experimental support for pg_anon

Signed-off-by: Luís Tavares <luis@neon.tech>

.github/PULL_REQUEST_TEMPLATE/release-pr.md

@@ -1,21 +0,0 @@
-## Release 202Y-MM-DD
-
-**NB: this PR must be merged only by 'Create a merge commit'!**
-
-### Checklist when preparing for release
-- [ ] Read or refresh [the release flow guide](https://www.notion.so/neondatabase/Release-general-flow-61f2e39fd45d4d14a70c7749604bd70b)
-- [ ] Ask in the [cloud Slack channel](https://neondb.slack.com/archives/C033A2WE6BZ) that you are going to rollout the release. Any blockers?
-- [ ] Does this release contain any db migrations? Destructive ones? What is the rollback plan?
-
-<!-- List everything that should be done **before** release, any issues / setting changes / etc -->
-
-### Checklist after release
-- [ ] Make sure instructions from PRs included in this release and labeled `manual_release_instructions` are executed (either by you or by people who wrote them).
-- [ ] Based on the merged commits write release notes and open a PR into `website` repo ([example](https://github.com/neondatabase/website/pull/219/files))
-- [ ] Check [#dev-production-stream](https://neondb.slack.com/archives/C03F5SM1N02) Slack channel
-- [ ] Check [stuck projects page](https://console.neon.tech/admin/projects?sort=last_active&order=desc&stuck=true)
-- [ ] Check [recent operation failures](https://console.neon.tech/admin/operations?action=create_timeline%2Cstart_compute%2Cstop_compute%2Csuspend_compute%2Capply_config%2Cdelete_timeline%2Cdelete_tenant%2Ccreate_branch%2Ccheck_availability&sort=updated_at&order=desc&had_retries=some)
-- [ ] Check [cloud SLO dashboard](https://neonprod.grafana.net/d/_oWcBMJ7k/cloud-slos?orgId=1)
-- [ ] Check [compute startup metrics dashboard](https://neonprod.grafana.net/d/5OkYJEmVz/compute-startup-time)
-
-<!-- List everything that should be done **after** release, any admin UI configuration / Grafana dashboard / alert changes / setting changes / etc -->

.github/scripts/generate_image_maps.py

@@ -1,14 +1,16 @@
 import itertools
 import json
 import os
+import sys
 
-build_tag = os.environ["BUILD_TAG"]
-branch = os.environ["BRANCH"]
-dev_acr = os.environ["DEV_ACR"]
-prod_acr = os.environ["PROD_ACR"]
-dev_aws = os.environ["DEV_AWS"]
-prod_aws = os.environ["PROD_AWS"]
-aws_region = os.environ["AWS_REGION"]
+source_tag = os.getenv("SOURCE_TAG")
+target_tag = os.getenv("TARGET_TAG")
+branch = os.getenv("BRANCH")
+dev_acr = os.getenv("DEV_ACR")
+prod_acr = os.getenv("PROD_ACR")
+dev_aws = os.getenv("DEV_AWS")
+prod_aws = os.getenv("PROD_AWS")
+aws_region = os.getenv("AWS_REGION")
 
 components = {
     "neon": ["neon"],
@@ -39,24 +41,23 @@ registries = {
 
 outputs: dict[str, dict[str, list[str]]] = {}
 
-target_tags = [build_tag, "latest"] if branch == "main" else [build_tag]
-target_stages = ["dev", "prod"] if branch.startswith("release") else ["dev"]
+target_tags = [target_tag, "latest"] if branch == "main" else [target_tag]
+target_stages = (
+    ["dev", "prod"] if branch in ["release", "release-proxy", "release-compute"] else ["dev"]
+)
 
 for component_name, component_images in components.items():
     for stage in target_stages:
-        outputs[f"{component_name}-{stage}"] = dict(
-            [
-                (
-                    f"docker.io/neondatabase/{component_image}:{build_tag}",
-                    [
-                        f"{combo[0]}/{component_image}:{combo[1]}"
-                        for combo in itertools.product(registries[stage], target_tags)
-                    ],
-                )
-                for component_image in component_images
+        outputs[f"{component_name}-{stage}"] = {
+            f"docker.io/neondatabase/{component_image}:{source_tag}": [
+                f"{registry}/{component_image}:{tag}"
+                for registry, tag in itertools.product(registries[stage], target_tags)
+                if not (registry == "docker.io/neondatabase" and tag == source_tag)
             ]
-        )
+            for component_image in component_images
+        }
 
-with open(os.environ["GITHUB_OUTPUT"], "a") as f:
+with open(os.getenv("GITHUB_OUTPUT", "/dev/null"), "a") as f:
     for key, value in outputs.items():
         f.write(f"{key}={json.dumps(value)}\n")
+        print(f"Image map for {key}:\n{json.dumps(value, indent=2)}\n\n", file=sys.stderr)

.github/scripts/lint-release-pr.sh

@@ -0,0 +1,110 @@
+#!/usr/bin/env bash
+
+set -euo pipefail
+
+DOCS_URL="https://docs.neon.build/overview/repositories/neon.html"
+
+message() {
+  if [[ -n "${GITHUB_PR_NUMBER:-}" ]]; then
+    gh pr comment --repo "${GITHUB_REPOSITORY}" "${GITHUB_PR_NUMBER}" --edit-last --body "$1" \
+      || gh pr comment --repo "${GITHUB_REPOSITORY}" "${GITHUB_PR_NUMBER}" --body "$1"
+  fi
+  echo "$1"
+}
+
+report_error() {
+  message "❌ $1
+  For more details, see the documentation: ${DOCS_URL}"
+
+  exit 1
+}
+
+case "$RELEASE_BRANCH" in
+  "release") COMPONENT="Storage" ;;
+  "release-proxy") COMPONENT="Proxy" ;;
+  "release-compute") COMPONENT="Compute" ;;
+  *)
+    report_error "Unknown release branch: ${RELEASE_BRANCH}"
+    ;;
+esac
+
+
+# Identify main and release branches
+MAIN_BRANCH="origin/main"
+REMOTE_RELEASE_BRANCH="origin/${RELEASE_BRANCH}"
+
+# Find merge base
+MERGE_BASE=$(git merge-base "${MAIN_BRANCH}" "${REMOTE_RELEASE_BRANCH}")
+echo "Merge base of ${MAIN_BRANCH} and ${RELEASE_BRANCH}: ${MERGE_BASE}"
+
+# Get the HEAD commit (last commit in PR, expected to be the merge commit)
+LAST_COMMIT=$(git rev-parse HEAD)
+
+MERGE_COMMIT_MESSAGE=$(git log -1 --format=%s "${LAST_COMMIT}")
+EXPECTED_MESSAGE_REGEX="^$COMPONENT release [0-9]{4}-[0-9]{2}-[0-9]{2}$"
+
+if ! [[ "${MERGE_COMMIT_MESSAGE}" =~ ${EXPECTED_MESSAGE_REGEX} ]]; then
+  report_error "Merge commit message does not match expected pattern: '<component> release YYYY-MM-DD'
+  Expected component: ${COMPONENT}
+  Found: '${MERGE_COMMIT_MESSAGE}'"
+fi
+echo "✅ Merge commit message is correctly formatted: '${MERGE_COMMIT_MESSAGE}'"
+
+LAST_COMMIT_PARENTS=$(git cat-file -p "${LAST_COMMIT}" | jq -sR '[capture("parent (?<parent>[0-9a-f]{40})"; "g") | .parent]')
+
+if [[ "$(echo "${LAST_COMMIT_PARENTS}" | jq 'length')" -ne 2 ]]; then
+  report_error "Last commit must be a merge commit with exactly two parents"
+fi
+
+EXPECTED_RELEASE_HEAD=$(git rev-parse "${REMOTE_RELEASE_BRANCH}")
+if echo "${LAST_COMMIT_PARENTS}" | jq -e --arg rel "${EXPECTED_RELEASE_HEAD}" 'index($rel) != null' > /dev/null; then
+  LINEAR_HEAD=$(echo "${LAST_COMMIT_PARENTS}" | jq -r '[.[] | select(. != $rel)][0]' --arg rel "${EXPECTED_RELEASE_HEAD}")
+else
+  report_error "Last commit must merge the release branch (${RELEASE_BRANCH})"
+fi
+echo "✅ Last commit correctly merges the previous commit and the release branch"
+echo "Top commit of linear history: ${LINEAR_HEAD}"
+
+MERGE_COMMIT_TREE=$(git rev-parse "${LAST_COMMIT}^{tree}")
+LINEAR_HEAD_TREE=$(git rev-parse "${LINEAR_HEAD}^{tree}")
+
+if [[ "${MERGE_COMMIT_TREE}" != "${LINEAR_HEAD_TREE}" ]]; then
+  report_error "Tree of merge commit (${MERGE_COMMIT_TREE}) does not match tree of linear history head (${LINEAR_HEAD_TREE})
+  This indicates that the merge of ${RELEASE_BRANCH} into this branch was not performed using the merge strategy 'ours'"
+fi
+echo "✅ Merge commit tree matches the linear history head"
+
+EXPECTED_PREVIOUS_COMMIT="${LINEAR_HEAD}"
+
+# Now traverse down the history, ensuring each commit has exactly one parent
+CURRENT_COMMIT="${EXPECTED_PREVIOUS_COMMIT}"
+while [[ "${CURRENT_COMMIT}" != "${MERGE_BASE}" && "${CURRENT_COMMIT}" != "${EXPECTED_RELEASE_HEAD}" ]]; do
+  CURRENT_COMMIT_PARENTS=$(git cat-file -p "${CURRENT_COMMIT}" | jq -sR '[capture("parent (?<parent>[0-9a-f]{40})"; "g") | .parent]')
+
+  if [[ "$(echo "${CURRENT_COMMIT_PARENTS}" | jq 'length')" -ne 1 ]]; then
+    report_error "Commit ${CURRENT_COMMIT} must have exactly one parent"
+  fi
+
+  NEXT_COMMIT=$(echo "${CURRENT_COMMIT_PARENTS}" | jq -r '.[0]')
+
+  if [[ "${NEXT_COMMIT}" == "${MERGE_BASE}" ]]; then
+    echo "✅ Reached merge base (${MERGE_BASE})"
+    PR_BASE="${MERGE_BASE}"
+  elif [[ "${NEXT_COMMIT}" == "${EXPECTED_RELEASE_HEAD}" ]]; then
+    echo "✅ Reached release branch (${EXPECTED_RELEASE_HEAD})"
+    PR_BASE="${EXPECTED_RELEASE_HEAD}"
+  elif [[ -z "${NEXT_COMMIT}" ]]; then
+    report_error "Unexpected end of commit history before reaching merge base"
+  fi
+
+  # Move to the next commit in the chain
+  CURRENT_COMMIT="${NEXT_COMMIT}"
+done
+
+echo "✅ All commits are properly ordered and linear"
+echo "✅ Release PR structure is valid"
+
+echo
+
+message "Commits that are part of this release:
+$(git log --oneline "${PR_BASE}..${LINEAR_HEAD}")"

.github/scripts/previous-releases.jq

@@ -17,6 +17,12 @@
 ({};
  .[$entry.component] |= (if . == null or $entry.version > .version then $entry else . end))
 
+# Ensure that each component exists, or fail
+| (["storage", "compute", "proxy"] - (keys)) as $missing
+| if ($missing | length) > 0 then
+    "Error: Found no release for \($missing | join(", "))!\n" | halt_error(1)
+  else . end
+
 # Convert the resulting object into an array of formatted strings
 | to_entries
 | map("\(.key)=\(.value.full)")

.github/workflows/_create-release-pr.yml

@@ -7,8 +7,8 @@ on:
         description: 'Component name'
         required: true
         type: string
-      release-branch:
-        description: 'Release branch'
+      source-branch:
+        description: 'Source branch'
         required: true
         type: string
     secrets:
@@ -30,17 +30,25 @@ jobs:
     steps:
     - uses: actions/checkout@v4
       with:
-        ref: main
+        ref: ${{ inputs.source-branch }}
+        fetch-depth: 0
 
     - name: Set variables
       id: vars
       env:
         COMPONENT_NAME: ${{ inputs.component-name }}
-        RELEASE_BRANCH: ${{ inputs.release-branch }}
+        RELEASE_BRANCH: >-
+          ${{
+            false
+            || inputs.component-name == 'Storage' && 'release'
+            || inputs.component-name == 'Proxy' && 'release-proxy'
+            || inputs.component-name == 'Compute' && 'release-compute'
+          }}
       run: |
         today=$(date +'%Y-%m-%d')
         echo "title=${COMPONENT_NAME} release ${today}" | tee -a ${GITHUB_OUTPUT}
         echo "rc-branch=rc/${RELEASE_BRANCH}/${today}"  | tee -a ${GITHUB_OUTPUT}
+        echo "release-branch=${RELEASE_BRANCH}"         | tee -a ${GITHUB_OUTPUT}
 
     - name: Configure git
       run: |
@@ -49,31 +57,36 @@ jobs:
 
     - name: Create RC branch
       env:
+        RELEASE_BRANCH: ${{ steps.vars.outputs.release-branch }}
         RC_BRANCH: ${{ steps.vars.outputs.rc-branch }}
         TITLE: ${{ steps.vars.outputs.title }}
       run: |
-        git checkout -b "${RC_BRANCH}"
+        git switch -c "${RC_BRANCH}"
 
-        # create an empty commit to distinguish workflow runs
-        # from other possible releases from the same commit
-        git commit --allow-empty -m "${TITLE}"
+        # Manually create a merge commit on the current branch, keeping the
+        # tree and setting the parents to the current HEAD and the HEAD of the
+        # release branch. This commit is what we'll fast-forward the release
+        # branch to when merging the release branch.
+        # For details on why, look at
+        # https://docs.neon.build/overview/repositories/neon.html#background-on-commit-history-of-release-prs
+        current_tree=$(git rev-parse 'HEAD^{tree}')
+        release_head=$(git rev-parse "origin/${RELEASE_BRANCH}")
+        current_head=$(git rev-parse HEAD)
+        merge_commit=$(git commit-tree -p "${current_head}" -p "${release_head}" -m "${TITLE}" "${current_tree}")
+
+        # Fast-forward the current branch to the newly created merge_commit
+        git merge --ff-only ${merge_commit}
 
         git push origin "${RC_BRANCH}"
 
-    - name: Create a PR into ${{ inputs.release-branch }}
+    - name: Create a PR into ${{ steps.vars.outputs.release-branch }}
       env:
         GH_TOKEN: ${{ secrets.ci-access-token }}
         RC_BRANCH: ${{ steps.vars.outputs.rc-branch }}
-        RELEASE_BRANCH: ${{ inputs.release-branch }}
+        RELEASE_BRANCH: ${{ steps.vars.outputs.release-branch }}
         TITLE: ${{ steps.vars.outputs.title }}
       run: |
-        cat << EOF > body.md
-          ## ${TITLE}
-
-          **Please merge this Pull Request using 'Create a merge commit' button**
-        EOF
-
         gh pr create --title "${TITLE}" \
-                     --body-file "body.md" \
+                     --body "" \
                      --head "${RC_BRANCH}" \
                      --base "${RELEASE_BRANCH}"

.github/workflows/_meta.yml

@@ -21,9 +21,16 @@ on:
       run-kind:
         description: "The kind of run we're currently in. Will be one of `push-main`, `storage-release`, `compute-release`, `proxy-release`, `storage-rc-pr`, `compute-rc-pr`,  `proxy-rc-pr`, `pr`, or `workflow-dispatch`"
         value: ${{ jobs.tags.outputs.run-kind }}
+      release-pr-run-id:
+        description: "Only available if `run-kind in [storage-release, proxy-release, compute-release]`. Contains the run ID of the `Build and Test` workflow, assuming one with the current commit can be found."
+        value: ${{ jobs.tags.outputs.release-pr-run-id }}
 
 permissions: {}
 
+defaults:
+  run:
+    shell: bash -euo pipefail {0}
+
 jobs:
   tags:
     runs-on: ubuntu-22.04
@@ -33,6 +40,7 @@ jobs:
       proxy: ${{ steps.previous-releases.outputs.proxy }}
       storage: ${{ steps.previous-releases.outputs.storage }}
       run-kind: ${{ steps.run-kind.outputs.run-kind }}
+      release-pr-run-id: ${{ steps.release-pr-run-id.outputs.release-pr-run-id }}
     permissions:
       contents: read
     steps:
@@ -83,7 +91,11 @@ jobs:
             echo "tag=release-compute-$(git rev-list --count HEAD)" | tee -a $GITHUB_OUTPUT
             ;;
           pr|storage-rc-pr|compute-rc-pr|proxy-rc-pr)
-            BUILD_AND_TEST_RUN_ID=$(gh run list -b $CURRENT_BRANCH -c $CURRENT_SHA -w 'Build and Test' -L 1 --json databaseId --jq '.[].databaseId')
+            BUILD_AND_TEST_RUN_ID=$(gh api --paginate \
+              -H "Accept: application/vnd.github+json" \
+              -H "X-GitHub-Api-Version: 2022-11-28" \
+              "/repos/${GITHUB_REPOSITORY}/actions/runs?head_sha=${CURRENT_SHA}&branch=${CURRENT_BRANCH}" \
+              | jq '[.workflow_runs[] | select(.name == "Build and Test")][0].id // ("Error: No matching workflow run found." | halt_error(1))')
             echo "tag=$BUILD_AND_TEST_RUN_ID" | tee -a $GITHUB_OUTPUT
             ;;
           workflow-dispatch)
@@ -105,3 +117,13 @@ jobs:
             "/repos/${GITHUB_REPOSITORY}/releases" \
           | jq -f .github/scripts/previous-releases.jq -r \
           | tee -a "${GITHUB_OUTPUT}"
+
+      - name: Get the release PR run ID
+        id: release-pr-run-id
+        if: ${{ contains(fromJson('["storage-release", "compute-release", "proxy-release"]'), steps.run-kind.outputs.run-kind) }}
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          CURRENT_SHA: ${{ github.event.pull_request.head.sha || github.sha }}
+        run: |
+          RELEASE_PR_RUN_ID=$(gh api "/repos/${GITHUB_REPOSITORY}/actions/runs?head_sha=$CURRENT_SHA" | jq '[.workflow_runs[] | select(.name == "Build and Test") | select(.head_branch | test("^rc/release(-(proxy)|(compute))?/[0-9]{4}-[0-9]{2}-[0-9]{2}$"; "s"))] | first | .id // ("Faied to find Build and Test run from  RC PR!" | halt_error(1))')
+          echo "release-pr-run-id=$RELEASE_PR_RUN_ID" | tee -a $GITHUB_OUTPUT

.github/workflows/build_and_test.yml

@@ -476,7 +476,7 @@ jobs:
         (
           !github.event.pull_request.draft
           || contains( github.event.pull_request.labels.*.name, 'run-e2e-tests-in-draft')
-          || contains(fromJSON('["push-main", "storage-release", "proxy-release", "compute-release"]'), needs.meta.outputs.run-kind)
+          || needs.meta.outputs.run-kind == 'push-main'
         ) && !failure() && !cancelled()
       }}
     needs: [ check-permissions, push-neon-image-dev, push-compute-image-dev, meta ]
@@ -487,7 +487,7 @@ jobs:
 
   neon-image-arch:
     needs: [ check-permissions, build-build-tools-image, meta ]
-    if: ${{ contains(fromJSON('["push-main", "pr", "storage-release", "storage-rc-pr", "proxy-release", "proxy-rc-pr"]'), needs.meta.outputs.run-kind) }}
+    if: ${{ contains(fromJSON('["push-main", "pr", "storage-rc-pr", "proxy-rc-pr"]'), needs.meta.outputs.run-kind) }}
     strategy:
       matrix:
         arch: [ x64, arm64 ]
@@ -537,7 +537,7 @@ jobs:
 
   neon-image:
     needs: [ neon-image-arch, meta ]
-    if: ${{ contains(fromJSON('["push-main", "pr", "storage-release", "storage-rc-pr", "proxy-release", "proxy-rc-pr"]'), needs.meta.outputs.run-kind) }}
+    if: ${{ contains(fromJSON('["push-main", "pr", "storage-rc-pr", "proxy-rc-pr"]'), needs.meta.outputs.run-kind) }}
     runs-on: ubuntu-22.04
     permissions:
       id-token: write # aws-actions/configure-aws-credentials
@@ -559,7 +559,7 @@ jobs:
 
   compute-node-image-arch:
     needs: [ check-permissions, build-build-tools-image, meta ]
-    if: ${{ contains(fromJSON('["push-main", "pr", "compute-release", "compute-rc-pr"]'), needs.meta.outputs.run-kind) }}
+    if: ${{ contains(fromJSON('["push-main", "pr", "compute-rc-pr"]'), needs.meta.outputs.run-kind) }}
     permissions:
       id-token: write # aws-actions/configure-aws-credentials
       statuses: write
@@ -651,7 +651,7 @@ jobs:
 
   compute-node-image:
     needs: [ compute-node-image-arch, meta ]
-    if: ${{ contains(fromJSON('["push-main", "pr", "compute-release", "compute-rc-pr"]'), needs.meta.outputs.run-kind) }}
+    if: ${{ contains(fromJSON('["push-main", "pr", "compute-rc-pr"]'), needs.meta.outputs.run-kind) }}
     permissions:
       id-token: write # aws-actions/configure-aws-credentials
       statuses: write
@@ -694,7 +694,7 @@ jobs:
 
   vm-compute-node-image-arch:
     needs: [ check-permissions, meta, compute-node-image ]
-    if: ${{ contains(fromJSON('["push-main", "pr", "compute-release", "compute-rc-pr"]'), needs.meta.outputs.run-kind) }}
+    if: ${{ contains(fromJSON('["push-main", "pr", "compute-rc-pr"]'), needs.meta.outputs.run-kind) }}
     runs-on: ${{ fromJson(format('["self-hosted", "{0}"]', matrix.arch == 'arm64' && 'large-arm64' || 'large')) }}
     strategy:
       fail-fast: false
@@ -747,7 +747,7 @@ jobs:
 
   vm-compute-node-image:
     needs: [ vm-compute-node-image-arch, meta ]
-    if: ${{ contains(fromJSON('["push-main", "pr", "compute-release", "compute-rc-pr"]'), needs.meta.outputs.run-kind) }}
+    if: ${{ contains(fromJSON('["push-main", "pr", "compute-rc-pr"]'), needs.meta.outputs.run-kind) }}
     runs-on: ubuntu-22.04
     strategy:
       matrix:
@@ -773,7 +773,12 @@ jobs:
   test-images:
     needs: [ check-permissions, meta, neon-image, compute-node-image ]
     # Depends on jobs that can get skipped
-    if: "!failure() && !cancelled()"
+    if: >-
+      ${{
+        !failure()
+        && !cancelled()
+        && contains(fromJSON('["push-main", "pr", "storage-rc-pr", "proxy-rc-pr", "compute-rc-pr"]'), needs.meta.outputs.run-kind)
+      }}
     strategy:
       fail-fast: false
       matrix:
@@ -800,7 +805,7 @@ jobs:
       # Ensure that we don't have bad versions.
       - name: Verify image versions
         shell: bash # ensure no set -e for better error messages
-        if: ${{ contains(fromJSON('["push-main", "pr", "storage-release", "storage-rc-pr", "proxy-release", "proxy-rc-pr"]'), needs.meta.outputs.run-kind) }}
+        if: ${{ contains(fromJSON('["push-main", "pr", "storage-rc-pr", "proxy-rc-pr"]'), needs.meta.outputs.run-kind) }}
         run: |
           pageserver_version=$(docker run --rm neondatabase/neon:${{ needs.meta.outputs.build-tag }} "/bin/sh" "-c" "/usr/local/bin/pageserver --version")
 
@@ -821,19 +826,19 @@ jobs:
         env:
           TAG: >-
             ${{
-              contains(fromJSON('["compute-release", "compute-rc-pr"]'), needs.meta.outputs.run-kind)
+              needs.meta.outputs.run-kind == 'compute-rc-pr'
               && needs.meta.outputs.previous-storage-release
               || needs.meta.outputs.build-tag
             }}
           COMPUTE_TAG: >-
             ${{
-              contains(fromJSON('["storage-release", "storage-rc-pr", "proxy-release", "proxy-rc-pr"]'), needs.meta.outputs.run-kind)
+              contains(fromJSON('["storage-rc-pr", "proxy-rc-pr"]'), needs.meta.outputs.run-kind)
               && needs.meta.outputs.previous-compute-release
               || needs.meta.outputs.build-tag
             }}
           TEST_EXTENSIONS_TAG: >-
             ${{
-              contains(fromJSON('["storage-release", "storage-rc-pr", "proxy-release", "proxy-rc-pr"]'), needs.meta.outputs.run-kind)
+              contains(fromJSON('["storage-rc-pr", "proxy-rc-pr"]'), needs.meta.outputs.run-kind)
               && 'latest'
               || needs.meta.outputs.build-tag
             }}
@@ -885,7 +890,13 @@ jobs:
         id: generate
         run: python3 .github/scripts/generate_image_maps.py
         env:
-          BUILD_TAG: "${{ needs.meta.outputs.build-tag }}"
+          SOURCE_TAG: >-
+            ${{
+              contains(fromJson('["storage-release", "compute-release", "proxy-release"]'), needs.meta.outputs.run-kind)
+              && needs.meta.outputs.release-pr-run-id
+              || needs.meta.outputs.build-tag
+            }}
+          TARGET_TAG: ${{ needs.meta.outputs.build-tag }}
           BRANCH: "${{ github.ref_name }}"
           DEV_ACR: "${{ vars.AZURE_DEV_REGISTRY_NAME }}"
           PROD_ACR: "${{ vars.AZURE_PROD_REGISTRY_NAME }}"
@@ -895,7 +906,7 @@ jobs:
 
   push-neon-image-dev:
     needs: [ meta, generate-image-maps, neon-image ]
-    if: ${{ contains(fromJSON('["push-main", "pr", "storage-release", "storage-rc-pr", "proxy-release", "proxy-rc-pr"]'), needs.meta.outputs.run-kind) }}
+    if: ${{ !failure() && !cancelled() && contains(fromJSON('["push-main", "pr", "storage-release", "storage-rc-pr", "proxy-release", "proxy-rc-pr"]'), needs.meta.outputs.run-kind) }}
     uses: ./.github/workflows/_push-to-container-registry.yml
     permissions:
       id-token: write  # Required for aws/azure login
@@ -913,7 +924,7 @@ jobs:
 
   push-compute-image-dev:
     needs: [ meta, generate-image-maps, vm-compute-node-image ]
-    if: ${{ contains(fromJSON('["push-main", "pr", "compute-release", "compute-rc-pr"]'), needs.meta.outputs.run-kind) }}
+    if: ${{ !failure() && !cancelled() && contains(fromJSON('["push-main", "pr", "compute-release", "compute-rc-pr"]'), needs.meta.outputs.run-kind) }}
     uses: ./.github/workflows/_push-to-container-registry.yml
     permissions:
       id-token: write  # Required for aws/azure login
@@ -1235,7 +1246,7 @@ jobs:
 
   # 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 ]
+    needs: [ meta, deploy ]
     permissions:
       id-token: write # aws-actions/configure-aws-credentials
       statuses: write
@@ -1245,37 +1256,6 @@ jobs:
 
     runs-on: ubuntu-22.04
     steps:
-      - name: Fetch GITHUB_RUN_ID and COMMIT_SHA for the last merged release PR
-        id: fetch-last-release-pr-info
-        env:
-          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-        run: |
-          branch_name_and_pr_number=$(gh pr list \
-            --repo "${GITHUB_REPOSITORY}" \
-            --base release \
-            --state merged \
-            --limit 10 \
-            --json mergeCommit,headRefName,number \
-            --jq ".[] | select(.mergeCommit.oid==\"${GITHUB_SHA}\") | { branch_name: .headRefName, pr_number: .number }")
-          branch_name=$(echo "${branch_name_and_pr_number}" | jq -r '.branch_name')
-          pr_number=$(echo "${branch_name_and_pr_number}" | jq -r '.pr_number')
-
-          run_id=$(gh run list \
-            --repo "${GITHUB_REPOSITORY}" \
-            --workflow build_and_test.yml \
-            --branch "${branch_name}" \
-            --json databaseId \
-            --limit 1 \
-            --jq '.[].databaseId')
-
-          last_commit_sha=$(gh pr view "${pr_number}" \
-            --repo "${GITHUB_REPOSITORY}" \
-            --json commits \
-            --jq '.commits[-1].oid')
-
-          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
@@ -1286,8 +1266,8 @@ jobs:
         env:
           BUCKET: neon-github-public-dev
           AWS_REGION: eu-central-1
-          COMMIT_SHA: ${{ steps.fetch-last-release-pr-info.outputs.commit-sha }}
-          RUN_ID: ${{ steps.fetch-last-release-pr-info.outputs.run-id }}
+          COMMIT_SHA: ${{ github.sha }}
+          RUN_ID: ${{ needs.meta.outputs.release-pr-run-id }}
         run: |
           old_prefix="artifacts/${COMMIT_SHA}/${RUN_ID}"
           new_prefix="artifacts/latest"
@@ -1376,5 +1356,5 @@ jobs:
           || needs.files-changed.result == 'skipped'
           || (needs.push-compute-image-dev.result == 'skipped' && contains(fromJSON('["push-main", "pr", "compute-release", "compute-rc-pr"]'), needs.meta.outputs.run-kind))
           || (needs.push-neon-image-dev.result == 'skipped' && contains(fromJSON('["push-main", "pr", "storage-release", "storage-rc-pr", "proxy-release", "proxy-rc-pr"]'), needs.meta.outputs.run-kind))
-          || needs.test-images.result == 'skipped'
+          || (needs.test-images.result == 'skipped' && contains(fromJSON('["push-main", "pr", "storage-rc-pr", "proxy-rc-pr", "compute-rc-pr"]'), needs.meta.outputs.run-kind))
           || (needs.trigger-custom-extensions-build-and-wait.result == 'skipped' && contains(fromJSON('["push-main", "pr", "compute-release", "compute-rc-pr"]'), needs.meta.outputs.run-kind))

.github/workflows/fast-forward.yml

@@ -0,0 +1,36 @@
+name: Fast forward merge
+on:
+  pull_request:
+    types: [labeled]
+    branches:
+      - release
+      - release-proxy
+      - release-compute
+
+jobs:
+  fast-forward:
+    if: ${{ github.event.label.name == 'fast-forward' }}
+    runs-on: ubuntu-22.04
+
+    steps:
+      - name: Remove fast-forward label to PR
+        env:
+          GH_TOKEN: ${{ secrets.CI_ACCESS_TOKEN }}
+        run: |
+          gh pr edit ${{ github.event.pull_request.number }} --repo "${GITHUB_REPOSITORY}" --remove-label "fast-forward"
+
+      - name: Fast forwarding
+        uses: sequoia-pgp/fast-forward@ea7628bedcb0b0b96e94383ada458d812fca4979
+        # See https://docs.github.com/en/graphql/reference/enums#mergestatestatus
+        if: ${{ github.event.pull_request.mergeable_state  == 'clean' }}
+        with:
+          merge: true
+          comment: on-error
+          github_token: ${{ secrets.CI_ACCESS_TOKEN }}
+
+      - name: Comment if mergeable_state is not clean
+        if: ${{ github.event.pull_request.mergeable_state  != 'clean' }}
+        run: |
+          gh pr comment ${{ github.event.pull_request.number }} \
+            --repo "${GITHUB_REPOSITORY}" \
+            --body "Not trying to forward pull-request, because \`mergeable_state\` is \`${{ github.event.pull_request.mergeable_state }}\`, not \`clean\`."

.github/workflows/lint-release-pr.yml

@@ -0,0 +1,24 @@
+name: Lint Release PR
+
+on:
+  pull_request:
+    branches:
+      - release
+      - release-proxy
+      - release-compute
+
+jobs:
+  lint-release-pr:
+    runs-on: ubuntu-22.04
+    steps:
+      - name: Checkout PR branch
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0  # Fetch full history for git operations
+          ref: ${{ github.event.pull_request.head.ref }}
+
+      - name: Run lint script
+        env:
+          RELEASE_BRANCH: ${{ github.base_ref }}
+        run: |
+          ./.github/scripts/lint-release-pr.sh

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

@@ -8,8 +8,6 @@ on:
       - .github/workflows/build-build-tools-image.yml
       - .github/workflows/pre-merge-checks.yml
   merge_group:
-    branches:
-      - main
 
 defaults:
   run:
@@ -19,11 +17,13 @@ defaults:
 permissions: {}
 
 jobs:
-  get-changed-files:
+  meta:
     runs-on: ubuntu-22.04
     outputs:
       python-changed: ${{ steps.python-src.outputs.any_changed }}
       rust-changed: ${{ steps.rust-src.outputs.any_changed }}
+      branch: ${{ steps.group-metadata.outputs.branch }}
+      pr-number: ${{ steps.group-metadata.outputs.pr-number }}
     steps:
       - uses: actions/checkout@v4
 
@@ -58,12 +58,20 @@ jobs:
           echo "${PYTHON_CHANGED_FILES}"
           echo "${RUST_CHANGED_FILES}"
 
+      - name: Merge group metadata
+        if: ${{ github.event_name == 'merge_group' }}
+        id: group-metadata
+        env:
+          MERGE_QUEUE_REF: ${{ github.event.merge_group.head_ref }}
+        run: |
+          echo $MERGE_QUEUE_REF | jq -Rr 'capture("refs/heads/gh-readonly-queue/(?<branch>.*)/pr-(?<pr_number>[0-9]+)-[0-9a-f]{40}") | ["branch=" + .branch, "pr-number=" + .pr_number] | .[]' | tee -a "${GITHUB_OUTPUT}"
+
   build-build-tools-image:
     if: |
       false
-      || needs.get-changed-files.outputs.python-changed == 'true'
-      || needs.get-changed-files.outputs.rust-changed == 'true'
-    needs: [ get-changed-files ]
+      || needs.meta.outputs.python-changed == 'true'
+      || needs.meta.outputs.rust-changed == 'true'
+    needs: [ meta ]
     uses: ./.github/workflows/build-build-tools-image.yml
     with:
       # Build only one combination to save time
@@ -72,8 +80,8 @@ jobs:
     secrets: inherit
 
   check-codestyle-python:
-    if: needs.get-changed-files.outputs.python-changed == 'true'
-    needs: [ get-changed-files, build-build-tools-image ]
+    if: needs.meta.outputs.python-changed == 'true'
+    needs: [ meta, build-build-tools-image ]
     uses: ./.github/workflows/_check-codestyle-python.yml
     with:
       # `-bookworm-x64` suffix should match the combination in `build-build-tools-image`
@@ -81,8 +89,8 @@ jobs:
     secrets: inherit
 
   check-codestyle-rust:
-    if: needs.get-changed-files.outputs.rust-changed == 'true'
-    needs: [ get-changed-files, build-build-tools-image ]
+    if: needs.meta.outputs.rust-changed == 'true'
+    needs: [ meta, build-build-tools-image ]
     uses: ./.github/workflows/_check-codestyle-rust.yml
     with:
       # `-bookworm-x64` suffix should match the combination in `build-build-tools-image`
@@ -101,7 +109,7 @@ jobs:
       statuses: write # for `github.repos.createCommitStatus(...)`
       contents: write
     needs:
-      - get-changed-files
+      - meta
       - check-codestyle-python
       - check-codestyle-rust
     runs-on: ubuntu-22.04
@@ -129,7 +137,20 @@ jobs:
         run: exit 1
         if: |
           false
-          || (needs.check-codestyle-python.result == 'skipped' && needs.get-changed-files.outputs.python-changed == 'true')
-          || (needs.check-codestyle-rust.result   == 'skipped' && needs.get-changed-files.outputs.rust-changed   == 'true')
+          || (github.event_name == 'merge_group' && needs.meta.outputs.branch != 'main')
+          || (needs.check-codestyle-python.result == 'skipped' && needs.meta.outputs.python-changed == 'true')
+          || (needs.check-codestyle-rust.result   == 'skipped' && needs.meta.outputs.rust-changed   == 'true')
           || contains(needs.*.result, 'failure')
           || contains(needs.*.result, 'cancelled')
+
+      - name: Add fast-forward label to PR to trigger fast-forward merge
+        if: >-
+          ${{
+            always()
+            && github.event_name == 'merge_group'
+            && contains(fromJson('["release", "release-proxy", "release-compute"]'), github.base_ref)
+          }}
+        env:
+          GH_TOKEN: ${{ secrets.CI_ACCESS_TOKEN }}
+        run: >-
+          gh pr edit ${{ needs.meta.outputs.pr-number }} --repo "${GITHUB_REPOSITORY}" --add-label "fast-forward"

.github/workflows/release.yml

@@ -38,7 +38,7 @@ jobs:
     uses: ./.github/workflows/_create-release-pr.yml
     with:
       component-name: 'Storage'
-      release-branch: 'release'
+      source-branch: ${{ github.ref_name }}
     secrets:
       ci-access-token: ${{ secrets.CI_ACCESS_TOKEN }}
 
@@ -51,7 +51,7 @@ jobs:
     uses: ./.github/workflows/_create-release-pr.yml
     with:
       component-name: 'Proxy'
-      release-branch: 'release-proxy'
+      source-branch: ${{ github.ref_name }}
     secrets:
       ci-access-token: ${{ secrets.CI_ACCESS_TOKEN }}
 
@@ -64,6 +64,6 @@ jobs:
     uses: ./.github/workflows/_create-release-pr.yml
     with:
       component-name: 'Compute'
-      release-branch: 'release-compute'
+      source-branch: ${{ github.ref_name }}
     secrets:
       ci-access-token: ${{ secrets.CI_ACCESS_TOKEN }}

Cargo.lock

@@ -191,7 +191,7 @@ checksum = "965c2d33e53cb6b267e148a4cb0760bc01f4904c1cd4bb4002a085bb016d1490"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
  "synstructure",
 ]
 
@@ -203,7 +203,7 @@ checksum = "7b18050c2cd6fe86c3a76584ef5e0baf286d038cda203eb6223df2cc413565f7"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -272,7 +272,7 @@ checksum = "16e62a023e7c117e27523144c5d2459f4397fcc3cab0085af8e2224f643a0193"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -283,7 +283,7 @@ checksum = "b9ccdd8f2a161be9bd5c023df56f1b2a0bd1d83872ae53b71a84a12c9bf6e842"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -1021,7 +1021,7 @@ dependencies = [
  "regex",
  "rustc-hash 2.1.1",
  "shlex",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -1248,7 +1248,7 @@ dependencies = [
  "heck",
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -1703,7 +1703,7 @@ checksum = "f46882e17999c6cc590af592290432be3bce0428cb0d5f8b6715e4dc7b383eb3"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -1727,7 +1727,7 @@ dependencies = [
  "proc-macro2",
  "quote",
  "strsim 0.10.0",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -1738,7 +1738,7 @@ checksum = "29a358ff9f12ec09c3e61fef9b5a9902623a695a46a917b07f269bff1445611a"
 dependencies = [
  "darling_core",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -1888,7 +1888,7 @@ dependencies = [
  "dsl_auto_type",
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -1908,7 +1908,7 @@ version = "0.2.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "209c735641a413bc68c4923a9d6ad4bcb3ca306b794edaa7eb0b3228a99ffb25"
 dependencies = [
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -1937,7 +1937,7 @@ checksum = "487585f4d0c6655fe74905e2504d8ad6908e4db67f744eb140876906c2f3175d"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -1960,7 +1960,7 @@ dependencies = [
  "heck",
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -2105,7 +2105,7 @@ dependencies = [
  "darling",
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -2115,28 +2115,19 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "186e05a59d4c50738528153b83b0b0194d3a29507dfec16eccd4b342903397d0"
 dependencies = [
  "log",
-]
-
-[[package]]
-name = "env_logger"
-version = "0.10.2"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "4cd405aab171cb85d6735e5c8d9db038c17d3ca007a4d2c25f337935c3d90580"
-dependencies = [
- "humantime",
- "is-terminal",
- "log",
  "regex",
- "termcolor",
 ]
 
 [[package]]
 name = "env_logger"
-version = "0.11.2"
+version = "0.11.7"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "6c012a26a7f605efc424dd53697843a72be7dc86ad2d01f7814337794a12231d"
+checksum = "c3716d7a920fb4fac5d84e9d4bce8ceb321e9414b4409da61b07b75c1e3d0697"
 dependencies = [
+ "anstream",
+ "anstyle",
  "env_filter",
+ "jiff",
  "log",
 ]
 
@@ -2157,7 +2148,7 @@ checksum = "3bf679796c0322556351f287a51b49e48f7c4986e727b5dd78c972d30e2e16cc"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -2417,7 +2408,7 @@ checksum = "162ee34ebcb7c64a8abebc059ce0fee27c2262618d7b60ed8faf72fef13c3650"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -2530,7 +2521,7 @@ checksum = "53010ccb100b96a67bc32c0175f0ed1426b31b655d562898e57325f81c023ac0"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -3148,7 +3139,7 @@ checksum = "1ec89e9337638ecdc08744df490b221a7399bf8d164eb52a665454e60e075ad6"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -3241,7 +3232,7 @@ dependencies = [
  "crossbeam-channel",
  "crossbeam-utils",
  "dashmap 6.1.0",
- "env_logger 0.11.2",
+ "env_logger",
  "indexmap 2.0.1",
  "itoa",
  "log",
@@ -3254,11 +3245,11 @@ dependencies = [
 
 [[package]]
 name = "inotify"
-version = "0.11.0"
+version = "0.9.6"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "f37dccff2791ab604f9babef0ba14fbe0be30bd368dc541e2b08d07c8aa908f3"
+checksum = "f8069d3ec154eb856955c1c0fbffefbf5f3c40a104ec912d4797314c1801abff"
 dependencies = [
- "bitflags 2.8.0",
+ "bitflags 1.3.2",
  "inotify-sys",
  "libc",
 ]
@@ -3364,6 +3355,30 @@ dependencies = [
  "tracing",
 ]
 
+[[package]]
+name = "jiff"
+version = "0.2.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d699bc6dfc879fb1bf9bdff0d4c56f0884fc6f0d0eb0fba397a6d00cd9a6b85e"
+dependencies = [
+ "jiff-static",
+ "log",
+ "portable-atomic",
+ "portable-atomic-util",
+ "serde",
+]
+
+[[package]]
+name = "jiff-static"
+version = "0.2.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "8d16e75759ee0aa64c57a56acbf43916987b20c77373cb7e808979e02b93c9f9"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn 2.0.100",
+]
+
 [[package]]
 name = "jobserver"
 version = "0.1.32"
@@ -3535,9 +3550,9 @@ dependencies = [
 
 [[package]]
 name = "log"
-version = "0.4.20"
+version = "0.4.26"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "b5e6163cb8c49088c2c36f57875e58ccd8c87c7427f7fbd50ea6710b2f3f2e8f"
+checksum = "30bde2b3dc3671ae49d8e2e9f044c7c005836e7a023ee57cffa25ab82764bb9e"
 
 [[package]]
 name = "lru"
@@ -3618,7 +3633,7 @@ dependencies = [
  "heck",
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -3732,6 +3747,18 @@ dependencies = [
  "adler2",
 ]
 
+[[package]]
+name = "mio"
+version = "0.8.11"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a4a650543ca06a924e8b371db273b2756685faae30f8487da1b56505a8f78b0c"
+dependencies = [
+ "libc",
+ "log",
+ "wasi 0.11.0+wasi-snapshot-preview1",
+ "windows-sys 0.48.0",
+]
+
 [[package]]
 name = "mio"
 version = "1.0.3"
@@ -3739,7 +3766,6 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "2886843bf800fba2e3377cff24abf6379b4c4d5c6681eaf9ea5b0d15090450bd"
 dependencies = [
  "libc",
- "log",
  "wasi 0.11.0+wasi-snapshot-preview1",
  "windows-sys 0.52.0",
 ]
@@ -3817,29 +3843,23 @@ checksum = "38bf9645c8b145698bb0b18a4637dcacbc421ea49bef2317e4fd8065a387cf21"
 
 [[package]]
 name = "notify"
-version = "8.0.0"
+version = "6.1.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "2fee8403b3d66ac7b26aee6e40a897d85dc5ce26f44da36b8b73e987cc52e943"
+checksum = "6205bd8bb1e454ad2e27422015fb5e4f2bcc7e08fa8f27058670d208324a4d2d"
 dependencies = [
  "bitflags 2.8.0",
+ "crossbeam-channel",
  "filetime",
  "fsevent-sys",
  "inotify",
  "kqueue",
  "libc",
  "log",
- "mio",
- "notify-types",
+ "mio 0.8.11",
  "walkdir",
- "windows-sys 0.59.0",
+ "windows-sys 0.48.0",
 ]
 
-[[package]]
-name = "notify-types"
-version = "2.0.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5e0826a989adedc2a244799e823aece04662b66609d96af8dff7ac6df9a8925d"
-
 [[package]]
 name = "ntapi"
 version = "0.4.1"
@@ -4308,6 +4328,7 @@ dependencies = [
  "tokio-util",
  "toml_edit",
  "tracing",
+ "tracing-utils",
  "url",
  "utils",
  "uuid",
@@ -4485,7 +4506,7 @@ dependencies = [
  "parquet",
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -4587,7 +4608,7 @@ checksum = "f6e859e6e5bd50440ab63c47e3ebabc90f26251f7c73c3d3e837b74a1cc3fa67"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -4683,6 +4704,15 @@ version = "1.10.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "280dc24453071f1b63954171985a0b0d30058d287960968b9b2aca264c8d4ee6"
 
+[[package]]
+name = "portable-atomic-util"
+version = "0.2.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d8a2f0d8d040d7848a709caf78912debcc3f33ee4b3cac47d73d1e1069e83507"
+dependencies = [
+ "portable-atomic",
+]
+
 [[package]]
 name = "postgres"
 version = "0.19.7"
@@ -4790,7 +4820,7 @@ dependencies = [
  "bytes",
  "crc32c",
  "criterion",
- "env_logger 0.10.2",
+ "env_logger",
  "log",
  "memoffset 0.9.0",
  "once_cell",
@@ -4889,7 +4919,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "8d3928fb5db768cb86f891ff014f0144589297e3c6a1aba6ed7cecfdace270c7"
 dependencies = [
  "proc-macro2",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -4903,9 +4933,9 @@ dependencies = [
 
 [[package]]
 name = "proc-macro2"
-version = "1.0.92"
+version = "1.0.94"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "37d3544b3f2748c54e147655edb5025752e2303145b5aefb3c3ea2c78b973bb0"
+checksum = "a31971752e70b8b2686d7e46ec17fb38dad4051d94024c88df49b667caea9c84"
 dependencies = [
  "unicode-ident",
 ]
@@ -4980,7 +5010,7 @@ checksum = "22505a5c94da8e3b7c2996394d1c933236c4d743e81a410bcca4e6989fc066a4"
 dependencies = [
  "bytes",
  "heck",
- "itertools 0.10.5",
+ "itertools 0.12.1",
  "log",
  "multimap",
  "once_cell",
@@ -4989,7 +5019,7 @@ dependencies = [
  "prost 0.12.6",
  "prost-types 0.12.6",
  "regex",
- "syn 2.0.90",
+ "syn 2.0.100",
  "tempfile",
 ]
 
@@ -5010,7 +5040,7 @@ dependencies = [
  "prost 0.13.3",
  "prost-types 0.13.3",
  "regex",
- "syn 2.0.90",
+ "syn 2.0.100",
  "tempfile",
 ]
 
@@ -5021,10 +5051,10 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "81bddcdb20abf9501610992b6759a4c888aef7d1a7247ef75e2404275ac24af1"
 dependencies = [
  "anyhow",
- "itertools 0.10.5",
+ "itertools 0.12.1",
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -5037,7 +5067,7 @@ dependencies = [
  "itertools 0.12.1",
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -5084,7 +5114,7 @@ dependencies = [
  "consumption_metrics",
  "ecdsa 0.16.9",
  "ed25519-dalek",
- "env_logger 0.10.2",
+ "env_logger",
  "fallible-iterator",
  "flate2",
  "framed-websockets",
@@ -5221,9 +5251,9 @@ dependencies = [
 
 [[package]]
 name = "quote"
-version = "1.0.37"
+version = "1.0.39"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "b5b9d34b8991d19d98081b46eacdd8eb58c6f2b201139f7c5f643cc155a633af"
+checksum = "c1f1914ce909e1658d9907913b4b91947430c7d9be598b15a1912935b8c04801"
 dependencies = [
  "proc-macro2",
 ]
@@ -5752,7 +5782,7 @@ dependencies = [
  "regex",
  "relative-path",
  "rustc_version",
- "syn 2.0.90",
+ "syn 2.0.100",
  "unicode-ident",
 ]
 
@@ -5967,7 +5997,7 @@ dependencies = [
  "crc32c",
  "criterion",
  "desim",
- "env_logger 0.10.2",
+ "env_logger",
  "fail",
  "futures",
  "hex",
@@ -6298,7 +6328,7 @@ checksum = "ad1e866f866923f252f05c889987993144fb74e722403468a4ebd70c3cd756c0"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -6380,7 +6410,7 @@ dependencies = [
  "darling",
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -6782,7 +6812,7 @@ dependencies = [
  "proc-macro2",
  "quote",
  "rustversion",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -6833,9 +6863,9 @@ dependencies = [
 
 [[package]]
 name = "syn"
-version = "2.0.90"
+version = "2.0.100"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "919d3b74a5dd0ccd15aeb8f93e7006bd9e14c295087c9896a110f490752bcf31"
+checksum = "b09a44accad81e1ba1cd74a32461ba89dee89095ba17b32f5d03683b1b1fc2a0"
 dependencies = [
  "proc-macro2",
  "quote",
@@ -6865,7 +6895,7 @@ checksum = "c8af7666ab7b6390ab78131fb5b0fce11d6b7a6951602017c35fa82800708971"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -6916,15 +6946,6 @@ dependencies = [
  "serde_json",
 ]
 
-[[package]]
-name = "termcolor"
-version = "1.2.0"
-source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "be55cf8942feac5c765c2c993422806843c9a9a45d4d5c407ad6dd2ea95eb9b6"
-dependencies = [
- "winapi-util",
-]
-
 [[package]]
 name = "test-context"
 version = "0.3.0"
@@ -6943,7 +6964,7 @@ checksum = "78ea17a2dc368aeca6f554343ced1b1e31f76d63683fa8016e5844bd7a5144a1"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -6972,7 +6993,7 @@ checksum = "4fee6c4efc90059e10f81e6d42c60a18f76588c3d74cb83a0b242a2b6c7504c1"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -6983,7 +7004,7 @@ checksum = "26afc1baea8a989337eeb52b6e72a039780ce45c3edfcc9c5b9d112feeb173c2"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -7123,7 +7144,7 @@ dependencies = [
  "backtrace",
  "bytes",
  "libc",
- "mio",
+ "mio 1.0.3",
  "parking_lot 0.12.1",
  "pin-project-lite",
  "signal-hook-registry",
@@ -7166,7 +7187,7 @@ checksum = "6e06d43f1345a3bcd39f6a56dbb7dcab2ba47e68e8ac134855e7e2bdbaf8cab8"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -7399,7 +7420,7 @@ dependencies = [
  "prost-build 0.13.3",
  "prost-types 0.13.3",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -7514,7 +7535,7 @@ checksum = "395ae124c09f9e6918a2310af6038fba074bcf474ac352496d5910dd59a2226d"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -7845,6 +7866,7 @@ dependencies = [
  "tracing",
  "tracing-error",
  "tracing-subscriber",
+ "tracing-utils",
  "walkdir",
 ]
 
@@ -7908,7 +7930,7 @@ dependencies = [
  "anyhow",
  "camino-tempfile",
  "clap",
- "env_logger 0.10.2",
+ "env_logger",
  "log",
  "postgres",
  "postgres_ffi",
@@ -8013,7 +8035,7 @@ dependencies = [
  "once_cell",
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
  "wasm-bindgen-shared",
 ]
 
@@ -8047,7 +8069,7 @@ checksum = "e94f17b526d0a461a191c78ea52bbce64071ed5c04c9ffe424dcb38f74171bb7"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
  "wasm-bindgen-backend",
  "wasm-bindgen-shared",
 ]
@@ -8354,6 +8376,7 @@ name = "workspace_hack"
 version = "0.1.0"
 dependencies = [
  "ahash",
+ "anstream",
  "anyhow",
  "base64 0.13.1",
  "base64 0.21.7",
@@ -8370,6 +8393,8 @@ dependencies = [
  "digest",
  "displaydoc",
  "either",
+ "env_filter",
+ "env_logger",
  "fail",
  "form_urlencoded",
  "futures-channel",
@@ -8387,7 +8412,6 @@ dependencies = [
  "hyper-util",
  "indexmap 1.9.3",
  "indexmap 2.0.1",
- "itertools 0.10.5",
  "itertools 0.12.1",
  "lazy_static",
  "libc",
@@ -8423,7 +8447,7 @@ dependencies = [
  "spki 0.7.3",
  "stable_deref_trait",
  "subtle",
- "syn 2.0.90",
+ "syn 2.0.100",
  "sync_wrapper 0.1.2",
  "tikv-jemalloc-ctl",
  "tikv-jemalloc-sys",
@@ -8540,7 +8564,7 @@ checksum = "2380878cad4ac9aac1e2435f3eb4020e8374b5f13c296cb75b4620ff8e229154"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
  "synstructure",
 ]
 
@@ -8562,7 +8586,7 @@ checksum = "b3c129550b3e6de3fd0ba67ba5c81818f9805e58b8d7fee80a3a59d2c9fc601a"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -8582,7 +8606,7 @@ checksum = "595eed982f7d355beb85837f651fa22e90b3c044842dc7f2c2842c086f295808"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
  "synstructure",
 ]
 
@@ -8604,7 +8628,7 @@ checksum = "ce36e65b0d2999d2aafac989fb249189a141aee1f53c612c1f37d72631959f69"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]
@@ -8626,7 +8650,7 @@ checksum = "6eafa6dfb17584ea3e2bd6e76e0cc15ad7af12b09abdd1ca55961bed9b1063c6"
 dependencies = [
  "proc-macro2",
  "quote",
- "syn 2.0.90",
+ "syn 2.0.100",
 ]
 
 [[package]]

Cargo.toml

@@ -126,7 +126,9 @@ measured = { version = "0.0.22", features=["lasso"] }
 measured-process = { version = "0.0.22" }
 memoffset = "0.9"
 nix = { version = "0.27", features = ["dir", "fs", "process", "socket", "signal", "poll"] }
-notify = "8.0.0"
+# Do not update to >= 7.0.0, at least. The update will have a significant impact
+# on compute startup metrics (start_postgres_ms), >= 25% degradation.
+notify = "6.0.0"
 num_cpus = "1.15"
 num-traits = "0.2.15"
 once_cell = "1.13"
@@ -219,7 +221,7 @@ zerocopy = { version = "0.7", features = ["derive"] }
 json-structural-diff = { version = "0.2.0" }
 
 ## TODO replace this with tracing
-env_logger = "0.10"
+env_logger = "0.11"
 log = "0.4"
 
 ## Libraries from neondatabase/ git forks, ideally with changes to be upstreamed

compute/compute-node.Dockerfile

@@ -1055,34 +1055,6 @@ RUN  if [ -d pg_embedding-src ]; then \
         make -j $(getconf _NPROCESSORS_ONLN) install; \
     fi
 
-#########################################################################################
-#
-# Layer "pg_anon-build"
-# compile anon extension
-#
-#########################################################################################
-FROM build-deps AS pg_anon-src
-ARG PG_VERSION
-
-# This is an experimental extension, never got to real production.
-# !Do not remove! It can be present in shared_preload_libraries and compute will fail to start if library is not found.
-WORKDIR /ext-src
-RUN case "${PG_VERSION:?}" in "v17") \
-    echo "postgresql_anonymizer does not yet support PG17" && exit 0;; \
-    esac && \
-    wget  https://github.com/neondatabase/postgresql_anonymizer/archive/refs/tags/neon_1.1.1.tar.gz -O pg_anon.tar.gz && \
-    echo "321ea8d5c1648880aafde850a2c576e4a9e7b9933a34ce272efc839328999fa9  pg_anon.tar.gz" | sha256sum --check && \
-    mkdir pg_anon-src && cd pg_anon-src && tar xzf ../pg_anon.tar.gz --strip-components=1 -C .
-
-FROM pg-build AS pg_anon-build
-COPY --from=pg_anon-src /ext-src/ /ext-src/
-WORKDIR /ext-src
-RUN if [ -d pg_anon-src ]; then \
-        cd pg_anon-src && \
-        make -j $(getconf _NPROCESSORS_ONLN) install && \
-        echo 'trusted = true' >> /usr/local/pgsql/share/extension/anon.control; \
-    fi
-
 #########################################################################################
 #
 # Layer "pg build with nonroot user and cargo installed"
@@ -1380,6 +1352,27 @@ COPY --from=pg_session_jwt-src /ext-src/ /ext-src/
 WORKDIR /ext-src/pg_session_jwt-src
 RUN cargo pgrx install --release
 
+#########################################################################################
+#
+# Layer "pg-anon-pg-build"
+# compile anon extension
+#
+#########################################################################################
+FROM rust-extensions-build-pgrx12 AS pg-anon-pg-build
+ARG PG_VERSION
+COPY --from=pg-build /usr/local/pgsql/ /usr/local/pgsql/
+
+# This is an experimental extension, never got to real production.
+# !Do not remove! It can be present in shared_preload_libraries and compute will fail to start if library is not found.
+ENV PATH="/usr/local/pgsql/bin/:$PATH"
+RUN wget https://gitlab.com/dalibo/postgresql_anonymizer/-/archive/latest/postgresql_anonymizer-latest.tar.gz -O pg_anon.tar.gz && \
+    mkdir pg_anon-src && cd pg_anon-src && tar xzf ../pg_anon.tar.gz --strip-components=1 -C . && \
+    find /usr/local/pgsql -type f | sed 's|^/usr/local/pgsql/||' > /before.txt && \
+    sed -i 's/pgrx = "0.12.9"/pgrx = { version = "=0.12.9", features = [ "unsafe-postgres" ] }/g' Cargo.toml && \
+    make -j $(getconf _NPROCESSORS_ONLN) extension PG_CONFIG=/usr/local/pgsql/bin/pg_config PGVER=pg$(echo "$PG_VERSION" | sed 's/^v//') && \
+    make -j $(getconf _NPROCESSORS_ONLN) install PG_CONFIG=/usr/local/pgsql/bin/pg_config PGVER=pg$(echo "$PG_VERSION" | sed 's/^v//') && \
+    echo 'trusted = true' >> /usr/local/pgsql/share/extension/anon.control
+
 #########################################################################################
 #
 # Layer "wal2json-build"
@@ -1677,7 +1670,6 @@ COPY --from=pg_roaringbitmap-build /usr/local/pgsql/ /usr/local/pgsql/
 COPY --from=pg_semver-build /usr/local/pgsql/ /usr/local/pgsql/
 COPY --from=pg_embedding-build /usr/local/pgsql/ /usr/local/pgsql/
 COPY --from=wal2json-build /usr/local/pgsql /usr/local/pgsql
-COPY --from=pg_anon-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/

compute_tools/src/compute.rs

@@ -645,9 +645,9 @@ impl ComputeNode {
         if pspec.spec.mode == ComputeMode::Primary {
             self.configure_as_primary(&compute_state)?;
 
-            let conf = self.get_conn_conf(None);
-            tokio::task::spawn_blocking(|| {
-                let res = get_installed_extensions(conf);
+            let conf = self.get_tokio_conn_conf(None);
+            tokio::task::spawn(async {
+                let res = get_installed_extensions(conf).await;
                 match res {
                     Ok(extensions) => {
                         info!(
@@ -855,12 +855,6 @@ impl ComputeNode {
             info!("Storage auth token not set");
         }
 
-        if let Some(spec) = &compute_state.pspec {
-            config.application_name(&format!("compute_ctl-{}", spec.spec.mode.to_type_str()));
-        } else {
-            config.application_name("compute_ctl");
-        }
-
         // Connect to pageserver
         let mut client = config.connect(NoTls)?;
         let pageserver_connect_micros = start_time.elapsed().as_micros() as u64;

compute_tools/src/config.rs

@@ -99,7 +99,6 @@ pub fn write_postgres_conf(
         writeln!(file, "lc_numeric='C.UTF-8'")?;
     }
 
-    writeln!(file, "neon.endpoint_type={}", spec.mode.to_type_str())?;
     match spec.mode {
         ComputeMode::Primary => {}
         ComputeMode::Static(lsn) => {

compute_tools/src/http/middleware/mod.rs

@@ -1 +1,2 @@
 pub(in crate::http) mod authorize;
+pub(in crate::http) mod request_id;

compute_tools/src/http/middleware/request_id.rs

@@ -0,0 +1,16 @@
+use axum::{extract::Request, middleware::Next, response::Response};
+use uuid::Uuid;
+
+use crate::http::headers::X_REQUEST_ID;
+
+/// This middleware function allows compute_ctl to generate its own request ID
+/// if one isn't supplied. The control plane will always send one as a UUID. The
+/// neon Postgres extension on the other hand does not send one.
+pub async fn maybe_add_request_id_header(mut request: Request, next: Next) -> Response {
+    let headers = request.headers_mut();
+    if !headers.contains_key(X_REQUEST_ID) {
+        headers.append(X_REQUEST_ID, Uuid::new_v4().to_string().parse().unwrap());
+    }
+
+    next.run(request).await
+}

compute_tools/src/http/server.rs

@@ -5,9 +5,8 @@ use std::time::Duration;
 
 use anyhow::Result;
 use axum::Router;
-use axum::extract::Request;
-use axum::middleware::{self, Next};
-use axum::response::{IntoResponse, Response};
+use axum::middleware::{self};
+use axum::response::IntoResponse;
 use axum::routing::{get, post};
 use http::StatusCode;
 use jsonwebtoken::jwk::JwkSet;
@@ -17,8 +16,8 @@ use tower_http::{
     auth::AsyncRequireAuthorizationLayer, request_id::PropagateRequestIdLayer, trace::TraceLayer,
 };
 use tracing::{Span, error, info};
-use uuid::Uuid;
 
+use super::middleware::request_id::maybe_add_request_id_header;
 use super::{
     headers::X_REQUEST_ID,
     middleware::authorize::Authorize,
@@ -219,15 +218,3 @@ impl Server {
         tokio::spawn(self.serve(state));
     }
 }
-
-/// This middleware function allows compute_ctl to generate its own request ID
-/// if one isn't supplied. The control plane will always send one as a UUID. The
-/// neon Postgres extension on the other hand does not send one.
-async fn maybe_add_request_id_header(mut request: Request, next: Next) -> Response {
-    let headers = request.headers_mut();
-    if headers.get(X_REQUEST_ID).is_none() {
-        headers.append(X_REQUEST_ID, Uuid::new_v4().to_string().parse().unwrap());
-    }
-
-    next.run(request).await
-}

compute_tools/src/installed_extensions.rs

@@ -2,7 +2,7 @@ use std::collections::HashMap;
 
 use anyhow::Result;
 use compute_api::responses::{InstalledExtension, InstalledExtensions};
-use postgres::{Client, NoTls};
+use tokio_postgres::{Client, Config, NoTls};
 
 use crate::metrics::INSTALLED_EXTENSIONS;
 
@@ -10,7 +10,7 @@ use crate::metrics::INSTALLED_EXTENSIONS;
 /// and to make database listing query here more explicit.
 ///
 /// Limit the number of databases to 500 to avoid excessive load.
-fn list_dbs(client: &mut Client) -> Result<Vec<String>> {
+async fn list_dbs(client: &mut Client) -> Result<Vec<String>> {
     // `pg_database.datconnlimit = -2` means that the database is in the
     // invalid state
     let databases = client
@@ -20,7 +20,8 @@ fn list_dbs(client: &mut Client) -> Result<Vec<String>> {
                 AND datconnlimit <> - 2
                 LIMIT 500",
             &[],
-        )?
+        )
+        .await?
         .iter()
         .map(|row| {
             let db: String = row.get("datname");
@@ -36,20 +37,36 @@ fn list_dbs(client: &mut Client) -> Result<Vec<String>> {
 /// Same extension can be installed in multiple databases with different versions,
 /// so we report a separate metric (number of databases where it is installed)
 /// for each extension version.
-pub fn get_installed_extensions(mut conf: postgres::config::Config) -> Result<InstalledExtensions> {
+pub async fn get_installed_extensions(mut conf: 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 databases: Vec<String> = {
+        let (mut client, connection) = conf.connect(NoTls).await?;
+        tokio::spawn(async move {
+            if let Err(e) = connection.await {
+                eprintln!("connection error: {}", e);
+            }
+        });
+
+        list_dbs(&mut client).await?
+    };
 
     let mut extensions_map: HashMap<(String, String, 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 (client, connection) = conf.connect(NoTls).await?;
+        tokio::spawn(async move {
+            if let Err(e) = connection.await {
+                eprintln!("connection error: {}", e);
+            }
+        });
+
+        let extensions: Vec<(String, String, i32)> = client
             .query(
                 "SELECT extname, extversion, extowner::integer FROM pg_catalog.pg_extension",
                 &[],
-            )?
+            )
+            .await?
             .iter()
             .map(|row| {
                 (

compute_tools/src/logger.rs

@@ -24,7 +24,8 @@ pub async fn init_tracing_and_logging(default_log_level: &str) -> anyhow::Result
         .with_writer(std::io::stderr);
 
     // Initialize OpenTelemetry
-    let otlp_layer = tracing_utils::init_tracing("compute_ctl").await;
+    let otlp_layer =
+        tracing_utils::init_tracing("compute_ctl", tracing_utils::ExportConfig::default()).await;
 
     // Put it all together
     tracing_subscriber::registry()

control_plane/src/bin/neon_local.rs

@@ -36,7 +36,9 @@ use pageserver_api::config::{
 use pageserver_api::controller_api::{
     NodeAvailabilityWrapper, PlacementPolicy, TenantCreateRequest,
 };
-use pageserver_api::models::{ShardParameters, TimelineCreateRequest, TimelineInfo};
+use pageserver_api::models::{
+    ShardParameters, TenantConfigRequest, TimelineCreateRequest, TimelineInfo,
+};
 use pageserver_api::shard::{ShardCount, ShardStripeSize, TenantShardId};
 use postgres_backend::AuthType;
 use postgres_connection::parse_host_port;
@@ -1129,12 +1131,16 @@ async fn handle_tenant(subcmd: &TenantCmd, env: &mut local_env::LocalEnv) -> any
             let tenant_id = get_tenant_id(args.tenant_id, env)?;
             let tenant_conf: HashMap<_, _> =
                 args.config.iter().flat_map(|c| c.split_once(':')).collect();
+            let config = PageServerNode::parse_config(tenant_conf)?;
 
-            pageserver
-                .tenant_config(tenant_id, tenant_conf)
+            let req = TenantConfigRequest { tenant_id, config };
+
+            let storage_controller = StorageController::from_env(env);
+            storage_controller
+                .set_tenant_config(&req)
                 .await
                 .with_context(|| format!("Tenant config failed for tenant with id {tenant_id}"))?;
-            println!("tenant {tenant_id} successfully configured on the pageserver");
+            println!("tenant {tenant_id} successfully configured via storcon");
         }
     }
     Ok(())

control_plane/src/storage_controller.rs

@@ -14,7 +14,7 @@ use pageserver_api::controller_api::{
     NodeConfigureRequest, NodeDescribeResponse, NodeRegisterRequest, TenantCreateRequest,
     TenantCreateResponse, TenantLocateResponse,
 };
-use pageserver_api::models::{TimelineCreateRequest, TimelineInfo};
+use pageserver_api::models::{TenantConfigRequest, TimelineCreateRequest, TimelineInfo};
 use pageserver_api::shard::TenantShardId;
 use pageserver_client::mgmt_api::ResponseErrorMessageExt;
 use postgres_backend::AuthType;
@@ -878,4 +878,9 @@ impl StorageController {
         )
         .await
     }
+
+    pub async fn set_tenant_config(&self, req: &TenantConfigRequest) -> anyhow::Result<()> {
+        self.dispatch(Method::PUT, "v1/tenant/config".to_string(), Some(req))
+            .await
+    }
 }

control_plane/storcon_cli/src/main.rs

@@ -14,8 +14,8 @@ use pageserver_api::controller_api::{
     TenantShardMigrateRequest, TenantShardMigrateResponse,
 };
 use pageserver_api::models::{
-    EvictionPolicy, EvictionPolicyLayerAccessThreshold, LocationConfigSecondary, ShardParameters,
-    TenantConfig, TenantConfigPatchRequest, TenantConfigRequest, TenantShardSplitRequest,
+    EvictionPolicy, EvictionPolicyLayerAccessThreshold, ShardParameters, TenantConfig,
+    TenantConfigPatchRequest, TenantConfigRequest, TenantShardSplitRequest,
     TenantShardSplitResponse,
 };
 use pageserver_api::shard::{ShardStripeSize, TenantShardId};
@@ -158,12 +158,6 @@ enum Command {
         #[arg(long)]
         tenant_id: TenantId,
     },
-    /// For a tenant which hasn't been onboarded to the storage controller yet, add it in secondary
-    /// mode so that it can warm up content on a pageserver.
-    TenantWarmup {
-        #[arg(long)]
-        tenant_id: TenantId,
-    },
     TenantSetPreferredAz {
         #[arg(long)]
         tenant_id: TenantId,
@@ -871,94 +865,6 @@ async fn main() -> anyhow::Result<()> {
                 )
                 .await?;
         }
-        Command::TenantWarmup { tenant_id } => {
-            let describe_response = storcon_client
-                .dispatch::<(), TenantDescribeResponse>(
-                    Method::GET,
-                    format!("control/v1/tenant/{tenant_id}"),
-                    None,
-                )
-                .await;
-            match describe_response {
-                Ok(describe) => {
-                    if matches!(describe.policy, PlacementPolicy::Secondary) {
-                        // Fine: it's already known to controller in secondary mode: calling
-                        // again to put it into secondary mode won't cause problems.
-                    } else {
-                        anyhow::bail!("Tenant already present with policy {:?}", describe.policy);
-                    }
-                }
-                Err(mgmt_api::Error::ApiError(StatusCode::NOT_FOUND, _)) => {
-                    // Fine: this tenant isn't know to the storage controller yet.
-                }
-                Err(e) => {
-                    // Unexpected API error
-                    return Err(e.into());
-                }
-            }
-
-            vps_client
-                .location_config(
-                    TenantShardId::unsharded(tenant_id),
-                    pageserver_api::models::LocationConfig {
-                        mode: pageserver_api::models::LocationConfigMode::Secondary,
-                        generation: None,
-                        secondary_conf: Some(LocationConfigSecondary { warm: true }),
-                        shard_number: 0,
-                        shard_count: 0,
-                        shard_stripe_size: ShardParameters::DEFAULT_STRIPE_SIZE.0,
-                        tenant_conf: TenantConfig::default(),
-                    },
-                    None,
-                    true,
-                )
-                .await?;
-
-            let describe_response = storcon_client
-                .dispatch::<(), TenantDescribeResponse>(
-                    Method::GET,
-                    format!("control/v1/tenant/{tenant_id}"),
-                    None,
-                )
-                .await?;
-
-            let secondary_ps_id = describe_response
-                .shards
-                .first()
-                .unwrap()
-                .node_secondary
-                .first()
-                .unwrap();
-
-            println!("Tenant {tenant_id} warming up on pageserver {secondary_ps_id}");
-            loop {
-                let (status, progress) = vps_client
-                    .tenant_secondary_download(
-                        TenantShardId::unsharded(tenant_id),
-                        Some(Duration::from_secs(10)),
-                    )
-                    .await?;
-                println!(
-                    "Progress: {}/{} layers, {}/{} bytes",
-                    progress.layers_downloaded,
-                    progress.layers_total,
-                    progress.bytes_downloaded,
-                    progress.bytes_total
-                );
-                match status {
-                    StatusCode::OK => {
-                        println!("Download complete");
-                        break;
-                    }
-                    StatusCode::ACCEPTED => {
-                        // Loop
-                    }
-                    _ => {
-                        anyhow::bail!("Unexpected download status: {status}");
-                    }
-                }
-            }
-        }
         Command::TenantDrop { tenant_id, unclean } => {
             if !unclean {
                 anyhow::bail!(

docs/storage_controller.md

@@ -101,7 +101,7 @@ changes such as a pageserver node becoming unavailable, or the tenant's shard co
 postgres clients to handle such changes, the storage controller calls an API hook when a tenant's pageserver
 location changes.
 
-The hook is configured using the storage controller's `--compute-hook-url` CLI option. If the hook requires
+The hook is configured using the storage controller's `--control-plane-url` CLI option. If the hook requires
 JWT auth, the token may be provided with `--control-plane-jwt-token`. The hook will be invoked with a `PUT` request.
 
 In the Neon cloud service, this hook is implemented by Neon's internal cloud control plane. In `neon_local` systems

libs/compute_api/src/spec.rs

@@ -272,18 +272,6 @@ pub enum ComputeMode {
     Replica,
 }
 
-impl ComputeMode {
-    /// Convert the compute mode to a string that can be used to identify the type of compute,
-    /// which means that if it's a static compute, the LSN will not be included.
-    pub fn to_type_str(&self) -> &'static str {
-        match self {
-            ComputeMode::Primary => "primary",
-            ComputeMode::Static(_) => "static",
-            ComputeMode::Replica => "replica",
-        }
-    }
-}
-
 /// Log level for audit logging
 /// Disabled, log, hipaa
 /// Default is Disabled

libs/pageserver_api/src/models.rs

@@ -1225,9 +1225,10 @@ pub struct TimelineInfo {
     pub last_record_lsn: Lsn,
     pub prev_record_lsn: Option<Lsn>,
 
-    /// Legacy field for compat with control plane.  Synonym of `min_readable_lsn`.
-    /// TODO: remove once control plane no longer reads it.
-    pub latest_gc_cutoff_lsn: Lsn,
+    /// Legacy field, retained for one version to enable old storage controller to
+    /// decode (it was a mandatory field).
+    #[serde(default, rename = "latest_gc_cutoff_lsn")]
+    pub _unused: Lsn,
 
     /// The LSN up to which GC has advanced: older data may still exist but it is not available for clients.
     /// This LSN is not suitable for deciding where to create branches etc: use [`TimelineInfo::min_readable_lsn`] instead,

libs/pageserver_api/src/shard.rs

@@ -112,6 +112,16 @@ impl ShardIdentity {
         }
     }
 
+    /// An unsharded identity with the given stripe size (if non-zero). This is typically used to
+    /// carry over a stripe size for an unsharded tenant from persistent storage.
+    pub fn unsharded_with_stripe_size(stripe_size: ShardStripeSize) -> Self {
+        let mut shard_identity = Self::unsharded();
+        if stripe_size.0 > 0 {
+            shard_identity.stripe_size = stripe_size;
+        }
+        shard_identity
+    }
+
     /// A broken instance of this type is only used for `TenantState::Broken` tenants,
     /// which are constructed in code paths that don't have access to proper configuration.
     ///

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

@@ -135,8 +135,8 @@ impl Type {
 pub enum Kind {
     /// A simple type like `VARCHAR` or `INTEGER`.
     Simple,
-    /// An enumerated type along with its variants.
-    Enum(Vec<String>),
+    /// An enumerated type.
+    Enum,
     /// A pseudo-type.
     Pseudo,
     /// An array type along with the type of its elements.
@@ -146,9 +146,9 @@ pub enum Kind {
     /// A multirange type along with the type of its elements.
     Multirange(Type),
     /// A domain type along with its underlying type.
-    Domain(Type),
-    /// A composite type along with information about its fields.
-    Composite(Vec<Field>),
+    Domain(Oid),
+    /// A composite type.
+    Composite(Oid),
 }
 
 /// Information about a field of a composite type.

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

@@ -19,10 +19,10 @@ use crate::config::{Host, SslMode};
 use crate::connection::{Request, RequestMessages};
 use crate::query::RowStream;
 use crate::simple_query::SimpleQueryStream;
-use crate::types::{Oid, ToSql, Type};
+use crate::types::{Oid, Type};
 use crate::{
-    CancelToken, Error, ReadyForQueryStatus, Row, SimpleQueryMessage, Statement, Transaction,
-    TransactionBuilder, query, simple_query, slice_iter,
+    CancelToken, Error, ReadyForQueryStatus, SimpleQueryMessage, Statement, Transaction,
+    TransactionBuilder, query, simple_query,
 };
 
 pub struct Responses {
@@ -54,26 +54,18 @@ impl Responses {
 /// A cache of type info and prepared statements for fetching type info
 /// (corresponding to the queries in the [crate::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](crate::prepare::TYPEINFO_QUERY) (or its
     /// fallback).
-    typeinfo: Option<Statement>,
-    /// A statement for getting information for a composite type from its OID.
-    /// Corresponds to [TYPEINFO_QUERY](crate::prepare::TYPEINFO_COMPOSITE_QUERY).
-    typeinfo_composite: Option<Statement>,
-    /// A statement for getting information for a composite type from its OID.
-    /// Corresponds to [TYPEINFO_QUERY](crate::prepare::TYPEINFO_COMPOSITE_QUERY) (or
-    /// its fallback).
-    typeinfo_enum: Option<Statement>,
+    pub(crate) typeinfo: Option<Statement>,
 
     /// Cache of types already looked up.
-    types: HashMap<Oid, Type>,
+    pub(crate) types: HashMap<Oid, Type>,
 }
 
 pub struct InnerClient {
     sender: mpsc::UnboundedSender<Request>,
-    cached_typeinfo: Mutex<CachedTypeInfo>,
 
     /// A buffer to use when writing out postgres commands.
     buffer: Mutex<BytesMut>,
@@ -91,38 +83,6 @@ 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
@@ -142,7 +102,6 @@ pub struct SocketConfig {
     pub host: Host,
     pub port: u16,
     pub connect_timeout: Option<Duration>,
-    // pub keepalive: Option<KeepaliveConfig>,
 }
 
 /// An asynchronous PostgreSQL client.
@@ -151,6 +110,7 @@ pub struct SocketConfig {
 /// through this client object.
 pub struct Client {
     inner: Arc<InnerClient>,
+    cached_typeinfo: CachedTypeInfo,
 
     socket_config: SocketConfig,
     ssl_mode: SslMode,
@@ -169,9 +129,9 @@ impl Client {
         Client {
             inner: Arc::new(InnerClient {
                 sender,
-                cached_typeinfo: Default::default(),
                 buffer: Default::default(),
             }),
+            cached_typeinfo: Default::default(),
 
             socket_config,
             ssl_mode,
@@ -189,55 +149,6 @@ impl Client {
         &self.inner
     }
 
-    /// Executes a statement, returning a vector of the resulting rows.
-    ///
-    /// A statement may contain parameters, specified by `$n`, where `n` is the index of the parameter of the list
-    /// provided, 1-indexed.
-    ///
-    /// The `statement` argument can either be a `Statement`, or a raw query string. If the same statement will be
-    /// repeatedly executed (perhaps with different query parameters), consider preparing the statement up front
-    /// with the `prepare` method.
-    ///
-    /// # Panics
-    ///
-    /// Panics if the number of parameters provided does not match the number expected.
-    pub async fn query(
-        &self,
-        statement: Statement,
-        params: &[&(dyn ToSql + Sync)],
-    ) -> Result<Vec<Row>, Error> {
-        self.query_raw(statement, slice_iter(params))
-            .await?
-            .try_collect()
-            .await
-    }
-
-    /// The maximally flexible version of [`query`].
-    ///
-    /// A statement may contain parameters, specified by `$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, I>(
-        &self,
-        statement: Statement,
-        params: I,
-    ) -> Result<RowStream, Error>
-    where
-        I: IntoIterator<Item = &'a (dyn ToSql + Sync)>,
-        I::IntoIter: ExactSizeIterator,
-    {
-        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>
@@ -284,14 +195,10 @@ impl Client {
         simple_query::batch_execute(self.inner(), query).await
     }
 
-    pub async fn discard_all(&self) -> Result<ReadyForQueryStatus, Error> {
+    pub async fn discard_all(&mut self) -> Result<ReadyForQueryStatus, Error> {
         // clear the prepared statements that are about to be nuked from the postgres session
-        {
-            let mut typeinfo = self.inner.cached_typeinfo.lock();
-            typeinfo.typeinfo = None;
-            typeinfo.typeinfo_composite = None;
-            typeinfo.typeinfo_enum = None;
-        }
+
+        self.cached_typeinfo.typeinfo = None;
 
         self.batch_execute("discard all").await
     }
@@ -359,8 +266,8 @@ 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
+    pub(crate) async fn get_type_inner(&mut self, oid: Oid) -> Result<Type, Error> {
+        crate::prepare::get_type(&self.inner, &mut self.cached_typeinfo, oid).await
     }
 
     /// Determines if the connection to the server has already closed.

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

@@ -22,7 +22,7 @@ pub trait GenericClient: private::Sealed {
         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 {}
@@ -38,8 +38,8 @@ impl GenericClient for Client {
     }
 
     /// Query for type information
-    async fn get_type(&self, oid: Oid) -> Result<Type, Error> {
-        crate::prepare::get_type(self.inner(), oid).await
+    async fn get_type(&mut self, oid: Oid) -> Result<Type, Error> {
+        self.get_type_inner(oid).await
     }
 }
 
@@ -56,7 +56,7 @@ impl GenericClient for Transaction<'_> {
     }
 
     /// Query for type information
-    async fn get_type(&self, oid: Oid) -> Result<Type, Error> {
-        self.client().get_type(oid).await
+    async fn get_type(&mut self, oid: Oid) -> Result<Type, Error> {
+        self.client_mut().get_type(oid).await
     }
 }

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

@@ -9,10 +9,10 @@ use log::debug;
 use postgres_protocol2::message::backend::Message;
 use postgres_protocol2::message::frontend;
 
-use crate::client::InnerClient;
+use crate::client::{CachedTypeInfo, InnerClient};
 use crate::codec::FrontendMessage;
 use crate::connection::RequestMessages;
-use crate::types::{Field, Kind, Oid, Type};
+use crate::types::{Kind, Oid, Type};
 use crate::{Column, Error, Statement, query, slice_iter};
 
 pub(crate) const TYPEINFO_QUERY: &str = "\
@@ -23,23 +23,7 @@ INNER JOIN pg_catalog.pg_namespace n ON t.typnamespace = n.oid
 WHERE t.oid = $1
 ";
 
-const TYPEINFO_ENUM_QUERY: &str = "\
-SELECT enumlabel
-FROM pg_catalog.pg_enum
-WHERE enumtypid = $1
-ORDER BY enumsortorder
-";
-
-pub(crate) const TYPEINFO_COMPOSITE_QUERY: &str = "\
-SELECT attname, atttypid
-FROM pg_catalog.pg_attribute
-WHERE attrelid = $1
-AND NOT attisdropped
-AND attnum > 0
-ORDER BY attnum
-";
-
-pub async fn prepare(
+async fn prepare_typecheck(
     client: &Arc<InnerClient>,
     name: &'static str,
     query: &str,
@@ -67,7 +51,7 @@ 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_ = Type::from_oid(oid).ok_or_else(Error::unexpected_message)?;
         parameters.push(type_);
     }
 
@@ -75,7 +59,7 @@ 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_ = Type::from_oid(field.type_oid()).ok_or_else(Error::unexpected_message)?;
             let column = Column::new(field.name().to_string(), type_, field);
             columns.push(column);
         }
@@ -84,15 +68,6 @@ pub async fn prepare(
     Ok(Statement::new(client, name, parameters, columns))
 }
 
-fn prepare_rec<'a>(
-    client: &'a Arc<InnerClient>,
-    name: &'static str,
-    query: &'a str,
-    types: &'a [Type],
-) -> Pin<Box<dyn Future<Output = Result<Statement, Error>> + 'a + Send>> {
-    Box::pin(prepare(client, name, query, types))
-}
-
 fn encode(client: &InnerClient, name: &str, query: &str, types: &[Type]) -> Result<Bytes, Error> {
     if types.is_empty() {
         debug!("preparing query {}: {}", name, query);
@@ -108,16 +83,20 @@ 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: &Arc<InnerClient>,
+    typecache: &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) {
-        return Ok(type_);
-    }
+    if let Some(type_) = typecache.types.get(&oid) {
+        return Ok(type_.clone());
+    };
 
-    let stmt = typeinfo_statement(client).await?;
+    let stmt = typeinfo_statement(client, typecache).await?;
 
     let rows = query::query(client, stmt, slice_iter(&[&oid])).await?;
     pin_mut!(rows);
@@ -136,100 +115,48 @@ pub async fn get_type(client: &Arc<InnerClient>, oid: Oid) -> Result<Type, Error
     let relid: Oid = row.try_get(6)?;
 
     let kind = if type_ == b'e' as i8 {
-        let variants = get_enum_variants(client, oid).await?;
-        Kind::Enum(variants)
+        Kind::Enum
     } else if type_ == b'p' as i8 {
         Kind::Pseudo
     } else if basetype != 0 {
-        let type_ = get_type_rec(client, basetype).await?;
-        Kind::Domain(type_)
+        Kind::Domain(basetype)
     } else if elem_oid != 0 {
-        let type_ = get_type_rec(client, elem_oid).await?;
+        let type_ = get_type_rec(client, typecache, elem_oid).await?;
         Kind::Array(type_)
     } else if relid != 0 {
-        let fields = get_composite_fields(client, relid).await?;
-        Kind::Composite(fields)
+        Kind::Composite(relid)
     } else if let Some(rngsubtype) = rngsubtype {
-        let type_ = get_type_rec(client, rngsubtype).await?;
+        let type_ = get_type_rec(client, typecache, rngsubtype).await?;
         Kind::Range(type_)
     } else {
         Kind::Simple
     };
 
     let type_ = Type::new(name, oid, kind, schema);
-    client.set_type(oid, &type_);
+    typecache.types.insert(oid, type_.clone());
 
     Ok(type_)
 }
 
 fn get_type_rec<'a>(
     client: &'a Arc<InnerClient>,
+    typecache: &'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, typecache, oid))
 }
 
-async fn typeinfo_statement(client: &Arc<InnerClient>) -> Result<Statement, Error> {
-    if let Some(stmt) = client.typeinfo() {
-        return Ok(stmt);
+async fn typeinfo_statement(
+    client: &Arc<InnerClient>,
+    typecache: &mut CachedTypeInfo,
+) -> Result<Statement, Error> {
+    if let Some(stmt) = &typecache.typeinfo {
+        return Ok(stmt.clone());
     }
 
     let typeinfo = "neon_proxy_typeinfo";
-    let stmt = prepare_rec(client, typeinfo, TYPEINFO_QUERY, &[]).await?;
+    let stmt = prepare_typecheck(client, typeinfo, TYPEINFO_QUERY, &[]).await?;
 
-    client.set_typeinfo(&stmt);
-    Ok(stmt)
-}
-
-async fn get_enum_variants(client: &Arc<InnerClient>, oid: Oid) -> Result<Vec<String>, Error> {
-    let stmt = typeinfo_enum_statement(client).await?;
-
-    query::query(client, stmt, slice_iter(&[&oid]))
-        .await?
-        .and_then(|row| async move { row.try_get(0) })
-        .try_collect()
-        .await
-}
-
-async fn typeinfo_enum_statement(client: &Arc<InnerClient>) -> Result<Statement, Error> {
-    if let Some(stmt) = client.typeinfo_enum() {
-        return Ok(stmt);
-    }
-
-    let typeinfo = "neon_proxy_typeinfo_enum";
-    let stmt = prepare_rec(client, typeinfo, TYPEINFO_ENUM_QUERY, &[]).await?;
-
-    client.set_typeinfo_enum(&stmt);
-    Ok(stmt)
-}
-
-async fn get_composite_fields(client: &Arc<InnerClient>, oid: Oid) -> Result<Vec<Field>, Error> {
-    let stmt = typeinfo_composite_statement(client).await?;
-
-    let rows = query::query(client, stmt, slice_iter(&[&oid]))
-        .await?
-        .try_collect::<Vec<_>>()
-        .await?;
-
-    let mut fields = vec![];
-    for row in rows {
-        let name = row.try_get(0)?;
-        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> {
-    if let Some(stmt) = client.typeinfo_composite() {
-        return Ok(stmt);
-    }
-
-    let typeinfo = "neon_proxy_typeinfo_composite";
-    let stmt = prepare_rec(client, typeinfo, TYPEINFO_COMPOSITE_QUERY, &[]).await?;
-
-    client.set_typeinfo_composite(&stmt);
+    typecache.typeinfo = Some(stmt.clone());
     Ok(stmt)
 }

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

@@ -72,4 +72,9 @@ impl<'a> Transaction<'a> {
     pub fn client(&self) -> &Client {
         self.client
     }
+
+    /// Returns a reference to the underlying `Client`.
+    pub fn client_mut(&mut self) -> &mut Client {
+        self.client
+    }
 }

libs/tracing-utils/src/lib.rs

@@ -21,7 +21,7 @@
 //!         .with_writer(std::io::stderr);
 //!
 //!     // Initialize OpenTelemetry. Exports tracing spans as OpenTelemetry traces
-//!     let otlp_layer = tracing_utils::init_tracing("my_application").await;
+//!     let otlp_layer = tracing_utils::init_tracing("my_application", tracing_utils::ExportConfig::default()).await;
 //!
 //!     // Put it all together
 //!     tracing_subscriber::registry()
@@ -38,8 +38,12 @@ pub mod http;
 
 use opentelemetry::KeyValue;
 use opentelemetry::trace::TracerProvider;
-use tracing::Subscriber;
+use opentelemetry_otlp::WithExportConfig;
+pub use opentelemetry_otlp::{ExportConfig, Protocol};
+use tracing::level_filters::LevelFilter;
+use tracing::{Dispatch, Subscriber};
 use tracing_subscriber::Layer;
+use tracing_subscriber::layer::SubscriberExt;
 use tracing_subscriber::registry::LookupSpan;
 
 /// Set up OpenTelemetry exporter, using configuration from environment variables.
@@ -69,19 +73,28 @@ use tracing_subscriber::registry::LookupSpan;
 ///
 /// This doesn't block, but is marked as 'async' to hint that this must be called in
 /// asynchronous execution context.
-pub async fn init_tracing<S>(service_name: &str) -> Option<impl Layer<S>>
+pub async fn init_tracing<S>(
+    service_name: &str,
+    export_config: ExportConfig,
+) -> Option<impl Layer<S>>
 where
     S: Subscriber + for<'span> LookupSpan<'span>,
 {
     if std::env::var("OTEL_SDK_DISABLED") == Ok("true".to_string()) {
         return None;
     };
-    Some(init_tracing_internal(service_name.to_string()))
+    Some(init_tracing_internal(
+        service_name.to_string(),
+        export_config,
+    ))
 }
 
 /// Like `init_tracing`, but creates a separate tokio Runtime for the tracing
 /// tasks.
-pub fn init_tracing_without_runtime<S>(service_name: &str) -> Option<impl Layer<S>>
+pub fn init_tracing_without_runtime<S>(
+    service_name: &str,
+    export_config: ExportConfig,
+) -> Option<impl Layer<S>>
 where
     S: Subscriber + for<'span> LookupSpan<'span>,
 {
@@ -112,16 +125,22 @@ where
     ));
     let _guard = runtime.enter();
 
-    Some(init_tracing_internal(service_name.to_string()))
+    Some(init_tracing_internal(
+        service_name.to_string(),
+        export_config,
+    ))
 }
 
-fn init_tracing_internal<S>(service_name: String) -> impl Layer<S>
+fn init_tracing_internal<S>(service_name: String, export_config: ExportConfig) -> impl Layer<S>
 where
     S: Subscriber + for<'span> LookupSpan<'span>,
 {
-    // Sets up exporter from the OTEL_EXPORTER_* environment variables.
+    // Sets up exporter from the provided [`ExportConfig`] parameter.
+    // If the endpoint is not specified, it is loaded from the
+    // OTEL_EXPORTER_OTLP_ENDPOINT environment variable.
     let exporter = opentelemetry_otlp::SpanExporter::builder()
         .with_http()
+        .with_export_config(export_config)
         .build()
         .expect("could not initialize opentelemetry exporter");
 
@@ -151,3 +170,51 @@ where
 pub fn shutdown_tracing() {
     opentelemetry::global::shutdown_tracer_provider();
 }
+
+pub enum OtelEnablement {
+    Disabled,
+    Enabled {
+        service_name: String,
+        export_config: ExportConfig,
+        runtime: &'static tokio::runtime::Runtime,
+    },
+}
+
+pub struct OtelGuard {
+    pub dispatch: Dispatch,
+}
+
+impl Drop for OtelGuard {
+    fn drop(&mut self) {
+        shutdown_tracing();
+    }
+}
+
+/// Initializes OTEL infrastructure for performance tracing according to the provided configuration
+///
+/// Performance tracing is handled by a different [`tracing::Subscriber`]. This functions returns
+/// an [`OtelGuard`] containing a [`tracing::Dispatch`] associated with a newly created subscriber.
+/// Applications should use this dispatch for their performance traces.
+///
+/// The lifetime of the guard should match taht of the application. On drop, it tears down the
+/// OTEL infra.
+pub fn init_performance_tracing(otel_enablement: OtelEnablement) -> Option<OtelGuard> {
+    let otel_subscriber = match otel_enablement {
+        OtelEnablement::Disabled => None,
+        OtelEnablement::Enabled {
+            service_name,
+            export_config,
+            runtime,
+        } => {
+            let otel_layer = runtime
+                .block_on(init_tracing(&service_name, export_config))
+                .with_filter(LevelFilter::INFO);
+            let otel_subscriber = tracing_subscriber::registry().with(otel_layer);
+            let otel_dispatch = Dispatch::new(otel_subscriber);
+
+            Some(otel_dispatch)
+        }
+    };
+
+    otel_subscriber.map(|dispatch| OtelGuard { dispatch })
+}

libs/utils/Cargo.toml

@@ -42,6 +42,7 @@ toml_edit = { workspace = true, features = ["serde"] }
 tracing.workspace = true
 tracing-error.workspace = true
 tracing-subscriber = { workspace = true, features = ["json", "registry"] }
+tracing-utils.workspace = true
 rand.workspace = true
 scopeguard.workspace = true
 strum.workspace = true

libs/utils/src/logging.rs

@@ -165,6 +165,7 @@ pub fn init(
         };
         log_layer.with_filter(rust_log_env_filter())
     });
+
     let r = r.with(
         TracingEventCountLayer(&TRACING_EVENT_COUNT_METRIC).with_filter(rust_log_env_filter()),
     );

pageserver/Cargo.toml

@@ -70,6 +70,7 @@ tokio-stream.workspace = true
 tokio-util.workspace = true
 toml_edit = { workspace = true, features = [ "serde" ] }
 tracing.workspace = true
+tracing-utils.workspace = true
 url.workspace = true
 walkdir.workspace = true
 metrics.workspace = true

pageserver/compaction/tests/tests.rs

@@ -12,7 +12,7 @@ pub(crate) fn setup_logging() {
             logging::TracingErrorLayerEnablement::EnableWithRustLogFilter,
             logging::Output::Stdout,
         )
-        .expect("Failed to init test logging")
+        .expect("Failed to init test logging");
     });
 }
 

pageserver/src/bin/pageserver.rs

@@ -111,6 +111,7 @@ fn main() -> anyhow::Result<()> {
     } else {
         TracingErrorLayerEnablement::Disabled
     };
+
     logging::init(
         conf.log_format,
         tracing_error_layer_enablement,

pageserver/src/http/openapi_spec.yml

@@ -1079,7 +1079,6 @@ components:
         - last_record_lsn
         - disk_consistent_lsn
         - state
-        - latest_gc_cutoff_lsn
       properties:
         timeline_id:
           type: string
@@ -1123,9 +1122,6 @@ components:
         min_readable_lsn:
           type: string
           format: hex
-        latest_gc_cutoff_lsn:
-          type: string
-          format: hex
         applied_gc_cutoff_lsn:
           type: string
           format: hex

pageserver/src/http/routes.rs

@@ -72,6 +72,7 @@ use crate::tenant::remote_timeline_client::{
 use crate::tenant::secondary::SecondaryController;
 use crate::tenant::size::ModelInputs;
 use crate::tenant::storage_layer::{IoConcurrency, LayerAccessStatsReset, LayerName};
+use crate::tenant::timeline::detach_ancestor::DetachBehavior;
 use crate::tenant::timeline::offload::{OffloadError, offload_timeline};
 use crate::tenant::timeline::{
     CompactFlags, CompactOptions, CompactRequest, CompactionError, Timeline, WaitLsnTimeout,
@@ -460,10 +461,7 @@ async fn build_timeline_info_common(
         initdb_lsn,
         last_record_lsn,
         prev_record_lsn: Some(timeline.get_prev_record_lsn()),
-        // Externally, expose the lowest LSN that can be used to create a branch as the "GC cutoff", although internally
-        // we distinguish between the "planned" GC cutoff (PITR point) and the "latest" GC cutoff (where we
-        // actually trimmed data to), which can pass each other when PITR is changed.
-        latest_gc_cutoff_lsn: min_readable_lsn,
+        _unused: Default::default(), // Unused, for legacy decode only
         min_readable_lsn,
         applied_gc_cutoff_lsn: *timeline.get_applied_gc_cutoff_lsn(),
         current_logical_size: current_logical_size.size_dont_care_about_accuracy(),
@@ -2508,6 +2506,8 @@ async fn timeline_detach_ancestor_handler(
     let tenant_shard_id: TenantShardId = parse_request_param(&request, "tenant_shard_id")?;
     check_permission(&request, Some(tenant_shard_id.tenant_id))?;
     let timeline_id: TimelineId = parse_request_param(&request, "timeline_id")?;
+    let behavior: Option<DetachBehavior> = parse_query_param(&request, "detach_behavior")?;
+    let behavior = behavior.unwrap_or_default();
 
     let span = tracing::info_span!("detach_ancestor", tenant_id=%tenant_shard_id.tenant_id, shard_id=%tenant_shard_id.shard_slug(), %timeline_id);
 
@@ -2557,7 +2557,7 @@ async fn timeline_detach_ancestor_handler(
         let ctx = &ctx.with_scope_timeline(&timeline);
 
         let progress = timeline
-            .prepare_to_detach_from_ancestor(&tenant, options, ctx)
+            .prepare_to_detach_from_ancestor(&tenant, options, behavior, ctx)
             .await?;
 
         // uncomment to allow early as possible Tenant::drop
@@ -2572,6 +2572,7 @@ async fn timeline_detach_ancestor_handler(
                         tenant_shard_id,
                         timeline_id,
                         prepared,
+                        behavior,
                         attempt,
                         ctx,
                     )

pageserver/src/tenant.rs

@@ -5754,7 +5754,7 @@ pub(crate) mod harness {
                 logging::TracingErrorLayerEnablement::EnableWithRustLogFilter,
                 logging::Output::Stdout,
             )
-            .expect("Failed to init test logging")
+            .expect("Failed to init test logging");
         });
     }
 

pageserver/src/tenant/config.rs

@@ -219,7 +219,11 @@ impl LocationConf {
         };
 
         let shard = if conf.shard_count == 0 {
-            ShardIdentity::unsharded()
+            // NB: carry over the persisted stripe size instead of using the default. This doesn't
+            // matter for most practical purposes, since unsharded tenants don't use the stripe
+            // size, but can cause inconsistencies between storcon and Pageserver and cause manual
+            // splits without `new_stripe_size` to use an unintended stripe size.
+            ShardIdentity::unsharded_with_stripe_size(ShardStripeSize(conf.shard_stripe_size))
         } else {
             ShardIdentity::new(
                 ShardNumber(conf.shard_number),

pageserver/src/tenant/ephemeral_file.rs

@@ -9,7 +9,7 @@ use camino::Utf8PathBuf;
 use num_traits::Num;
 use pageserver_api::shard::TenantShardId;
 use tokio_epoll_uring::{BoundedBuf, Slice};
-use tracing::error;
+use tracing::{error, info_span};
 use utils::id::TimelineId;
 
 use crate::assert_u64_eq_usize::{U64IsUsize, UsizeIsU64};
@@ -76,6 +76,7 @@ impl EphemeralFile {
                 || IoBufferMut::with_capacity(TAIL_SZ),
                 gate.enter()?,
                 ctx,
+                info_span!(parent: None, "ephemeral_file_buffered_writer", tenant_id=%tenant_shard_id.tenant_id, shard_id=%tenant_shard_id.shard_slug(), timeline_id=%timeline_id, path = %filename),
             ),
             _gate_guard: gate.enter()?,
         })

pageserver/src/tenant/metadata.rs

@@ -300,9 +300,8 @@ impl TimelineMetadata {
 
     /// Returns true if anything was changed
     pub fn detach_from_ancestor(&mut self, branchpoint: &(TimelineId, Lsn)) {
-        if let Some(ancestor) = self.body.ancestor_timeline {
-            assert_eq!(ancestor, branchpoint.0);
-        }
+        // Detaching from ancestor now doesn't always detach directly to the direct ancestor, but we
+        // ensure the LSN is the same. So we don't check the timeline ID.
         if self.body.ancestor_lsn != Lsn(0) {
             assert_eq!(self.body.ancestor_lsn, branchpoint.1);
         }

pageserver/src/tenant/mgr.rs

@@ -1914,6 +1914,7 @@ impl TenantManager {
         tenant_shard_id: TenantShardId,
         timeline_id: TimelineId,
         prepared: PreparedTimelineDetach,
+        behavior: detach_ancestor::DetachBehavior,
         mut attempt: detach_ancestor::Attempt,
         ctx: &RequestContext,
     ) -> Result<HashSet<TimelineId>, detach_ancestor::Error> {
@@ -1957,7 +1958,14 @@ impl TenantManager {
             .map_err(Error::NotFound)?;
 
         let resp = timeline
-            .detach_from_ancestor_and_reparent(&tenant, prepared, ctx)
+            .detach_from_ancestor_and_reparent(
+                &tenant,
+                prepared,
+                attempt.ancestor_timeline_id,
+                attempt.ancestor_lsn,
+                behavior,
+                ctx,
+            )
             .await?;
 
         let mut slot_guard = slot_guard;

pageserver/src/tenant/remote_timeline_client/download.rs

@@ -229,6 +229,7 @@ async fn download_object(
                     || IoBufferMut::with_capacity(super::BUFFER_SIZE),
                     gate.enter().map_err(|_| DownloadError::Cancelled)?,
                     ctx,
+                    tracing::info_span!(parent: None, "download_object_buffered_writer", %dst_path),
                 );
 
                 // TODO: use vectored write (writev) once supported by tokio-epoll-uring.

pageserver/src/tenant/timeline.rs

@@ -5388,9 +5388,10 @@ impl Timeline {
         self: &Arc<Timeline>,
         tenant: &crate::tenant::Tenant,
         options: detach_ancestor::Options,
+        behavior: detach_ancestor::DetachBehavior,
         ctx: &RequestContext,
     ) -> Result<detach_ancestor::Progress, detach_ancestor::Error> {
-        detach_ancestor::prepare(self, tenant, options, ctx).await
+        detach_ancestor::prepare(self, tenant, behavior, options, ctx).await
     }
 
     /// Second step of detach from ancestor; detaches the `self` from it's current ancestor and
@@ -5406,9 +5407,21 @@ impl Timeline {
         self: &Arc<Timeline>,
         tenant: &crate::tenant::Tenant,
         prepared: detach_ancestor::PreparedTimelineDetach,
+        ancestor_timeline_id: TimelineId,
+        ancestor_lsn: Lsn,
+        behavior: detach_ancestor::DetachBehavior,
         ctx: &RequestContext,
     ) -> Result<detach_ancestor::DetachingAndReparenting, detach_ancestor::Error> {
-        detach_ancestor::detach_and_reparent(self, tenant, prepared, ctx).await
+        detach_ancestor::detach_and_reparent(
+            self,
+            tenant,
+            prepared,
+            ancestor_timeline_id,
+            ancestor_lsn,
+            behavior,
+            ctx,
+        )
+        .await
     }
 
     /// Final step which unblocks the GC.

pageserver/src/tenant/timeline/detach_ancestor.rs

@@ -32,6 +32,9 @@ pub(crate) enum Error {
     #[error("too many ancestors")]
     TooManyAncestors,
 
+    #[error("ancestor is not empty")]
+    AncestorNotEmpty,
+
     #[error("shutting down, please retry later")]
     ShuttingDown,
 
@@ -89,7 +92,9 @@ impl From<Error> for ApiError {
     fn from(value: Error) -> Self {
         match value {
             Error::NoAncestor => ApiError::Conflict(value.to_string()),
-            Error::TooManyAncestors => ApiError::BadRequest(anyhow::anyhow!("{value}")),
+            Error::TooManyAncestors | Error::AncestorNotEmpty => {
+                ApiError::BadRequest(anyhow::anyhow!("{value}"))
+            }
             Error::ShuttingDown => ApiError::ShuttingDown,
             Error::Archived(_) => ApiError::BadRequest(anyhow::anyhow!("{value}")),
             Error::OtherTimelineDetachOngoing(_) | Error::FailedToReparentAll => {
@@ -127,13 +132,37 @@ pub(crate) struct PreparedTimelineDetach {
     layers: Vec<Layer>,
 }
 
-/// TODO: this should be part of PageserverConf because we cannot easily modify cplane arguments.
+// TODO: this should be part of PageserverConf because we cannot easily modify cplane arguments.
 #[derive(Debug)]
 pub(crate) struct Options {
     pub(crate) rewrite_concurrency: std::num::NonZeroUsize,
     pub(crate) copy_concurrency: std::num::NonZeroUsize,
 }
 
+/// Controls the detach ancestor behavior.
+/// - When set to `NoAncestorAndReparent`, we will only detach a branch if its ancestor is a root branch. It will automatically reparent any children of the ancestor before and at the branch point.
+/// - When set to `MultiLevelAndNoReparent`, we will detach a branch from multiple levels of ancestors, and no reparenting will happen at all.
+#[derive(Debug, Clone, Copy, Default)]
+pub enum DetachBehavior {
+    #[default]
+    NoAncestorAndReparent,
+    MultiLevelAndNoReparent,
+}
+
+impl std::str::FromStr for DetachBehavior {
+    type Err = &'static str;
+
+    fn from_str(s: &str) -> Result<Self, Self::Err> {
+        match s {
+            "no_ancestor_and_reparent" => Ok(DetachBehavior::NoAncestorAndReparent),
+            "multi_level_and_no_reparent" => Ok(DetachBehavior::MultiLevelAndNoReparent),
+            "v1" => Ok(DetachBehavior::NoAncestorAndReparent),
+            "v2" => Ok(DetachBehavior::MultiLevelAndNoReparent),
+            _ => Err("cannot parse detach behavior"),
+        }
+    }
+}
+
 impl Default for Options {
     fn default() -> Self {
         Self {
@@ -147,7 +176,8 @@ impl Default for Options {
 #[derive(Debug)]
 pub(crate) struct Attempt {
     pub(crate) timeline_id: TimelineId,
-
+    pub(crate) ancestor_timeline_id: TimelineId,
+    pub(crate) ancestor_lsn: Lsn,
     _guard: completion::Completion,
     gate_entered: Option<utils::sync::gate::GateGuard>,
 }
@@ -167,25 +197,30 @@ impl Attempt {
 pub(super) async fn prepare(
     detached: &Arc<Timeline>,
     tenant: &Tenant,
+    behavior: DetachBehavior,
     options: Options,
     ctx: &RequestContext,
 ) -> Result<Progress, Error> {
     use Error::*;
 
-    let Some((ancestor, ancestor_lsn)) = detached
+    let Some((mut ancestor, mut ancestor_lsn)) = detached
         .ancestor_timeline
         .as_ref()
         .map(|tl| (tl.clone(), detached.ancestor_lsn))
     else {
+        let ancestor_id;
+        let ancestor_lsn;
         let still_in_progress = {
             let accessor = detached.remote_client.initialized_upload_queue()?;
 
             // we are safe to inspect the latest uploaded, because we can only witness this after
             // restart is complete and ancestor is no more.
             let latest = accessor.latest_uploaded_index_part();
-            if latest.lineage.detached_previous_ancestor().is_none() {
+            let Some((id, lsn)) = latest.lineage.detached_previous_ancestor() else {
                 return Err(NoAncestor);
             };
+            ancestor_id = id;
+            ancestor_lsn = lsn;
 
             latest
                 .gc_blocking
@@ -196,7 +231,8 @@ pub(super) async fn prepare(
         if still_in_progress {
             // gc is still blocked, we can still reparent and complete.
             // we are safe to reparent remaining, because they were locked in in the beginning.
-            let attempt = continue_with_blocked_gc(detached, tenant).await?;
+            let attempt =
+                continue_with_blocked_gc(detached, tenant, ancestor_id, ancestor_lsn).await?;
 
             // because the ancestor of detached is already set to none, we have published all
             // of the layers, so we are still "prepared."
@@ -224,13 +260,34 @@ pub(super) async fn prepare(
 
     check_no_archived_children_of_ancestor(tenant, detached, &ancestor, ancestor_lsn)?;
 
-    if ancestor.ancestor_timeline.is_some() {
+    if let DetachBehavior::MultiLevelAndNoReparent = behavior {
+        // If the ancestor has an ancestor, we might be able to fast-path detach it if the current ancestor does not have any data written/used by the detaching timeline.
+        while let Some(ancestor_of_ancestor) = ancestor.ancestor_timeline.clone() {
+            if ancestor_lsn != ancestor.ancestor_lsn {
+                // non-technical requirement; we could flatten still if ancestor LSN does not match but that needs
+                // us to copy and cut more layers.
+                return Err(AncestorNotEmpty);
+            }
+            // Use the ancestor of the ancestor as the new ancestor (only when the ancestor LSNs are the same)
+            ancestor_lsn = ancestor.ancestor_lsn; // Get the LSN first before resetting the `ancestor` variable
+            ancestor = ancestor_of_ancestor;
+            // TODO: do we still need to check if we don't want to reparent?
+            check_no_archived_children_of_ancestor(tenant, detached, &ancestor, ancestor_lsn)?;
+        }
+    } else if ancestor.ancestor_timeline.is_some() {
         // non-technical requirement; we could flatten N ancestors just as easily but we chose
         // not to, at least initially
         return Err(TooManyAncestors);
     }
 
-    let attempt = start_new_attempt(detached, tenant).await?;
+    tracing::info!(
+        "attempt to detach the timeline from the ancestor: {}@{}, behavior={:?}",
+        ancestor.timeline_id,
+        ancestor_lsn,
+        behavior
+    );
+
+    let attempt = start_new_attempt(detached, tenant, ancestor.timeline_id, ancestor_lsn).await?;
 
     utils::pausable_failpoint!("timeline-detach-ancestor::before_starting_after_locking-pausable");
 
@@ -450,8 +507,13 @@ pub(super) async fn prepare(
     Ok(Progress::Prepared(attempt, prepared))
 }
 
-async fn start_new_attempt(detached: &Timeline, tenant: &Tenant) -> Result<Attempt, Error> {
-    let attempt = obtain_exclusive_attempt(detached, tenant)?;
+async fn start_new_attempt(
+    detached: &Timeline,
+    tenant: &Tenant,
+    ancestor_timeline_id: TimelineId,
+    ancestor_lsn: Lsn,
+) -> Result<Attempt, Error> {
+    let attempt = obtain_exclusive_attempt(detached, tenant, ancestor_timeline_id, ancestor_lsn)?;
 
     // insert the block in the index_part.json, if not already there.
     let _dont_care = tenant
@@ -466,13 +528,23 @@ async fn start_new_attempt(detached: &Timeline, tenant: &Tenant) -> Result<Attem
     Ok(attempt)
 }
 
-async fn continue_with_blocked_gc(detached: &Timeline, tenant: &Tenant) -> Result<Attempt, Error> {
+async fn continue_with_blocked_gc(
+    detached: &Timeline,
+    tenant: &Tenant,
+    ancestor_timeline_id: TimelineId,
+    ancestor_lsn: Lsn,
+) -> Result<Attempt, Error> {
     // FIXME: it would be nice to confirm that there is an in-memory version, since we've just
     // verified there is a persistent one?
-    obtain_exclusive_attempt(detached, tenant)
+    obtain_exclusive_attempt(detached, tenant, ancestor_timeline_id, ancestor_lsn)
 }
 
-fn obtain_exclusive_attempt(detached: &Timeline, tenant: &Tenant) -> Result<Attempt, Error> {
+fn obtain_exclusive_attempt(
+    detached: &Timeline,
+    tenant: &Tenant,
+    ancestor_timeline_id: TimelineId,
+    ancestor_lsn: Lsn,
+) -> Result<Attempt, Error> {
     use Error::{OtherTimelineDetachOngoing, ShuttingDown};
 
     // ensure we are the only active attempt for this tenant
@@ -493,6 +565,8 @@ fn obtain_exclusive_attempt(detached: &Timeline, tenant: &Tenant) -> Result<Atte
 
     Ok(Attempt {
         timeline_id: detached.timeline_id,
+        ancestor_timeline_id,
+        ancestor_lsn,
         _guard: guard,
         gate_entered: Some(_gate_entered),
     })
@@ -795,6 +869,9 @@ pub(super) async fn detach_and_reparent(
     detached: &Arc<Timeline>,
     tenant: &Tenant,
     prepared: PreparedTimelineDetach,
+    ancestor_timeline_id: TimelineId,
+    ancestor_lsn: Lsn,
+    behavior: DetachBehavior,
     _ctx: &RequestContext,
 ) -> Result<DetachingAndReparenting, Error> {
     let PreparedTimelineDetach { layers } = prepared;
@@ -822,7 +899,30 @@ pub(super) async fn detach_and_reparent(
         "cannot (detach? reparent)? complete if the operation is not still ongoing"
     );
 
-    let ancestor = match (detached.ancestor_timeline.as_ref(), recorded_branchpoint) {
+    let ancestor_to_detach = match detached.ancestor_timeline.as_ref() {
+        Some(mut ancestor) => {
+            while ancestor.timeline_id != ancestor_timeline_id {
+                match ancestor.ancestor_timeline.as_ref() {
+                    Some(found) => {
+                        if ancestor_lsn != ancestor.ancestor_lsn {
+                            return Err(Error::DetachReparent(anyhow::anyhow!(
+                                "cannot find the ancestor timeline to detach from: wrong ancestor lsn"
+                            )));
+                        }
+                        ancestor = found;
+                    }
+                    None => {
+                        return Err(Error::DetachReparent(anyhow::anyhow!(
+                            "cannot find the ancestor timeline to detach from"
+                        )));
+                    }
+                }
+            }
+            Some(ancestor)
+        }
+        None => None,
+    };
+    let ancestor = match (ancestor_to_detach, recorded_branchpoint) {
         (Some(ancestor), None) => {
             assert!(
                 !layers.is_empty(),
@@ -895,6 +995,11 @@ pub(super) async fn detach_and_reparent(
         Ancestor::Detached(ancestor, ancestor_lsn) => (ancestor, ancestor_lsn, false),
     };
 
+    if let DetachBehavior::MultiLevelAndNoReparent = behavior {
+        // Do not reparent if the user requests to behave so.
+        return Ok(DetachingAndReparenting::Reparented(HashSet::new()));
+    }
+
     let mut tasks = tokio::task::JoinSet::new();
 
     // Returns a single permit semaphore which will be used to make one reparenting succeed,
@@ -1032,6 +1137,11 @@ pub(super) async fn complete(
 }
 
 /// Query against a locked `Tenant::timelines`.
+///
+/// A timeline is reparentable if:
+///
+/// - It is not the timeline being detached.
+/// - It has the same ancestor as the timeline being detached. Note that the ancestor might not be the direct ancestor.
 fn reparentable_timelines<'a, I>(
     timelines: I,
     detached: &'a Arc<Timeline>,

pageserver/src/virtual_file.rs

@@ -1299,9 +1299,8 @@ impl OwnedAsyncWriter for VirtualFile {
         buf: FullSlice<Buf>,
         offset: u64,
         ctx: &RequestContext,
-    ) -> std::io::Result<FullSlice<Buf>> {
-        let (buf, res) = VirtualFile::write_all_at(self, buf, offset, ctx).await;
-        res.map(|_| buf)
+    ) -> (FullSlice<Buf>, std::io::Result<()>) {
+        VirtualFile::write_all_at(self, buf, offset, ctx).await
     }
 }
 

pageserver/src/virtual_file/owned_buffers_io/write.rs

@@ -31,7 +31,7 @@ pub trait OwnedAsyncWriter {
         buf: FullSlice<Buf>,
         offset: u64,
         ctx: &RequestContext,
-    ) -> impl std::future::Future<Output = std::io::Result<FullSlice<Buf>>> + Send;
+    ) -> impl std::future::Future<Output = (FullSlice<Buf>, std::io::Result<()>)> + Send;
 }
 
 /// A wrapper aorund an [`OwnedAsyncWriter`] that uses a [`Buffer`] to batch
@@ -66,6 +66,7 @@ where
         buf_new: impl Fn() -> B,
         gate_guard: utils::sync::gate::GateGuard,
         ctx: &RequestContext,
+        flush_task_span: tracing::Span,
     ) -> Self {
         Self {
             writer: writer.clone(),
@@ -75,6 +76,7 @@ where
                 buf_new(),
                 gate_guard,
                 ctx.attached_child(),
+                flush_task_span,
             ),
             bytes_submitted: 0,
         }
@@ -269,12 +271,12 @@ mod tests {
             buf: FullSlice<Buf>,
             offset: u64,
             _: &RequestContext,
-        ) -> std::io::Result<FullSlice<Buf>> {
+        ) -> (FullSlice<Buf>, std::io::Result<()>) {
             self.writes
                 .lock()
                 .unwrap()
                 .push((Vec::from(&buf[..]), offset));
-            Ok(buf)
+            (buf, Ok(()))
         }
     }
 
@@ -293,6 +295,7 @@ mod tests {
             || IoBufferMut::with_capacity(2),
             gate.enter()?,
             ctx,
+            tracing::Span::none(),
         );
 
         writer.write_buffered_borrowed(b"abc", ctx).await?;

pageserver/src/virtual_file/owned_buffers_io/write/flush.rs

@@ -1,9 +1,14 @@
+use std::ops::ControlFlow;
 use std::sync::Arc;
 
+use once_cell::sync::Lazy;
+use tokio_util::sync::CancellationToken;
+use tracing::{Instrument, info, info_span, warn};
 use utils::sync::duplex;
 
 use super::{Buffer, CheapCloneForRead, OwnedAsyncWriter};
 use crate::context::RequestContext;
+use crate::virtual_file::MaybeFatalIo;
 use crate::virtual_file::owned_buffers_io::io_buf_aligned::IoBufAligned;
 use crate::virtual_file::owned_buffers_io::io_buf_ext::FullSlice;
 
@@ -118,6 +123,7 @@ where
         buf: B,
         gate_guard: utils::sync::gate::GateGuard,
         ctx: RequestContext,
+        span: tracing::Span,
     ) -> Self
     where
         B: Buffer<IoBuf = Buf> + Send + 'static,
@@ -125,11 +131,14 @@ where
         // It is fine to buffer up to only 1 message. We only 1 message in-flight at a time.
         let (front, back) = duplex::mpsc::channel(1);
 
-        let join_handle = tokio::spawn(async move {
-            FlushBackgroundTask::new(back, file, gate_guard, ctx)
-                .run(buf.flush())
-                .await
-        });
+        let join_handle = tokio::spawn(
+            async move {
+                FlushBackgroundTask::new(back, file, gate_guard, ctx)
+                    .run(buf.flush())
+                    .await
+            }
+            .instrument(span),
+        );
 
         FlushHandle {
             inner: Some(FlushHandleInner {
@@ -236,6 +245,7 @@ where
     /// The passed in slice is immediately sent back to the flush handle through the duplex channel.
     async fn run(mut self, slice: FullSlice<Buf>) -> std::io::Result<Arc<W>> {
         // Sends the extra buffer back to the handle.
+        // TODO: can this ever await and or fail? I think not.
         self.channel.send(slice).await.map_err(|_| {
             std::io::Error::new(std::io::ErrorKind::BrokenPipe, "flush handle closed early")
         })?;
@@ -251,10 +261,47 @@ where
             }
 
             // Write slice to disk at `offset`.
-            let slice = self
-                .writer
-                .write_all_at(request.slice, request.offset, &self.ctx)
-                .await?;
+            //
+            // Error handling happens according to the current policy of crashing
+            // on fatal IO errors and retrying in place otherwise (deeming all other errors retryable).
+            // (The upper layers of the Pageserver write path are not equipped to retry write errors
+            //  becasuse they often deallocate the buffers that were already written).
+            //
+            // TODO: cancellation sensitiity.
+            // Without it, if we hit a bug where retrying is never successful,
+            // then we can't shut down the timeline/tenant/pageserver cleanly because
+            // layers of the Pageserver write path are holding the gate open for EphemeralFile.
+            //
+            // TODO: use utils::backoff::retry once async closures are actually usable
+            //
+            let mut slice_storage = Some(request.slice);
+            for attempt in 1.. {
+                let result = async {
+                    if attempt > 1 {
+                        info!("retrying flush");
+                    }
+                    let slice = slice_storage.take().expect(
+                        "likely previous invocation of this future didn't get polled to completion",
+                    );
+                    let (slice, res) = self.writer.write_all_at(slice, request.offset, &self.ctx).await;
+                    slice_storage = Some(slice);
+                    let res = res.maybe_fatal_err("owned_buffers_io flush");
+                    let Err(err) = res else {
+                        return ControlFlow::Break(());
+                    };
+                    warn!(%err, "error flushing buffered writer buffer to disk, retrying after backoff");
+                    static NO_CANCELLATION: Lazy<CancellationToken> = Lazy::new(CancellationToken::new);
+                    utils::backoff::exponential_backoff(attempt, 1.0, 10.0, &NO_CANCELLATION).await;
+                    ControlFlow::Continue(())
+                }
+                .instrument(info_span!("flush_attempt", %attempt))
+                .await;
+                match result {
+                    ControlFlow::Break(()) => break,
+                    ControlFlow::Continue(()) => continue,
+                }
+            }
+            let slice = slice_storage.expect("loop must have run at least once");
 
             #[cfg(test)]
             {

pgxn/neon/libpagestore.c

@@ -50,20 +50,6 @@
 #define MIN_RECONNECT_INTERVAL_USEC 1000
 #define MAX_RECONNECT_INTERVAL_USEC 1000000
 
-
-enum NeonEndpointType {
-	EP_TYPE_PRIMARY = 1,
-	EP_TYPE_REPLICA,
-	EP_TYPE_STATIC
-};
-
-static const struct config_enum_entry neon_endpoint_types[] = {
-	{"primary", EP_TYPE_PRIMARY, false},
-	{"replica", EP_TYPE_REPLICA, false},
-	{"static", EP_TYPE_STATIC, false},
-	{NULL, 0, false}
-};
-
 /* GUCs */
 char	   *neon_timeline;
 char	   *neon_tenant;
@@ -76,8 +62,6 @@ int			flush_every_n_requests = 8;
 
 int         neon_protocol_version = 2;
 
-static int neon_endpoint_type = 0;
-
 static int	max_reconnect_attempts = 60;
 static int	stripe_size;
 
@@ -408,7 +392,7 @@ pageserver_connect(shardno_t shard_no, int elevel)
 	{
 		const char *keywords[4];
 		const char *values[4];
-		char pid_str[24];
+		char pid_str[16];
 		int			n_pgsql_params;
 		TimestampTz	now;
 		int64		us_since_last_attempt;
@@ -461,22 +445,7 @@ pageserver_connect(shardno_t shard_no, int elevel)
 		 */
 		keywords[n_pgsql_params] = "application_name";
 		{
-			int ret;
-			switch (neon_endpoint_type)
-			{
-				case EP_TYPE_PRIMARY:
-					ret = snprintf(pid_str, sizeof(pid_str), "%d-%s", MyProcPid, "primary");
-					break;
-				case EP_TYPE_REPLICA:
-					ret = snprintf(pid_str, sizeof(pid_str), "%d-%s", MyProcPid, "replica");
-					break;
-				case EP_TYPE_STATIC:
-					ret = snprintf(pid_str, sizeof(pid_str), "%d-%s", MyProcPid, "static");
-					break;
-				default:
-					ret = snprintf(pid_str, sizeof(pid_str), "%d", MyProcPid);
-					break;
-			}
+			int ret = snprintf(pid_str, sizeof(pid_str), "%d", MyProcPid);
 			if (ret < 0 || ret >= (int)(sizeof(pid_str)))
 				elog(FATAL, "stack-allocated buffer too small to hold pid");
 		}
@@ -1401,17 +1370,6 @@ pg_init_libpagestore(void)
 							GUC_UNIT_MS,
 							NULL, NULL, NULL);
 
-	DefineCustomEnumVariable(
-							"neon.endpoint_type",
-							"The compute endpoint node type",
-							NULL,
-							&neon_endpoint_type,
-							EP_TYPE_PRIMARY,
-							neon_endpoint_types,
-							PGC_POSTMASTER,
-							0,
-							NULL, NULL, NULL);
-
 	relsize_hash_init();
 
 	if (page_server != NULL)

pgxn/neon/pagestore_smgr.c

@@ -76,6 +76,10 @@
 #include "access/xlogrecovery.h"
 #endif
 
+#if PG_VERSION_NUM < 160000
+typedef PGAlignedBlock PGIOAlignedBlock;
+#endif
+
 /*
  * If DEBUG_COMPARE_LOCAL is defined, we pass through all the SMGR API
  * calls to md.c, and *also* do the calls to the Page Server. On every
@@ -1803,7 +1807,7 @@ static XLogRecPtr
 log_newpage_copy(NRelFileInfo * rinfo, ForkNumber forkNum, BlockNumber blkno,
 				 Page page, bool page_std)
 {
-	PGAlignedBlock copied_buffer;
+	PGIOAlignedBlock copied_buffer;
 
 	memcpy(copied_buffer.data, page, BLCKSZ);
 	return log_newpage(rinfo, forkNum, blkno, copied_buffer.data, page_std);
@@ -1820,7 +1824,7 @@ static XLogRecPtr
 log_newpages_copy(NRelFileInfo * rinfo, ForkNumber forkNum, BlockNumber blkno,
 				  BlockNumber nblocks, Page *pages, bool page_std)
 {
-	PGAlignedBlock copied_buffer[XLR_MAX_BLOCK_ID];
+	PGIOAlignedBlock copied_buffer[XLR_MAX_BLOCK_ID];
 	BlockNumber	blknos[XLR_MAX_BLOCK_ID];
 	Page		pageptrs[XLR_MAX_BLOCK_ID];
 	int			nregistered = 0;
@@ -1858,7 +1862,7 @@ log_newpages_copy(NRelFileInfo * rinfo, ForkNumber forkNum, BlockNumber blkno,
 static bool
 PageIsEmptyHeapPage(char *buffer)
 {
-	PGAlignedBlock empty_page;
+	PGIOAlignedBlock empty_page;
 
 	PageInit((Page) empty_page.data, BLCKSZ, 0);
 
@@ -2847,7 +2851,7 @@ static void
 neon_zeroextend(SMgrRelation reln, ForkNumber forkNum, BlockNumber blocknum,
 				int nblocks, bool skipFsync)
 {
-	const PGAlignedBlock buffer = {0};
+	const PGIOAlignedBlock buffer = {0};
 	int			remblocks = nblocks;
 	XLogRecPtr	lsn = 0;
 
@@ -3389,15 +3393,16 @@ neon_read(SMgrRelation reln, ForkNumber forkNum, BlockNumber blkno, void *buffer
 	if (forkNum == MAIN_FORKNUM && IS_LOCAL_REL(reln))
 	{
 		char		pageserver_masked[BLCKSZ];
-		char		mdbuf[BLCKSZ];
-		char		mdbuf_masked[BLCKSZ];
+		PGIOAlignedBlock mdbuf;
+		PGIOAlignedBlock mdbuf_masked;
+		XLogRecPtr  request_lsn = request_lsns.request_lsn;
 
-		mdread(reln, forkNum, blkno, mdbuf);
+		mdread(reln, forkNum, blkno, mdbuf.data);
 
 		memcpy(pageserver_masked, buffer, BLCKSZ);
-		memcpy(mdbuf_masked, mdbuf, BLCKSZ);
+		memcpy(mdbuf_masked.data, mdbuf.data, BLCKSZ);
 
-		if (PageIsNew((Page) mdbuf))
+		if (PageIsNew((Page) mdbuf.data))
 		{
 			if (!PageIsNew((Page) pageserver_masked))
 			{
@@ -3416,41 +3421,41 @@ neon_read(SMgrRelation reln, ForkNumber forkNum, BlockNumber blkno, void *buffer
 				 RelFileInfoFmt(InfoFromSMgrRel(reln)),
 				 forkNum,
 				 (uint32) (request_lsn >> 32), (uint32) request_lsn,
-				 hexdump_page(mdbuf));
+				 hexdump_page(mdbuf.data));
 		}
-		else if (PageGetSpecialSize(mdbuf) == 0)
+		else if (PageGetSpecialSize(mdbuf.data) == 0)
 		{
 			/* assume heap */
-			RmgrTable[RM_HEAP_ID].rm_mask(mdbuf_masked, blkno);
+			RmgrTable[RM_HEAP_ID].rm_mask(mdbuf_masked.data, blkno);
 			RmgrTable[RM_HEAP_ID].rm_mask(pageserver_masked, blkno);
 
-			if (memcmp(mdbuf_masked, pageserver_masked, BLCKSZ) != 0)
+			if (memcmp(mdbuf_masked.data, pageserver_masked, BLCKSZ) != 0)
 			{
 				neon_log(PANIC, "heap buffers differ at blk %u in rel %u/%u/%u fork %u (request LSN %X/%08X):\n------ MD ------\n%s\n------ Page Server ------\n%s\n",
 					 blkno,
 					 RelFileInfoFmt(InfoFromSMgrRel(reln)),
 					 forkNum,
 					 (uint32) (request_lsn >> 32), (uint32) request_lsn,
-					 hexdump_page(mdbuf_masked),
+					 hexdump_page(mdbuf_masked.data),
 					 hexdump_page(pageserver_masked));
 			}
 		}
-		else if (PageGetSpecialSize(mdbuf) == MAXALIGN(sizeof(BTPageOpaqueData)))
+		else if (PageGetSpecialSize(mdbuf.data) == MAXALIGN(sizeof(BTPageOpaqueData)))
 		{
-			if (((BTPageOpaqueData *) PageGetSpecialPointer(mdbuf))->btpo_cycleid < MAX_BT_CYCLE_ID)
+			if (((BTPageOpaqueData *) PageGetSpecialPointer(mdbuf.data))->btpo_cycleid < MAX_BT_CYCLE_ID)
 			{
 				/* assume btree */
-				RmgrTable[RM_BTREE_ID].rm_mask(mdbuf_masked, blkno);
+				RmgrTable[RM_BTREE_ID].rm_mask(mdbuf_masked.data, blkno);
 				RmgrTable[RM_BTREE_ID].rm_mask(pageserver_masked, blkno);
 
-				if (memcmp(mdbuf_masked, pageserver_masked, BLCKSZ) != 0)
+				if (memcmp(mdbuf_masked.data, pageserver_masked, BLCKSZ) != 0)
 				{
 					neon_log(PANIC, "btree buffers differ at blk %u in rel %u/%u/%u fork %u (request LSN %X/%08X):\n------ MD ------\n%s\n------ Page Server ------\n%s\n",
 						 blkno,
 						 RelFileInfoFmt(InfoFromSMgrRel(reln)),
 						 forkNum,
 						 (uint32) (request_lsn >> 32), (uint32) request_lsn,
-						 hexdump_page(mdbuf_masked),
+						 hexdump_page(mdbuf_masked.data),
 						 hexdump_page(pageserver_masked));
 				}
 			}
@@ -3542,77 +3547,85 @@ neon_readv(SMgrRelation reln, ForkNumber forknum, BlockNumber blocknum,
 	prefetch_pump_state(false);
 
 #ifdef DEBUG_COMPARE_LOCAL
-	if (forkNum == MAIN_FORKNUM && IS_LOCAL_REL(reln))
+	if (forknum == MAIN_FORKNUM && IS_LOCAL_REL(reln))
 	{
 		char		pageserver_masked[BLCKSZ];
-		char		mdbuf[BLCKSZ];
-		char		mdbuf_masked[BLCKSZ];
+		PGIOAlignedBlock mdbuf;
+		PGIOAlignedBlock mdbuf_masked;
+		XLogRecPtr  request_lsn = request_lsns->request_lsn;
 
 		for (int i = 0; i < nblocks; i++)
 		{
+			BlockNumber blkno = blocknum + i;
+			if (!BITMAP_ISSET(read, i))
+				continue;
+
 #if PG_MAJORVERSION_NUM >= 17
-			mdreadv(reln, forkNum, blkno + i, &mdbuf, 1);
+			{
+				void* mdbuffers[1] = { mdbuf.data };
+				mdreadv(reln, forknum, blkno, mdbuffers, 1);
+			}
 #else
-			mdread(reln, forkNum, blkno + i, mdbuf);
+			mdread(reln, forknum, blkno, mdbuf.data);
 #endif
 
-			memcpy(pageserver_masked, buffer, BLCKSZ);
-			memcpy(mdbuf_masked, mdbuf, BLCKSZ);
+			memcpy(pageserver_masked, buffers[i], BLCKSZ);
+			memcpy(mdbuf_masked.data, mdbuf.data, BLCKSZ);
 
-			if (PageIsNew((Page) mdbuf))
+			if (PageIsNew((Page) mdbuf.data))
 			{
 				if (!PageIsNew((Page) pageserver_masked))
 				{
 					neon_log(PANIC, "page is new in MD but not in Page Server at blk %u in rel %u/%u/%u fork %u (request LSN %X/%08X):\n%s\n",
 						 blkno,
 						 RelFileInfoFmt(InfoFromSMgrRel(reln)),
-						 forkNum,
+						 forknum,
 						 (uint32) (request_lsn >> 32), (uint32) request_lsn,
-						 hexdump_page(buffer));
+						 hexdump_page(buffers[i]));
 				}
 			}
-			else if (PageIsNew((Page) buffer))
+			else if (PageIsNew((Page) buffers[i]))
 			{
 				neon_log(PANIC, "page is new in Page Server but not in MD at blk %u in rel %u/%u/%u fork %u (request LSN %X/%08X):\n%s\n",
 					 blkno,
 					 RelFileInfoFmt(InfoFromSMgrRel(reln)),
-					 forkNum,
+					 forknum,
 					 (uint32) (request_lsn >> 32), (uint32) request_lsn,
-					 hexdump_page(mdbuf));
+					 hexdump_page(mdbuf.data));
 			}
-			else if (PageGetSpecialSize(mdbuf) == 0)
+			else if (PageGetSpecialSize(mdbuf.data) == 0)
 			{
 				/* assume heap */
-				RmgrTable[RM_HEAP_ID].rm_mask(mdbuf_masked, blkno);
+				RmgrTable[RM_HEAP_ID].rm_mask(mdbuf_masked.data, blkno);
 				RmgrTable[RM_HEAP_ID].rm_mask(pageserver_masked, blkno);
 
-				if (memcmp(mdbuf_masked, pageserver_masked, BLCKSZ) != 0)
+				if (memcmp(mdbuf_masked.data, pageserver_masked, BLCKSZ) != 0)
 				{
 					neon_log(PANIC, "heap buffers differ at blk %u in rel %u/%u/%u fork %u (request LSN %X/%08X):\n------ MD ------\n%s\n------ Page Server ------\n%s\n",
 						 blkno,
 						 RelFileInfoFmt(InfoFromSMgrRel(reln)),
-						 forkNum,
+						 forknum,
 						 (uint32) (request_lsn >> 32), (uint32) request_lsn,
-						 hexdump_page(mdbuf_masked),
+						 hexdump_page(mdbuf_masked.data),
 						 hexdump_page(pageserver_masked));
 				}
 			}
-			else if (PageGetSpecialSize(mdbuf) == MAXALIGN(sizeof(BTPageOpaqueData)))
+			else if (PageGetSpecialSize(mdbuf.data) == MAXALIGN(sizeof(BTPageOpaqueData)))
 			{
-				if (((BTPageOpaqueData *) PageGetSpecialPointer(mdbuf))->btpo_cycleid < MAX_BT_CYCLE_ID)
+				if (((BTPageOpaqueData *) PageGetSpecialPointer(mdbuf.data))->btpo_cycleid < MAX_BT_CYCLE_ID)
 				{
 					/* assume btree */
-					RmgrTable[RM_BTREE_ID].rm_mask(mdbuf_masked, blkno);
+					RmgrTable[RM_BTREE_ID].rm_mask(mdbuf_masked.data, blkno);
 					RmgrTable[RM_BTREE_ID].rm_mask(pageserver_masked, blkno);
 	
-					if (memcmp(mdbuf_masked, pageserver_masked, BLCKSZ) != 0)
+					if (memcmp(mdbuf_masked.data, pageserver_masked, BLCKSZ) != 0)
 					{
 						neon_log(PANIC, "btree buffers differ at blk %u in rel %u/%u/%u fork %u (request LSN %X/%08X):\n------ MD ------\n%s\n------ Page Server ------\n%s\n",
 							 blkno,
 							 RelFileInfoFmt(InfoFromSMgrRel(reln)),
-							 forkNum,
+							 forknum,
 							 (uint32) (request_lsn >> 32), (uint32) request_lsn,
-							 hexdump_page(mdbuf_masked),
+							 hexdump_page(mdbuf_masked.data),
 							 hexdump_page(pageserver_masked));
 					}
 				}
@@ -3664,6 +3677,7 @@ neon_write(SMgrRelation reln, ForkNumber forknum, BlockNumber blocknum, const vo
 	switch (reln->smgr_relpersistence)
 	{
 		case 0:
+#ifndef DEBUG_COMPARE_LOCAL
 			/* This is a bit tricky. Check if the relation exists locally */
 			if (mdexists(reln, forknum))
 			{
@@ -3682,6 +3696,7 @@ neon_write(SMgrRelation reln, ForkNumber forknum, BlockNumber blocknum, const vo
 				 */
 				return;
 			}
+#endif
 			break;
 
 		case RELPERSISTENCE_PERMANENT:
@@ -3732,6 +3747,7 @@ neon_writev(SMgrRelation reln, ForkNumber forknum, BlockNumber blkno,
 	switch (reln->smgr_relpersistence)
 	{
 		case 0:
+#ifndef DEBUG_COMPARE_LOCAL
 			/* This is a bit tricky. Check if the relation exists locally */
 			if (mdexists(reln, forknum))
 			{
@@ -3747,6 +3763,7 @@ neon_writev(SMgrRelation reln, ForkNumber forknum, BlockNumber blkno,
 				 */
 				return;
 			}
+#endif
 			break;
 
 		case RELPERSISTENCE_PERMANENT:
@@ -3768,7 +3785,7 @@ neon_writev(SMgrRelation reln, ForkNumber forknum, BlockNumber blkno,
 
 #ifdef DEBUG_COMPARE_LOCAL
 	if (IS_LOCAL_REL(reln))
-		mdwritev(reln, forknum, blocknum, &buffer, 1, skipFsync);
+		mdwritev(reln, forknum, blkno, buffers, nblocks, skipFsync);
 #endif
 }
 

proxy/src/logging.rs

@@ -46,7 +46,8 @@ pub async fn init() -> anyhow::Result<LoggingGuard> {
                 .expect("this should be a valid filter directive"),
         );
 
-    let otlp_layer = tracing_utils::init_tracing("proxy").await;
+    let otlp_layer =
+        tracing_utils::init_tracing("proxy", tracing_utils::ExportConfig::default()).await;
 
     let json_log_layer = if logfmt == LogFormat::Json {
         Some(JsonLoggingLayer::new(

proxy/src/serverless/sql_over_http.rs

@@ -860,7 +860,13 @@ impl QueryData {
         let cancel_token = inner.cancel_token();
 
         let res = match select(
-            pin!(query_to_json(config, &*inner, self, &mut 0, parsed_headers)),
+            pin!(query_to_json(
+                config,
+                &mut *inner,
+                self,
+                &mut 0,
+                parsed_headers
+            )),
             pin!(cancel.cancelled()),
         )
         .await
@@ -944,7 +950,7 @@ impl BatchQueryData {
             builder = builder.deferrable(true);
         }
 
-        let transaction = builder
+        let mut transaction = builder
             .start()
             .await
             .inspect_err(|_| {
@@ -957,7 +963,7 @@ impl BatchQueryData {
         let json_output = match query_batch(
             config,
             cancel.child_token(),
-            &transaction,
+            &mut transaction,
             self,
             parsed_headers,
         )
@@ -1009,7 +1015,7 @@ impl BatchQueryData {
 async fn query_batch(
     config: &'static HttpConfig,
     cancel: CancellationToken,
-    transaction: &Transaction<'_>,
+    transaction: &mut Transaction<'_>,
     queries: BatchQueryData,
     parsed_headers: HttpHeaders,
 ) -> Result<String, SqlOverHttpError> {
@@ -1047,7 +1053,7 @@ async fn query_batch(
 
 async fn query_to_json<T: GenericClient>(
     config: &'static HttpConfig,
-    client: &T,
+    client: &mut T,
     data: QueryData,
     current_size: &mut usize,
     parsed_headers: HttpHeaders,

storage_controller/src/compute_hook.rs

@@ -624,7 +624,16 @@ impl ComputeHook {
             MaybeSendResult::Transmit((request, lock)) => (request, lock),
         };
 
-        let result = if let Some(notify_url) = &self.config.compute_hook_url {
+        let compute_hook_url = if let Some(control_plane_url) = &self.config.control_plane_url {
+            Some(if control_plane_url.ends_with('/') {
+                format!("{control_plane_url}notify-attach")
+            } else {
+                format!("{control_plane_url}/notify-attach")
+            })
+        } else {
+            self.config.compute_hook_url.clone()
+        };
+        let result = if let Some(notify_url) = &compute_hook_url {
             self.do_notify(notify_url, &request, cancel).await
         } else {
             self.do_notify_local(&request).await.map_err(|e| {

storage_controller/src/main.rs

@@ -71,6 +71,10 @@ struct Cli {
     #[arg(long)]
     compute_hook_url: Option<String>,
 
+    /// URL to control plane storage API prefix
+    #[arg(long)]
+    control_plane_url: Option<String>,
+
     /// URL to connect to postgres, like postgresql://localhost:1234/storage_controller
     #[arg(long)]
     database_url: Option<String>,
@@ -313,11 +317,13 @@ async fn async_main() -> anyhow::Result<()> {
                 "Insecure config!  One or more secrets is not set.  This is only permitted in `--dev` mode"
             );
         }
-        StrictMode::Strict if args.compute_hook_url.is_none() => {
-            // Production systems should always have a compute hook set, to prevent falling
+        StrictMode::Strict
+            if args.compute_hook_url.is_none() && args.control_plane_url.is_none() =>
+        {
+            // Production systems should always have a control plane URL set, to prevent falling
             // back to trying to use neon_local.
             anyhow::bail!(
-                "`--compute-hook-url` is not set: this is only permitted in `--dev` mode"
+                "neither `--compute-hook-url` nor `--control-plane-url` are set: this is only permitted in `--dev` mode"
             );
         }
         StrictMode::Strict => {
@@ -343,6 +349,7 @@ async fn async_main() -> anyhow::Result<()> {
         control_plane_jwt_token: secrets.control_plane_jwt_token,
         peer_jwt_token: secrets.peer_jwt_token,
         compute_hook_url: args.compute_hook_url,
+        control_plane_url: args.control_plane_url,
         max_offline_interval: args
             .max_offline_interval
             .map(humantime::Duration::into)

storage_controller/src/persistence.rs

@@ -1613,23 +1613,49 @@ pub(crate) struct TenantShardPersistence {
 }
 
 impl TenantShardPersistence {
+    fn get_shard_count(&self) -> Result<ShardCount, ShardConfigError> {
+        self.shard_count
+            .try_into()
+            .map(ShardCount)
+            .map_err(|_| ShardConfigError::InvalidCount)
+    }
+
+    fn get_shard_number(&self) -> Result<ShardNumber, ShardConfigError> {
+        self.shard_number
+            .try_into()
+            .map(ShardNumber)
+            .map_err(|_| ShardConfigError::InvalidNumber)
+    }
+
+    fn get_stripe_size(&self) -> Result<ShardStripeSize, ShardConfigError> {
+        self.shard_stripe_size
+            .try_into()
+            .map(ShardStripeSize)
+            .map_err(|_| ShardConfigError::InvalidStripeSize)
+    }
+
     pub(crate) fn get_shard_identity(&self) -> Result<ShardIdentity, ShardConfigError> {
         if self.shard_count == 0 {
-            Ok(ShardIdentity::unsharded())
+            // NB: carry over the stripe size from the persisted record, to avoid consistency check
+            // failures if the persisted value differs from the default stripe size. The stripe size
+            // doesn't really matter for unsharded tenants anyway.
+            Ok(ShardIdentity::unsharded_with_stripe_size(
+                self.get_stripe_size()?,
+            ))
         } else {
             Ok(ShardIdentity::new(
-                ShardNumber(self.shard_number as u8),
-                ShardCount::new(self.shard_count as u8),
-                ShardStripeSize(self.shard_stripe_size as u32),
+                self.get_shard_number()?,
+                self.get_shard_count()?,
+                self.get_stripe_size()?,
             )?)
         }
     }
 
-    pub(crate) fn get_tenant_shard_id(&self) -> Result<TenantShardId, hex::FromHexError> {
+    pub(crate) fn get_tenant_shard_id(&self) -> anyhow::Result<TenantShardId> {
         Ok(TenantShardId {
             tenant_id: TenantId::from_str(self.tenant_id.as_str())?,
-            shard_number: ShardNumber(self.shard_number as u8),
-            shard_count: ShardCount::new(self.shard_count as u8),
+            shard_number: self.get_shard_number()?,
+            shard_count: self.get_shard_count()?,
         })
     }
 }

storage_controller/src/service.rs

@@ -363,6 +363,15 @@ pub struct Config {
     /// assume it is running in a test environment and try to update neon_local.
     pub compute_hook_url: Option<String>,
 
+    /// Prefix for storage API endpoints of the control plane. We use this prefix to compute
+    /// URLs that we use to send pageserver and safekeeper attachment locations.
+    /// If this is None, the compute hook will assume it is running in a test environment
+    /// and try to invoke neon_local instead.
+    ///
+    /// For now, there is also `compute_hook_url` which allows configuration of the pageserver
+    /// specific endpoint, but it is in the process of being phased out.
+    pub control_plane_url: Option<String>,
+
     /// Grace period within which a pageserver does not respond to heartbeats, but is still
     /// considered active. Once the grace period elapses, the next heartbeat failure will
     /// mark the pagseserver offline.
@@ -1995,21 +2004,41 @@ impl Service {
         tracing::info!("Loaded {} LocationConfigs", configs.tenant_shards.len());
 
         let mut cleanup = Vec::new();
+        let mut mismatched_locations = 0;
         {
             let mut locked = self.inner.write().unwrap();
 
-            for (tenant_shard_id, observed_loc) in configs.tenant_shards {
+            for (tenant_shard_id, reported) in configs.tenant_shards {
                 let Some(tenant_shard) = locked.tenants.get_mut(&tenant_shard_id) else {
                     cleanup.push(tenant_shard_id);
                     continue;
                 };
-                tenant_shard
+
+                let on_record = &mut tenant_shard
                     .observed
                     .locations
-                    .insert(node.get_id(), ObservedStateLocation { conf: observed_loc });
+                    .entry(node.get_id())
+                    .or_insert_with(|| ObservedStateLocation { conf: None })
+                    .conf;
+
+                // If the location reported by the node does not match our observed state,
+                // then we mark it as uncertain and let the background reconciliation loop
+                // deal with it.
+                //
+                // Note that this also covers net new locations reported by the node.
+                if *on_record != reported {
+                    mismatched_locations += 1;
+                    *on_record = None;
+                }
             }
         }
 
+        if mismatched_locations > 0 {
+            tracing::info!(
+                "Set observed state to None for {mismatched_locations} mismatched locations"
+            );
+        }
+
         for tenant_shard_id in cleanup {
             tracing::info!("Detaching {tenant_shard_id}");
             match node
@@ -7865,6 +7894,9 @@ impl Service {
     /// At most one tenant will be split per call: the one with the largest max logical size. It
     /// will split 1 → 8 shards.
     ///
+    /// An unsharded tenant will get DEFAULT_STRIPE_SIZE, regardless of what its ShardIdentity says.
+    /// A sharded tenant will retain its stripe size, as splits do not allow changing it.
+    ///
     /// TODO: consider splitting based on total logical size rather than max logical size.
     ///
     /// TODO: consider spawning multiple splits in parallel: this is only called once every 20
@@ -7910,6 +7942,16 @@ impl Service {
             "Auto-splitting tenant for size threshold {split_threshold}: current size {split_candidate:?}"
         );
 
+        // Retain the stripe size of sharded tenants, as splits don't allow changing it. Otherwise,
+        // use DEFAULT_STRIPE_SIZE for unsharded tenants -- their stripe size doesn't really matter,
+        // and if we change the default stripe size we want to use the new default rather than an
+        // old, persisted stripe size.
+        let new_stripe_size = match split_candidate.id.shard_count.count() {
+            0 => panic!("invalid shard count 0"),
+            1 => Some(ShardParameters::DEFAULT_STRIPE_SIZE),
+            2.. => None,
+        };
+
         let this = self.clone();
         tokio::spawn(
             async move {
@@ -7923,7 +7965,7 @@ impl Service {
                             // because our max shard count is relatively low anyway. This policy
                             // will be adjusted in future once we support higher shard count.
                             new_shard_count: MAX_SHARDS.literal(),
-                            new_stripe_size: Some(ShardParameters::DEFAULT_STRIPE_SIZE),
+                            new_stripe_size,
                         },
                     )
                     .await

test_runner/fixtures/pageserver/http.py

@@ -1070,11 +1070,14 @@ class PageserverHttpClient(requests.Session, MetricsGetter):
         tenant_id: TenantId | TenantShardId,
         timeline_id: TimelineId,
         batch_size: int | None = None,
+        behavior_v2: bool = False,
         **kwargs,
     ) -> set[TimelineId]:
-        params = {}
+        params: dict[str, Any] = {}
         if batch_size is not None:
             params["batch_size"] = batch_size
+        if behavior_v2:
+            params["detach_behavior"] = "v2"
         res = self.put(
             f"http://localhost:{self.port}/v1/tenant/{tenant_id}/timeline/{timeline_id}/detach_ancestor",
             params=params,

test_runner/performance/test_compute_ctl_api.py

@@ -41,24 +41,24 @@ def test_compute_ctl_api_latencies(
     zenbenchmark.record(
         "status_response_latency_p50_us",
         status_response_latency_us[len(status_response_latency_us) // 2],
-        "microseconds",
+        "μs",
         MetricReport.LOWER_IS_BETTER,
     )
     zenbenchmark.record(
         "metrics_response_latency_p50_us",
         metrics_response_latency_us[len(metrics_response_latency_us) // 2],
-        "microseconds",
+        "μs",
         MetricReport.LOWER_IS_BETTER,
     )
     zenbenchmark.record(
         "status_response_latency_p99_us",
         status_response_latency_us[len(status_response_latency_us) * 99 // 100],
-        "microseconds",
+        "μs",
         MetricReport.LOWER_IS_BETTER,
     )
     zenbenchmark.record(
         "metrics_response_latency_p99_us",
         metrics_response_latency_us[len(metrics_response_latency_us) * 99 // 100],
-        "microseconds",
+        "μs",
         MetricReport.LOWER_IS_BETTER,
     )

test_runner/performance/test_logical_replication.py

@@ -23,6 +23,25 @@ if TYPE_CHECKING:
     from psycopg2.extensions import connection, cursor
 
 
+"""
+These benchmarks stress test logical replication within Neon. In order to run
+them locally, they require setting up some infrastructure. See
+https://docs.neon.build/compute/logical_replication_benchmarks.html for how to
+do that. After setting that up, run the following shell commands.
+
+# These are the project IDs setup for the purposes of running these benchmarks
+export BENCHMARK_PROJECT_ID_PUB=
+export BENCHMARK_PROJECT_ID_SUB=
+
+# See https://neon.tech/docs/manage/api-keys
+export NEON_API_KEY=
+
+# Fiddling with the --timeout parameter may be required depending on the
+# performance of the benchmark
+pytest -m remote_cluster 'test_runner/performance/test_logical_replication.py'
+"""
+
+
 @pytest.mark.timeout(1000)
 def test_logical_replication(neon_simple_env: NeonEnv, pg_bin: PgBin, vanilla_pg: VanillaPostgres):
     env = neon_simple_env

test_runner/regress/test_compaction.py

@@ -524,6 +524,42 @@ def test_pageserver_gc_compaction_trigger(neon_env_builder: NeonEnvBuilder):
     workload.validate(env.pageserver.id)
 
 
+def test_pageserver_small_tenant_compaction(neon_env_builder: NeonEnvBuilder):
+    """
+    Create a small tenant that rarely needs compaction and ensure that everything works.
+    """
+    SMOKE_CONF = {
+        # Run both gc and gc-compaction.
+        "gc_period": "5s",
+        "compaction_period": "5s",
+        # No PiTR interval and small GC horizon
+        "pitr_interval": "0s",
+        "gc_horizon": 1024,
+        "lsn_lease_length": "0s",
+    }
+
+    env = neon_env_builder.init_start(initial_tenant_conf=SMOKE_CONF)
+    tenant_id = env.initial_tenant
+    timeline_id = env.initial_timeline
+
+    ps_http = env.pageserver.http_client()
+
+    workload = Workload(env, tenant_id, timeline_id)
+    workload.init(env.pageserver.id)
+
+    log.info("Writing initial data ...")
+    workload.write_rows(10000, env.pageserver.id)
+
+    for _ in range(100):
+        workload.churn_rows(10, env.pageserver.id, upload=False, ingest=False)
+        ps_http.timeline_checkpoint(tenant_id, timeline_id, wait_until_uploaded=True)
+        ps_http.timeline_compact(tenant_id, timeline_id)
+        ps_http.timeline_gc(tenant_id, timeline_id, None)
+
+    log.info("Validating at workload end ...")
+    workload.validate(env.pageserver.id)
+
+
 # Stripe sizes in number of pages.
 TINY_STRIPES = 16
 LARGE_STRIPES = 32768

test_runner/regress/test_storage_controller.py

@@ -1749,18 +1749,23 @@ def test_storage_controller_re_attach(neon_env_builder: NeonEnvBuilder):
     # Restart the failed pageserver
     victim_ps.start()
 
+    env.storage_controller.reconcile_until_idle()
+
     # We expect that the re-attach call correctly tipped off the pageserver that its locations
     # are all secondaries now.
     locations = victim_ps.http_client().tenant_list_locations()["tenant_shards"]
     assert len(locations) == 2
     assert all(loc[1]["mode"] == "Secondary" for loc in locations)
 
-    # We expect that this situation resulted from the re_attach call, and not any explicit
-    # Reconciler runs: assert that the reconciliation count has not gone up since we restarted.
+    # We expect that this situation resulted from background reconciliations
+    # Reconciler runs: assert that the reconciliation count has gone up by exactly
+    # one for each shard
     reconciles_after_restart = env.storage_controller.get_metric_value(
         "storage_controller_reconcile_complete_total", filter={"status": "ok"}
     )
-    assert reconciles_after_restart == reconciles_before_restart
+
+    assert reconciles_before_restart is not None
+    assert reconciles_after_restart == reconciles_before_restart + 2
 
 
 def test_storage_controller_shard_scheduling_policy(neon_env_builder: NeonEnvBuilder):

test_runner/regress/test_tenant_size.py

@@ -436,7 +436,7 @@ def test_single_branch_get_tenant_size_grows(
         # when our tenant is configured with a tiny pitr interval, dropping a table should
         # cause synthetic size to go down immediately
         tenant_config["pitr_interval"] = "0s"
-        env.pageserver.http_client().set_tenant_config(tenant_id, tenant_config)
+        env.storage_controller.pageserver_api().set_tenant_config(tenant_id, tenant_config)
         (current_lsn, size) = get_current_consistent_size(
             env, endpoint, size_debug_file, http_client, tenant_id, timeline_id
         )

test_runner/regress/test_timeline_archive.py

@@ -42,6 +42,14 @@ def test_timeline_archive(neon_env_builder: NeonEnvBuilder, shard_count: int):
         # If we run the unsharded version, talk to the storage controller
         ps_http = env.storage_controller.pageserver_api()
 
+    for ps in env.pageservers:
+        # We make /archival_config requests that are intended to fail.
+        # It's expected that storcon drops requests to other pageservers after
+        # it gets the first error (https://github.com/neondatabase/neon/issues/11177)
+        ps.allowed_errors.append(
+            ".*WARN.* path=/v1/tenant/.*/archival_config .*request was dropped before completing",
+        )
+
     # first try to archive a non existing timeline for an existing tenant:
     invalid_timeline_id = TimelineId.generate()
     with pytest.raises(PageserverApiException, match="timeline not found") as exc:

test_runner/regress/test_timeline_detach_ancestor.py

@@ -319,8 +319,9 @@ def test_ancestor_detach_reparents_earlier(neon_env_builder: NeonEnvBuilder):
             # this does not contain Z in the end, so fromisoformat accepts it
             # it is to be in line with the deletion timestamp.. well, almost.
             when = original_ancestor[2][:26]
-            when_ts = datetime.datetime.fromisoformat(when)
-            assert when_ts < datetime.datetime.now()
+            when_ts = datetime.datetime.fromisoformat(when).replace(tzinfo=datetime.UTC)
+            now = datetime.datetime.utcnow().replace(tzinfo=datetime.UTC)
+            assert when_ts < now
             assert len(lineage.get("reparenting_history", [])) == 0
         elif expected_ancestor == timeline_id:
             assert len(lineage.get("original_ancestor", [])) == 0
@@ -342,6 +343,138 @@ def test_ancestor_detach_reparents_earlier(neon_env_builder: NeonEnvBuilder):
     wait_timeline_detail_404(client, env.initial_tenant, env.initial_timeline)
 
 
+def test_ancestor_detach_behavior_v2(neon_env_builder: NeonEnvBuilder):
+    """
+    Test the v2 behavior of ancestor detach.
+
+    old main -------|---------X--------->
+                    |         |         |
+                    |         |         +-> after
+                    |         +--X empty snapshot branch
+                    |            |
+                    |            +-> branch-to-detach
+                    |
+                    +-> earlier
+
+    Ends up as:
+
+    old main -------|---------X--------->
+                    |         |         |
+                    |         |         +-> after
+                    |         +--X empty snapshot branch
+                    |
+                    +-> earlier
+
+
+    new main -------|---------|----> branch-to-detach
+    """
+
+    env = neon_env_builder.init_start()
+
+    env.pageserver.allowed_errors.extend(SHUTDOWN_ALLOWED_ERRORS)
+
+    client = env.pageserver.http_client()
+
+    with env.endpoints.create_start("main", tenant_id=env.initial_tenant) as ep:
+        ep.safe_psql("CREATE TABLE foo (i BIGINT);")
+        ep.safe_psql("CREATE TABLE audit AS SELECT 1 as starts;")
+
+        branchpoint_pipe = wait_for_last_flush_lsn(
+            env, ep, env.initial_tenant, env.initial_timeline
+        )
+
+        ep.safe_psql("INSERT INTO foo SELECT i::bigint FROM generate_series(0, 8191) g(i);")
+
+        branchpoint_x = wait_for_last_flush_lsn(env, ep, env.initial_tenant, env.initial_timeline)
+        client.timeline_checkpoint(env.initial_tenant, env.initial_timeline)
+
+        ep.safe_psql("INSERT INTO foo SELECT i::bigint FROM generate_series(8192, 16383) g(i);")
+        wait_for_last_flush_lsn(env, ep, env.initial_tenant, env.initial_timeline)
+
+    earlier = env.create_branch(
+        "earlier", ancestor_branch_name="main", ancestor_start_lsn=branchpoint_pipe
+    )
+
+    snapshot_branchpoint = env.create_branch(
+        "snapshot_branchpoint", ancestor_branch_name="main", ancestor_start_lsn=branchpoint_x
+    )
+
+    branch_to_detach = env.create_branch(
+        "branch_to_detach",
+        ancestor_branch_name="snapshot_branchpoint",
+        ancestor_start_lsn=branchpoint_x,
+    )
+
+    after = env.create_branch("after", ancestor_branch_name="main", ancestor_start_lsn=None)
+
+    all_reparented = client.detach_ancestor(env.initial_tenant, branch_to_detach, behavior_v2=True)
+    assert set(all_reparented) == set()
+
+    env.pageserver.quiesce_tenants()
+
+    # checking the ancestor after is much faster than waiting for the endpoint not start
+    expected_result = [
+        ("main", env.initial_timeline, None, 16384, 1),
+        ("after", after, env.initial_timeline, 16384, 1),
+        ("snapshot_branchpoint", snapshot_branchpoint, env.initial_timeline, 8192, 1),
+        ("branch_to_detach", branch_to_detach, None, 8192, 1),
+        ("earlier", earlier, env.initial_timeline, 0, 1),
+    ]
+
+    assert isinstance(env.pageserver_remote_storage, LocalFsStorage)
+
+    for branch_name, queried_timeline, expected_ancestor, _, _ in expected_result:
+        details = client.timeline_detail(env.initial_tenant, queried_timeline)
+        ancestor_timeline_id = details["ancestor_timeline_id"]
+        if expected_ancestor is None:
+            assert ancestor_timeline_id is None
+        else:
+            assert (
+                TimelineId(ancestor_timeline_id) == expected_ancestor
+            ), f"when checking branch {branch_name}, mapping={expected_result}"
+
+        index_part = env.pageserver_remote_storage.index_content(
+            env.initial_tenant, queried_timeline
+        )
+        lineage = index_part["lineage"]
+        assert lineage is not None
+
+        assert lineage.get("reparenting_history_overflown", "false") == "false"
+
+        if queried_timeline == branch_to_detach:
+            original_ancestor = lineage["original_ancestor"]
+            assert original_ancestor is not None
+            assert original_ancestor[0] == str(env.initial_timeline)
+            assert original_ancestor[1] == str(branchpoint_x)
+
+            # this does not contain Z in the end, so fromisoformat accepts it
+            # it is to be in line with the deletion timestamp.. well, almost.
+            when = original_ancestor[2][:26]
+            when_ts = datetime.datetime.fromisoformat(when).replace(tzinfo=datetime.UTC)
+            now = datetime.datetime.utcnow().replace(tzinfo=datetime.UTC)
+            assert when_ts < now
+            assert len(lineage.get("reparenting_history", [])) == 0
+        elif expected_ancestor == branch_to_detach:
+            assert len(lineage.get("original_ancestor", [])) == 0
+            assert lineage["reparenting_history"] == [str(env.initial_timeline)]
+        else:
+            assert len(lineage.get("original_ancestor", [])) == 0
+            assert len(lineage.get("reparenting_history", [])) == 0
+
+    for name, _, _, rows, starts in expected_result:
+        with env.endpoints.create_start(name, tenant_id=env.initial_tenant) as ep:
+            assert ep.safe_psql("SELECT count(*) FROM foo;")[0][0] == rows
+            assert ep.safe_psql(f"SELECT count(*) FROM audit WHERE starts = {starts}")[0][0] == 1
+
+    # delete the new timeline to confirm it doesn't carry over the anything from the old timeline
+    client.timeline_delete(env.initial_tenant, branch_to_detach)
+    wait_timeline_detail_404(client, env.initial_tenant, branch_to_detach)
+
+    # delete the after timeline
+    client.timeline_delete(env.initial_tenant, after)
+    wait_timeline_detail_404(client, env.initial_tenant, after)
+
+
 def test_detached_receives_flushes_while_being_detached(neon_env_builder: NeonEnvBuilder):
     """
     Makes sure that the timeline is able to receive writes through-out the detach process.

workspace_hack/Cargo.toml

@@ -16,6 +16,7 @@ license.workspace = true
 ### BEGIN HAKARI SECTION
 [dependencies]
 ahash = { version = "0.8" }
+anstream = { version = "0.6" }
 anyhow = { version = "1", features = ["backtrace"] }
 base64-594e8ee84c453af0 = { package = "base64", version = "0.13", features = ["alloc"] }
 base64-647d43efb71741da = { package = "base64", version = "0.21" }
@@ -30,6 +31,8 @@ der = { version = "0.7", default-features = false, features = ["oid", "pem", "st
 deranged = { version = "0.3", default-features = false, features = ["powerfmt", "serde", "std"] }
 digest = { version = "0.10", features = ["mac", "oid", "std"] }
 either = { version = "1" }
+env_filter = { version = "0.1", default-features = false, features = ["regex"] }
+env_logger = { version = "0.11" }
 fail = { version = "0.5", default-features = false, features = ["failpoints"] }
 form_urlencoded = { version = "1" }
 futures-channel = { version = "0.3", features = ["sink"] }
@@ -47,8 +50,7 @@ hyper-dff4ba8e3ae991db = { package = "hyper", version = "1", features = ["full"]
 hyper-util = { version = "0.1", features = ["client-legacy", "http1", "http2", "server", "service"] }
 indexmap-dff4ba8e3ae991db = { package = "indexmap", version = "1", default-features = false, features = ["std"] }
 indexmap-f595c2ba2a3f28df = { package = "indexmap", version = "2", features = ["serde"] }
-itertools-5ef9efb8ec2df382 = { package = "itertools", version = "0.12" }
-itertools-93f6ce9d446188ac = { package = "itertools", version = "0.10" }
+itertools = { version = "0.12" }
 lazy_static = { version = "1", default-features = false, features = ["spin_no_std"] }
 libc = { version = "0.2", features = ["extra_traits", "use_std"] }
 log = { version = "0.4", default-features = false, features = ["std"] }
@@ -115,8 +117,7 @@ half = { version = "2", default-features = false, features = ["num-traits"] }
 hashbrown = { version = "0.14", features = ["raw"] }
 indexmap-dff4ba8e3ae991db = { package = "indexmap", version = "1", default-features = false, features = ["std"] }
 indexmap-f595c2ba2a3f28df = { package = "indexmap", version = "2", features = ["serde"] }
-itertools-5ef9efb8ec2df382 = { package = "itertools", version = "0.12" }
-itertools-93f6ce9d446188ac = { package = "itertools", version = "0.10" }
+itertools = { version = "0.12" }
 libc = { version = "0.2", features = ["extra_traits", "use_std"] }
 log = { version = "0.4", default-features = false, features = ["std"] }
 memchr = { version = "2" }