_KEYS_PER_DELETE {
-                        if 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
+        Update tenant's config.
-
-        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.
 
         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)))?;
-
+    tenant.set_new_tenant_config(new_tenant_conf);
-    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)))?;
 
     json_response(StatusCode::OK, ())
 }
@@ -2077,30 +2036,15 @@ async fn timeline_compact_handler(
         parse_query_param::<_, bool>(&request, "wait_until_scheduled_compaction_done")?
             .unwrap_or(false);
 
-    let sub_compaction = compact_request
-        .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()),
+            .and_then(|r| r.compact_range.clone()),
-        compact_lsn_range: compact_request
+        compact_below_lsn: compact_request.as_ref().and_then(|r| r.compact_below_lsn),
-            .as_ref()
-            .and_then(|r| r.compact_lsn_range.clone()),
         flags,
-        sub_compaction,
-        sub_compaction_max_job_size_mb,
     };
 
-    let scheduled = compact_request
+    let scheduled = compact_request.map(|r| r.scheduled).unwrap_or(false);
-        .as_ref()
-        .map(|r| r.scheduled)
-        .unwrap_or(false);
 
     async {
         let ctx = RequestContext::new(TaskKind::MgmtRequest, DownloadBehavior::Download);
@@ -2109,7 +2053,7 @@ async fn timeline_compact_handler(
             let tenant = state
                 .tenant_manager
                 .get_attached_tenant_shard(tenant_shard_id)?;
-            let rx = tenant.schedule_compaction(timeline_id, options).await.map_err(ApiError::InternalServerError)?;
+            let rx = tenant.schedule_compaction(timeline_id, options).await;
             if wait_until_scheduled_compaction_done {
                 // It is possible that this will take a long time, dropping the HTTP request will not cancel the compaction.
                 rx.await.ok();
@@ -3336,9 +3280,6 @@ 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",
@@ -1213,163 +1223,54 @@ pub(crate) mod virtual_file_io_engine {
     });
 }
 
-pub(crate) struct SmgrOpTimer(Option<SmgrOpTimerInner>);
+pub(crate) struct SmgrOpTimer {
-pub(crate) struct SmgrOpTimerInner {
+    global_latency_histo: Histogram,
-    global_execution_latency_histo: Histogram,
-    per_timeline_execution_latency_histo: Option<Histogram>,
 
-    global_batch_wait_time: Histogram,
+    // Optional because not all op types are tracked per-timeline
-    per_timeline_batch_wait_time: Histogram,
+    per_timeline_latency_histo: Option<Histogram>,
 
-    global_flush_in_progress_micros: IntCounter,
+    start: Instant,
-    per_timeline_flush_in_progress_micros: IntCounter,
+    throttled: Duration,
-
+    op: SmgrQueryType,
-    timings: SmgrOpTimerState,
-}
-
-#[derive(Debug)]
-enum SmgrOpTimerState {
-    Received {
-        received_at: Instant,
-    },
-    ThrottleDoneExecutionStarting {
-        received_at: Instant,
-        throttle_started_at: Instant,
-        started_execution_at: Instant,
-    },
-}
-
-pub(crate) struct SmgrOpFlushInProgress {
-    flush_started_at: 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 inner = self.0.as_mut().expect("other public methods consume self");
+        let Some(throttle) = throttle else {
-        match (&mut inner.timings, throttle) {
+            return;
-            (SmgrOpTimerState::Received { received_at }, throttle) => match throttle {
+        };
-                ThrottleResult::NotThrottled { start } => {
+        self.throttled += *throttle;
-                    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:?}"),
-        }
-    }
-
-    pub(crate) fn observe_smgr_op_completion_and_start_flushing(mut self) -> SmgrOpFlushInProgress {
-        let (flush_start, inner) = self
-            .smgr_op_end()
-            .expect("this method consume self, and the only other caller is drop handler");
-        let SmgrOpTimerInner {
-            global_flush_in_progress_micros,
-            per_timeline_flush_in_progress_micros,
-            ..
-        } = inner;
-        SmgrOpFlushInProgress {
-            flush_started_at: flush_start,
-            global_micros: global_flush_in_progress_micros,
-            per_timeline_micros: per_timeline_flush_in_progress_micros,
-        }
-    }
-
-    /// Returns `None`` if this method has already been called, `Some` otherwise.
-    fn smgr_op_end(&mut self) -> Option<(Instant, SmgrOpTimerInner)> {
-        let inner = self.0.take()?;
-
-        let now = Instant::now();
-
-        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();
-            }
-            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);
-
-        // 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);
-        }
-
-        Some((now, inner))
     }
 }
 
 impl Drop for SmgrOpTimer {
     fn drop(&mut self) {
-        self.smgr_op_end();
+        let elapsed = self.start.elapsed();
-    }
-}
 
-impl SmgrOpFlushInProgress {
+        let elapsed = match elapsed.checked_sub(self.throttled) {
-    pub(crate) async fn measure<Fut, O>(mut self, mut fut: Fut) -> O
+            Some(elapsed) => elapsed,
-    where
+            None => {
-        Fut: std::future::Future<Output = O>,
+                use utils::rate_limit::RateLimit;
-    {
+                static LOGGED: Lazy<Mutex<enum_map::EnumMap<SmgrQueryType, RateLimit>>> =
-        let mut fut = std::pin::pin!(fut);
+                    Lazy::new(|| {
-
+                        Mutex::new(enum_map::EnumMap::from_array(std::array::from_fn(|_| {
-        let now = Instant::now();
+                            RateLimit::new(Duration::from_secs(10))
-        // Whenever observe_guard gets called, or dropped,
+                        })))
-        // it adds the time elapsed since its last call to metrics.
+                    });
-        // Last call is tracked in `now`.
+                let mut guard = LOGGED.lock().unwrap();
-        let mut observe_guard = scopeguard::guard(
+                let rate_limit = &mut guard[self.op];
-            || {
+                rate_limit.call(|| {
-                let elapsed = now - self.flush_started_at;
+                    warn!(op=?self.op, ?elapsed, ?self.throttled, "implementation error: time spent throttled exceeds total request wall clock time");
-                self.global_micros
+                });
-                    .inc_by(u64::try_from(elapsed.as_micros()).unwrap());
+                elapsed // un-throttled time, more info than just saturating to 0
-                self.per_timeline_micros
-                    .inc_by(u64::try_from(elapsed.as_micros()).unwrap());
-                self.flush_started_at = now;
-            },
-            |mut observe| {
-                observe();
-            },
-        );
-
-        loop {
-            match tokio::time::timeout(Duration::from_secs(10), &mut fut).await {
-                Ok(v) => return v,
-                Err(_timeout) => {
-                    (*observe_guard)();
-                }
             }
+        };
+
+        let elapsed = elapsed.as_secs_f64();
+
+        self.global_latency_histo.observe(elapsed);
+        if let Some(per_timeline_getpage_histo) = &self.per_timeline_latency_histo {
+            per_timeline_getpage_histo.observe(elapsed);
         }
     }
 }
@@ -1401,10 +1302,6 @@ pub(crate) struct SmgrQueryTimePerTimeline {
     per_timeline_getpage_latency: Histogram,
     global_batch_size: Histogram,
     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 +1324,12 @@ 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 +1387,7 @@ 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(),
     )
@@ -1570,45 +1464,6 @@ fn set_page_service_config_max_batch_size(conf: &PageServicePipeliningConfig) {
         .set(value.try_into().unwrap());
 }
 
-static PAGE_SERVICE_SMGR_FLUSH_INPROGRESS_MICROS: Lazy<IntCounterVec> = Lazy::new(|| {
-    register_int_counter_vec!(
-        "pageserver_page_service_pagestream_flush_in_progress_micros",
-        "Counter that sums up the microseconds that a pagestream response was being flushed into the TCP connection. \
-         If the flush is particularly slow, this counter will be updated periodically to make slow flushes \
-         easily discoverable in monitoring. \
-         Hence, this is NOT a completion latency historgram.",
-        &["tenant_id", "shard_id", "timeline_id"],
-    )
-    .expect("failed to define a metric")
-});
-
-static PAGE_SERVICE_SMGR_FLUSH_INPROGRESS_MICROS_GLOBAL: Lazy<IntCounter> = Lazy::new(|| {
-    register_int_counter!(
-        "pageserver_page_service_pagestream_flush_in_progress_micros_global",
-        "Like pageserver_page_service_pagestream_flush_in_progress_seconds, but instance-wide.",
-    )
-    .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();
@@ -1649,17 +1504,6 @@ 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
-            .get_metric_with_label_values(&[&tenant_id, &shard_slug, &timeline_id])
-            .unwrap();
-
         Self {
             global_started,
             global_latency,
@@ -1667,13 +1511,9 @@ impl SmgrQueryTimePerTimeline {
             per_timeline_getpage_started,
             global_batch_size,
             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) {
@@ -1683,17 +1523,13 @@ impl SmgrQueryTimePerTimeline {
             None
         };
 
-        SmgrOpTimer(Some(SmgrOpTimerInner {
+        SmgrOpTimer {
-            global_execution_latency_histo: self.global_latency[op as usize].clone(),
+            global_latency_histo: self.global_latency[op as usize].clone(),
-            per_timeline_execution_latency_histo: per_timeline_latency_histo,
+            per_timeline_latency_histo,
-            timings: SmgrOpTimerState::Received { received_at },
+            start: started_at,
-            global_flush_in_progress_micros: self.global_flush_in_progress_micros.clone(),
+            op,
-            per_timeline_flush_in_progress_micros: self
+            throttled: Duration::ZERO,
-                .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(),
-        }))
     }
 
     pub(crate) fn observe_getpage_batch_start(&self, batch_size: usize) {
@@ -2368,15 +2204,6 @@ pub(crate) static WAL_INGEST: Lazy<WalIngestMetrics> = Lazy::new(|| WalIngestMet
     .expect("failed to define a metric"),
 });
 
-pub(crate) static PAGESERVER_TIMELINE_WAL_RECORDS_RECEIVED: Lazy<IntCounterVec> = Lazy::new(|| {
-    register_int_counter_vec!(
-        "pageserver_timeline_wal_records_received",
-        "Number of WAL records received per shard",
-        &["tenant_id", "shard_id", "timeline_id"]
-    )
-    .expect("failed to define a metric")
-});
-
 pub(crate) static WAL_REDO_TIME: Lazy<Histogram> = Lazy::new(|| {
     register_histogram!(
         "pageserver_wal_redo_seconds",
@@ -2577,6 +2404,7 @@ 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,
@@ -2603,7 +2431,6 @@ pub(crate) struct TimelineMetrics {
     pub evictions_with_low_residence_duration: std::sync::RwLock<EvictionsWithLowResidenceDuration>,
     /// Number of valid LSN leases.
     pub valid_lsn_lease_count_gauge: UIntGauge,
-    pub wal_records_received: IntCounter,
     shutdown: std::sync::atomic::AtomicBool,
 }
 
@@ -2622,6 +2449,9 @@ 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,
@@ -2758,15 +2588,12 @@ impl TimelineMetrics {
             .get_metric_with_label_values(&[&tenant_id, &shard_id, &timeline_id])
             .unwrap();
 
-        let wal_records_received = PAGESERVER_TIMELINE_WAL_RECORDS_RECEIVED
-            .get_metric_with_label_values(&[&tenant_id, &shard_id, &timeline_id])
-            .unwrap();
-
         TimelineMetrics {
             tenant_id,
             shard_id,
             timeline_id,
             flush_time_histo,
+            flush_wait_upload_time_gauge,
             compact_time_histo,
             create_images_time_histo,
             logical_size_histo,
@@ -2793,7 +2620,6 @@ impl TimelineMetrics {
                 evictions_with_low_residence_duration,
             ),
             valid_lsn_lease_count_gauge,
-            wal_records_received,
             shutdown: std::sync::atomic::AtomicBool::default(),
         }
     }
@@ -2816,6 +2642,14 @@ 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 +2667,7 @@ 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());
@@ -2922,21 +2757,6 @@ impl TimelineMetrics {
             shard_id,
             timeline_id,
         ]);
-        let _ = PAGESERVER_TIMELINE_WAL_RECORDS_RECEIVED.remove_label_values(&[
-            tenant_id,
-            shard_id,
-            timeline_id,
-        ]);
-        let _ = PAGE_SERVICE_SMGR_FLUSH_INPROGRESS_MICROS.remove_label_values(&[
-            tenant_id,
-            shard_id,
-            timeline_id,
-        ]);
-        let _ = PAGE_SERVICE_SMGR_BATCH_WAIT_TIME.remove_label_values(&[
-            tenant_id,
-            shard_id,
-            timeline_id,
-        ]);
     }
 }
 
@@ -2967,7 +2787,6 @@ 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 +3641,6 @@ 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 +3688,6 @@ 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(
+    async fn throttle(&mut self, cancel: &CancellationToken) -> Result<(), QueryError> {
-        &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(())
     }
@@ -1020,8 +1017,10 @@ impl PageServerHandler {
         // Map handler result to protocol behavior.
         // Some handler errors cause exit from pagestream protocol.
         // Other handler errors are sent back as an error message and we stay in pagestream protocol.
+        let mut timers: smallvec::SmallVec<[_; 1]> =
+            smallvec::SmallVec::with_capacity(handler_results.len());
         for handler_result in handler_results {
-            let (response_msg, timer) = match handler_result {
+            let response_msg = match handler_result {
                 Err(e) => match &e {
                     PageStreamError::Shutdown => {
                         // If we fail to fulfil a request during shutdown, which may be _because_ of
@@ -1045,66 +1044,34 @@ impl PageServerHandler {
                         span.in_scope(|| {
                             error!("error reading relation or page version: {full:#}")
                         });
-                        (
+                        PagestreamBeMessage::Error(PagestreamErrorResponse {
-                            PagestreamBeMessage::Error(PagestreamErrorResponse {
+                            message: e.to_string(),
-                                message: e.to_string(),
+                        })
-                            }),
-                            None, // TODO: measure errors
-                        )
                     }
                 },
-                Ok((response_msg, timer)) => (response_msg, Some(timer)),
+                Ok((response_msg, timer)) => {
+                    // Extending the lifetime of the timers so observations on drop
+                    // include the flush time.
+                    timers.push(timer);
+                    response_msg
+                }
             };
 
-            //
             // marshal & transmit response message
-            //
-
             pgb_writer.write_message_noflush(&BeMessage::CopyData(&response_msg.serialize()))?;
-
-            // We purposefully don't count flush time into the timer.
-            //
-            // The reason is that current compute client will not perform protocol processing
-            // if the postgres backend process is doing things other than `->smgr_read()`.
-            // This is especially the case for prefetch.
-            //
-            // If the compute doesn't read from the connection, eventually TCP will backpressure
-            // all the way into our flush call below.
-            //
-            // The timer's underlying metric is used for a storage-internal latency SLO and
-            // we don't want to include latency in it that we can't control.
-            // And as pointed out above, in this case, we don't control the time that flush will take.
-            let flushing_timer =
-                timer.map(|timer| timer.observe_smgr_op_completion_and_start_flushing());
-
-            // what we want to do
-            let flush_fut = pgb_writer.flush();
-            // metric for how long flushing takes
-            let flush_fut = match flushing_timer {
-                Some(flushing_timer) => {
-                    futures::future::Either::Left(flushing_timer.measure(flush_fut))
-                }
-                None => futures::future::Either::Right(flush_fut),
-            };
-            // do it while respecting cancellation
-            let _: () = async move {
-                tokio::select! {
-                    biased;
-                    _ = cancel.cancelled() => {
-                        // We were requested to shut down.
-                        info!("shutdown request received in page handler");
-                        return Err(QueryError::Shutdown)
-                    }
-                    res = flush_fut => {
-                        res?;
-                    }
-                }
-                Ok(())
-            }
-            // and log the info! line inside the request span
-            .instrument(span.clone())
-            .await?;
         }
+        tokio::select! {
+            biased;
+            _ = cancel.cancelled() => {
+                // We were requested to shut down.
+                info!("shutdown request received in page handler");
+                return Err(QueryError::Shutdown)
+            }
+            res = pgb_writer.flush() => {
+                res?;
+            }
+        }
+        drop(timers);
         Ok(())
     }
 
@@ -1233,7 +1200,7 @@ 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 +1367,7 @@ impl PageServerHandler {
                             return Err(e);
                         }
                     };
-                    batch
+                    batch.throttle(&self.cancel).await?;
-                        .throttle_and_record_start_processing(&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);
-
+        self.put(slru_dir_to_key(SlruKind::Clog), empty_dir.clone());
-        // Initialize SLRUs on shard 0 only: creating these on other shards would be
+        self.pending_directory_entries
-        // harmless but they'd just be dropped on later compaction.
+            .push((DirectoryKind::SlruSegment(SlruKind::Clog), 0));
-        if self.tline.tenant_shard_id.is_shard_zero() {
+        self.put(
-            self.put(slru_dir_to_key(SlruKind::Clog), empty_dir.clone());
+            slru_dir_to_key(SlruKind::MultiXactMembers),
-            self.pending_directory_entries
+            empty_dir.clone(),
-                .push((DirectoryKind::SlruSegment(SlruKind::Clog), 0));
+        );
-            self.put(
+        self.pending_directory_entries
-                slru_dir_to_key(SlruKind::MultiXactMembers),
+            .push((DirectoryKind::SlruSegment(SlruKind::Clog), 0));
-                empty_dir.clone(),
+        self.put(slru_dir_to_key(SlruKind::MultiXactOffsets), empty_dir);
-            );
+        self.pending_directory_entries
-            self.pending_directory_entries
+            .push((DirectoryKind::SlruSegment(SlruKind::MultiXactOffsets), 0));
-                .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,