./github/scripts/auto-backport.py: don't remove backport label when backport process has an error

Today, when the `Fixes` prefix is missing or the developer is not a collaborator with `scylladbbot` we remove the backport labels to prevent the process from starting and notifying the developers.

Developers are worried that removing these backport labels will cause us to forget we need to do these backports. @nyh suggested to add a `scylladbbot/backport_error` label instead

Applied those changes, so when a `Fixes` prefix is missing we will add a `scylladbbot/backport_error` label and stop the process

When a user doesn't accept the invite we will still open the PR but he will not be assigned and will not be able to edit the branch when we have conflicts

Fixes: https://github.com/scylladb/scylla-pkg/issues/4898
Fixes: https://github.com/scylladb/scylla-pkg/issues/4897

.github/CODEOWNERS

@@ -1,5 +1,5 @@
 # AUTH
-auth/* @nuivall @ptrsmrn
+auth/* @nuivall @ptrsmrn @KrzaQ
 
 # CACHE
 row_cache* @tgrabiec
@@ -25,15 +25,15 @@ compaction/* @raphaelsc
 transport/*
 
 # CQL QUERY LANGUAGE
-cql3/* @tgrabiec @nuivall @ptrsmrn
+cql3/* @tgrabiec @nuivall @ptrsmrn @KrzaQ
 
 # COUNTERS
-counters* @nuivall @ptrsmrn
-tests/counter_test* @nuivall @ptrsmrn
+counters* @nuivall @ptrsmrn @KrzaQ
+tests/counter_test* @nuivall @ptrsmrn @KrzaQ
 
 # DOCS
 docs/* @annastuchlik @tzach
-docs/alternator @annastuchlik @tzach @nyh
+docs/alternator @annastuchlik @tzach @nyh @nuivall @ptrsmrn @KrzaQ
 
 # GOSSIP
 gms/* @tgrabiec @asias @kbr-scylla
@@ -74,8 +74,8 @@ streaming/* @tgrabiec @asias
 service/storage_service.* @tgrabiec @asias
 
 # ALTERNATOR
-alternator/* @nyh
-test/alternator/* @nyh
+alternator/* @nyh @nuivall @ptrsmrn @KrzaQ
+test/alternator/* @nyh @nuivall @ptrsmrn @KrzaQ
 
 # HINTED HANDOFF
 db/hints/* @piodul @vladzcloudius @eliransin

.github/ISSUE_TEMPLATE/bug_report.yml

@@ -1,86 +1,15 @@
-name: "Report a bug"
-description: "File a bug report."
-title: "[Bug]: "
-type: "bug"
-labels: bug
-body:
-  - type: checkboxes
-    id: terms
-    attributes:
-      label: Code of Conduct
-      description: "This is Scylla's bug tracker, to be used for reporting bugs only.
+This is Scylla's bug tracker, to be used for reporting bugs only.
 If you have a question about Scylla, and not a bug, please ask it in
-our forum at https://forum.scylladb.com/ or in our slack channel https://slack.scylladb.com/ "
-      options:
-        - label: I have read the disclaimer above and am reporting a suspected malfunction in Scylla.
-          required: true
+our mailing-list at scylladb-dev@googlegroups.com or in our slack channel.
 
-  - type: input
-    id: product-version
-    attributes:
-      label: product version
-      description: Scylla version (or git commit hash)
-      placeholder: ex. scylla-6.1.1
-    validations:
-      required: true
-      
-  - type: input
-    id: cluster-size
-    attributes:
-      label: Cluster Size
-    validations:
-      required: true  
-      
-  - type: input
-    id: os
-    attributes:
-      label: OS
-      placeholder: RHEL/CentOS/Ubuntu/AWS AMI
-    validations:
-      required: true
-      
-  - type: textarea
-    id: additional-data
-    attributes:
-      label: Additional Environmental Data
-      #description: 
-      placeholder: Add additional data
-      value: "Platform (physical/VM/cloud instance type/docker):\n
-Hardware: sockets=   cores=   hyperthreading=   memory=\n
-Disks: (SSD/HDD, count)"
-    validations:
-      required: false
-      
-  - type: textarea
-    id: reproducer-steps
-    attributes:
-      label: Reproduction Steps
-      placeholder: Describe how to reproduce the problem
-      value: "The steps to reproduce the problem are:"
-    validations:
-      required: true
-      
-  - type: textarea
-    id: the-problem
-    attributes:
-      label: What is the problem?
-      placeholder: Describe the problem you found
-      value: "The problem is that"
-    validations:
-      required: true
-      
-  - type: textarea
-    id: what-happened
-    attributes:
-      label: Expected behavior?
-      placeholder: Describe what should have happened
-      value: "I expected that "
-    validations:
-      required: true
-      
-  - type: textarea
-    id: logs
-    attributes:
-      label: Relevant log output
-      description: Please copy and paste any relevant log output. This will be automatically formatted into code, so no need for backticks.
-      render: shell
+- [] I have read the disclaimer above, and I am reporting a suspected malfunction in Scylla.
+
+*Installation details*
+Scylla version (or git commit hash):
+Cluster size:
+OS (RHEL/CentOS/Ubuntu/AWS AMI):
+
+*Hardware details (for performance issues)*          Delete if unneeded
+Platform (physical/VM/cloud instance type/docker):
+Hardware: sockets= cores= hyperthreading= memory=
+Disks: (SSD/HDD, count)

.github/scripts/auto-backport.py

@@ -47,29 +47,13 @@ def create_pull_request(repo, new_branch_name, base_branch_name, pr, backport_pr
             draft=is_draft
         )
         logging.info(f"Pull request created: {backport_pr.html_url}")
-        labels_to_add = []
-        priority_labels = {"P0", "P1"}
-        parent_pr_labels = [label.name for label in pr.labels]
-        for label in priority_labels:
-            if label in parent_pr_labels:
-                labels_to_add.append(label)
-                labels_to_add.append("force_on_cloud")
-                logging.info(f"Adding {label} and force_on_cloud labels from parent PR to backport PR")
-                break  # Only apply the highest priority label
-        
         if is_collaborator:
             backport_pr.add_to_assignees(pr.user)
         if is_draft:
-            labels_to_add.append("conflicts")
+            backport_pr.add_to_labels("conflicts")
             pr_comment = f"@{pr.user.login} - This PR was marked as draft because it has conflicts\n"
             pr_comment += "Please resolve them and mark this PR as ready for review"
             backport_pr.create_issue_comment(pr_comment)
-        
-        # Apply all labels at once if we have any
-        if labels_to_add:
-            backport_pr.add_to_labels(*labels_to_add)
-            logging.info(f"Added labels to backport PR: {labels_to_add}")
-
         logging.info(f"Assigned PR to original author: {pr.user}")
         return backport_pr
     except GithubException as e:
@@ -128,15 +112,10 @@ def backport(repo, pr, version, commits, backport_base_branch, is_collaborator):
                     is_draft = True
                     repo_local.git.add(A=True)
                     repo_local.git.cherry_pick('--continue')
-            # Check if the branch already exists in the remote fork
-            remote_refs = repo_local.git.ls_remote('--heads', fork_repo, new_branch_name)
-            if not remote_refs:
-                # Branch does not exist, create it with a regular push
-                repo_local.git.push(fork_repo, new_branch_name)
-                create_pull_request(repo, new_branch_name, backport_base_branch, pr, backport_pr_title, commits,
-                                    is_draft, is_collaborator)
-            else:
-                logging.info(f"Remote branch {new_branch_name} already exists in fork. Skipping push.")
+            repo_local.git.push(fork_repo, new_branch_name, force=True)
+            create_pull_request(repo, new_branch_name, backport_base_branch, pr, backport_pr_title, commits,
+                                is_draft, is_collaborator)
+
         except GitCommandError as e:
             logging.warning(f"GitCommandError: {e}")
 

.github/scripts/sync_labels.py

@@ -30,13 +30,8 @@ def copy_labels_from_linked_issues(repo, pr_number):
             try:
                 issue = repo.get_issue(int(issue_number))
                 for label in issue.labels:
-                    # Copy ALL labels from issues to PR when PR is opened
                     pr.add_to_labels(label.name)
-                    print(f"Copied label '{label.name}' from issue #{issue_number} to PR #{pr_number}")
-                    if label.name in ['P0', 'P1']:
-                        pr.add_to_labels('force_on_cloud')
-                        print(f"Added force_on_cloud label to PR #{pr_number} due to {label.name} label")
-                print(f"All labels from issue #{issue_number} copied to PR #{pr_number}")
+                print(f"Labels from issue #{issue_number} copied to PR #{pr_number}")
             except Exception as e:
                 print(f"Error processing issue #{issue_number}: {e}")
 
@@ -79,22 +74,9 @@ def sync_labels(repo, number, label, action, is_issue=False):
             target = repo.get_issue(int(pr_or_issue_number))
         if action == 'labeled':
             target.add_to_labels(label)
-            if label in ['P0', 'P1'] and is_issue:
-                # Only add force_on_cloud to PRs when P0/P1 is added to an issue
-                target.add_to_labels('force_on_cloud')
-                print(f"Added 'force_on_cloud' label to PR #{pr_or_issue_number} due to {label} label")
             print(f"Label '{label}' successfully added.")
         elif action == 'unlabeled':
             target.remove_from_labels(label)
-            if label in ['P0', 'P1'] and is_issue:
-                # Check if any other P0/P1 labels remain before removing force_on_cloud
-                remaining_priority_labels = [l.name for l in target.labels if l.name in ['P0', 'P1']]
-                if not remaining_priority_labels:
-                    try:
-                        target.remove_from_labels('force_on_cloud')
-                        print(f"Removed 'force_on_cloud' label from PR #{pr_or_issue_number} as no P0/P1 labels remain")
-                    except Exception as e:
-                        print(f"Warning: Could not remove force_on_cloud label: {e}")
             print(f"Label '{label}' successfully removed.")
         elif action == 'opened':
             copy_labels_from_linked_issues(repo, number)

.github/seastar-bad-include.json

@@ -1,16 +0,0 @@
-{
-    "problemMatcher": [
-        {
-            "owner": "seastar-bad-include",
-            "severity": "error",
-            "pattern": [
-                {
-                    "regexp": "^(.+):(\\d+):(.+)$",
-                    "file": 1,
-                    "line": 2,
-                    "message": 3
-                }
-            ]
-        }
-    ]
-}

.github/workflows/add-label-when-promoted.yaml

@@ -54,13 +54,10 @@ jobs:
           GITHUB_TOKEN: ${{ secrets.AUTO_BACKPORT_TOKEN }}
         run: python .github/scripts/auto-backport.py --repo ${{ github.repository }} --base-branch ${{ github.ref }} --commits ${{ github.event.before }}..${{ github.sha }}
       - name: Check if a valid backport label exists and no backport_error
-        env:
-          LABELS_JSON: ${{ toJson(github.event.pull_request.labels) }}
         id: check_label
         run: |
-          labels_json="$LABELS_JSON"
-          echo "Checking labels:"
-          echo "$labels_json" | jq -r '.[].name'
+          labels_json='${{ toJson(github.event.pull_request.labels) }}'
+          echo "Checking labels: $(echo "$labels_json" | jq -r '.[].name')"
 
           # Check if a valid backport label exists
           if echo "$labels_json" | jq -e 'any(.[] | .name; test("backport/[0-9]+\\.[0-9]+$"))' > /dev/null; then

.github/workflows/call_jira_status_in_progress.yml

@@ -1,12 +0,0 @@
-name: Call Jira Status In Progress
-
-on:
-  pull_request_target:
-    types: [opened]
-
-jobs:
-  call-jira-status-in-progress:
-    uses: scylladb/github-automation/.github/workflows/main_update_jira_status_to_in_progress.yml@main
-    secrets:
-      caller_jira_auth: ${{ secrets.USER_AND_KEY_FOR_JIRA_AUTOMATION }}
-

.github/workflows/call_jira_status_in_review.yml

@@ -1,12 +0,0 @@
-name: Call Jira Status In Review
-
-on:
-  pull_request_target:
-    types: [ready_for_review, review_requested]
-
-jobs:
-  call-jira-status-in-review:
-    uses: scylladb/github-automation/.github/workflows/main_update_jira_status_to_in_review.yml@main
-    secrets:
-      caller_jira_auth: ${{ secrets.USER_AND_KEY_FOR_JIRA_AUTOMATION }}
-

.github/workflows/call_jira_status_ready_for_merge.yml

@@ -1,12 +0,0 @@
-name: Call Jira Status Ready For Merge
-
-on:
-  pull_request_target:
-    types: [labeled]
-
-jobs:
-  call-jira-status-update:
-    uses: scylladb/github-automation/.github/workflows/main_update_jira_status_to_ready_for_merge.yml@main
-    secrets:
-      caller_jira_auth: ${{ secrets.USER_AND_KEY_FOR_JIRA_AUTOMATION }}
-

.github/workflows/conflict_reminder.yaml

@@ -1,16 +1,9 @@
 name: Notify PR Authors of Conflicts
 
-permissions:
-  issues: write
-  pull-requests: write
-
 on:
-  push:
-    branches:
-      - 'master'
-      - 'branch-*'
   schedule:
-    - cron: '0 10 * * 1'  # Runs every Monday at 10:00am
+    - cron: '0 10 * * 1,4'  # Runs every Monday and Thursday at 10:00am
+  workflow_dispatch:      # Manual trigger for testing
 
 jobs:
   notify_conflict_prs:
@@ -21,134 +14,32 @@ jobs:
         uses: actions/github-script@v7
         with:
           script: |
-            console.log("Starting conflict reminder script...");
-            // Print trigger event
-            if (process.env.GITHUB_EVENT_NAME) {
-              console.log(`Workflow triggered by: ${process.env.GITHUB_EVENT_NAME}`);
-            } else {
-              console.log("Could not determine workflow trigger event.");
-            }
-            const isPushEvent = process.env.GITHUB_EVENT_NAME === 'push';
-            console.log(`isPushEvent: ${isPushEvent}`);
-            const twoMonthsAgo = new Date();
-            twoMonthsAgo.setMonth(twoMonthsAgo.getMonth() - 2);
             const prs = await github.paginate(github.rest.pulls.list, {
               owner: context.repo.owner,
               repo: context.repo.repo,
               state: 'open',
               per_page: 100
             });
-            console.log(`Fetched ${prs.length} open PRs`);
-            const recentPrs = prs.filter(pr => new Date(pr.created_at) >= twoMonthsAgo);
-            const validBaseBranches = ['master'];
             const branchPrefix = 'branch-';
-            const oneWeekAgo = new Date();
-            const conflictLabel = 'conflicts';
-            oneWeekAgo.setDate(oneWeekAgo.getDate() - 7);
-            console.log(`One week ago: ${oneWeekAgo.toISOString()}`);
-
-            for (const pr of recentPrs) {
-              console.log(`Checking PR #${pr.number} on base branch '${pr.base.ref}'`);
-              const isBranchX = pr.base.ref.startsWith(branchPrefix);
-              const isMaster = validBaseBranches.includes(pr.base.ref);
-              if (!(isBranchX || isMaster)) {
-                console.log(`PR #${pr.number} skipped: base branch is not 'master' or does not start with '${branchPrefix}'`);
-                continue;
-              }
+            const threeDaysAgo = new Date();
+            const conflictLabel = 'conflicts';          
+            threeDaysAgo.setDate(threeDaysAgo.getDate() - 3);
+            for (const pr of prs) {
+              if (!pr.base.ref.startsWith(branchPrefix)) continue;
+              const hasConflictLabel = pr.labels.some(label => label.name === conflictLabel);
+              if (!hasConflictLabel) continue;
               const updatedDate = new Date(pr.updated_at);
-              console.log(`PR #${pr.number} last updated at: ${updatedDate.toISOString()}`);
-              if (!isPushEvent && updatedDate >= oneWeekAgo) {
-                console.log(`PR #${pr.number} skipped: updated within last week`);
-                continue;
-              }
-              if (pr.assignee === null) {
-                console.log(`PR #${pr.number} skipped: no assignee`);
-                continue;
-              }
-
-              // Fetch PR details to check mergeability
-              let { data: prDetails } = await github.rest.pulls.get({
-                owner: context.repo.owner,
-                repo: context.repo.repo,
-                pull_number: pr.number,
-              });
-              console.log(`PR #${pr.number} mergeable: ${prDetails.mergeable}`);
-
-              // Wait and re-fetch if mergeable is null
-              if (prDetails.mergeable === null) {
-                console.log(`PR #${pr.number} mergeable is null, waiting 2 seconds and retrying...`);
-                await new Promise(resolve => setTimeout(resolve, 2000)); // wait 2 seconds
-                prDetails = (await github.rest.pulls.get({
-                  owner: context.repo.owner,
-                  repo: context.repo.repo,
-                  pull_number: pr.number,
-                })).data;
-                console.log(`PR #${pr.number} mergeable after retry: ${prDetails.mergeable}`);
-              }
-
-              if (prDetails.mergeable === false) {
-                const hasConflictLabel = pr.labels.some(label => label.name === conflictLabel);
-                console.log(`PR #${pr.number} has conflict label: ${hasConflictLabel}`);
-
-                // Fetch comments to check for existing notifications
-                const comments = await github.paginate(github.rest.issues.listComments, {
+              if (updatedDate >= threeDaysAgo) continue;
+              if (pr.assignee === null) continue;
+              const assignee = pr.assignee.login;
+              if (assignee) {
+                await github.rest.issues.createComment({
                   owner: context.repo.owner,
                   repo: context.repo.repo,
                   issue_number: pr.number,
-                  per_page: 100,
+                  body: `@${assignee}, this PR has been open with conflicts. Please resolve the conflicts so we can merge it.`,
                 });
-                
-                // Find last notification comment from the bot
-                const notificationPrefix = `@${pr.assignee.login}, this PR has merge conflicts with the base branch.`;
-                const lastNotification = comments
-                  .filter(c =>
-                    c.user.type === "Bot" &&
-                    c.body.startsWith(notificationPrefix)
-                  )
-                  .sort((a, b) => new Date(b.created_at) - new Date(a.created_at))[0];
-
-                // Check if we should skip notification based on recent notification
-                let shouldSkipNotification = false;
-                if (lastNotification) {
-                  const lastNotified = new Date(lastNotification.created_at);
-                  if (lastNotified >= oneWeekAgo) {
-                    console.log(`PR #${pr.number} skipped: last notification was less than 1 week ago`);
-                    shouldSkipNotification = true;
-                  }
-                }
-
-                // Additional check for push events on draft PRs with conflict labels
-                if (
-                  isPushEvent &&
-                  pr.draft === true &&
-                  hasConflictLabel &&
-                  shouldSkipNotification
-                ) {
-                  continue;
-                }
-
-                if (!hasConflictLabel) {
-                  await github.rest.issues.addLabels({
-                    owner: context.repo.owner,
-                    repo: context.repo.repo,
-                    issue_number: pr.number,
-                    labels: [conflictLabel],
-                  });
-                  console.log(`Added 'conflicts' label to PR #${pr.number}`);
-                }
-                
-                const assignee = pr.assignee.login;
-                if (assignee && !shouldSkipNotification) {
-                  await github.rest.issues.createComment({
-                    owner: context.repo.owner,
-                    repo: context.repo.repo,
-                    issue_number: pr.number,
-                    body: `@${assignee}, this PR has merge conflicts with the base branch. Please resolve the conflicts so we can merge it.`,
-                  });
-                  console.log(`Notified @${assignee} for PR #${pr.number}`);
-                }
-              } else {
-                console.log(`PR #${pr.number} is mergeable, no action needed.`);
-              }
+                console.log(`Notified @${assignee} for PR #${pr.number}`);
+              } 
             }
             console.log(`Total PRs checked: ${prs.length}`);

.github/workflows/iwyu.yaml

@@ -11,8 +11,7 @@ env:
   CLEANER_OUTPUT_PATH: build/clang-include-cleaner.log
   # the "idl" subdirectory does not contain C++ source code. the .hh files in it are
   # supposed to be processed by idl-compiler.py, so we don't check them using the cleaner
-  CLEANER_DIRS: test/unit exceptions alternator api auth cdc compaction db dht gms index lang message mutation mutation_writer node_ops raft redis replica service
-  SEASTAR_BAD_INCLUDE_OUTPUT_PATH: build/seastar-bad-include.log
+  CLEANER_DIRS: test/unit exceptions alternator api auth cdc compaction db dht gms index lang message mutation mutation_writer node_ops redis replica
 
 permissions: {}
 
@@ -81,24 +80,7 @@ jobs:
           done
       - run: |
           echo "::remove-matcher owner=clang-include-cleaner::"
-      - run: |
-          echo "::add-matcher::.github/seastar-bad-include.json"
-      - name: check for seastar includes
-        run: |
-          git -c safe.directory="$PWD"    \
-            grep -nE '#include +"seastar/' \
-            | tee "$SEASTAR_BAD_INCLUDE_OUTPUT_PATH"
-      - run: |
-          echo "::remove-matcher owner=seastar-bad-include::"
       - uses: actions/upload-artifact@v4
         with:
-          name: Logs
-          path: |
-            ${{ env.CLEANER_OUTPUT_PATH }}
-            ${{ env.SEASTAR_BAD_INCLUDE_OUTPUT_PATH }}
-      - name: fail if seastar headers are included as an internal library
-        run: |
-          if [ -s "$SEASTAR_BAD_INCLUDE_OUTPUT_PATH" ]; then
-            echo "::error::Found #include \"seastar/ in the source code. Use angle brackets instead."
-            exit 1
-          fi
+          name: Logs (clang-include-cleaner)
+          path: "./${{ env.CLEANER_OUTPUT_PATH }}"

.github/workflows/make-pr-ready-for-review.yaml

@@ -16,13 +16,6 @@ jobs:
       pull-requests: write
 
     steps:
-      - name: Checkout repository
-        uses: actions/checkout@v4
-        with:
-          repository: ${{ github.repository }}
-          ref: ${{ env.DEFAULT_BRANCH }}
-          token: ${{ secrets.AUTO_BACKPORT_TOKEN }}
-          fetch-depth: 1
       - name: Mark pull request as ready for review
         run:  gh pr ready "${{ github.event.pull_request.number }}"
         env:

.github/workflows/pr-require-backport-label.yaml

@@ -13,8 +13,6 @@ jobs:
       issues: write
       pull-requests: write
     steps:
-      - name: Wait for label to be added
-        run: sleep 1m
       - uses: mheap/github-action-required-labels@v5
         with:
           mode: minimum

.github/workflows/sync-labels.yaml

@@ -37,13 +37,13 @@ jobs:
         run: python .github/scripts/sync_labels.py --repo ${{ github.repository }} --number ${{ github.event.number }} --action ${{ github.event.action }}
 
       - name: Pull request labeled or unlabeled event
-        if: github.event_name == 'pull_request_target' && (startsWith(github.event.label.name, 'backport/') || github.event.label.name == 'P0' || github.event.label.name == 'P1')
+        if: github.event_name == 'pull_request_target' && startsWith(github.event.label.name, 'backport/')
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
         run: python .github/scripts/sync_labels.py --repo ${{ github.repository }} --number ${{ github.event.number }} --action ${{ github.event.action }} --label ${{ github.event.label.name }}
 
       - name: Issue labeled or unlabeled event
-        if: github.event_name == 'issues' && (startsWith(github.event.label.name, 'backport/') || github.event.label.name == 'P0' || github.event.label.name == 'P1')
+        if: github.event_name == 'issues' && startsWith(github.event.label.name, 'backport/')
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
         run: python .github/scripts/sync_labels.py --repo ${{ github.repository }} --number ${{ github.event.issue.number }} --action ${{ github.event.action }} --is_issue --label ${{ github.event.label.name }}

.github/workflows/trigger-scylla-ci.yaml

@@ -1,21 +0,0 @@
-name: Trigger Scylla CI Route
-
-on:
-  issue_comment:
-    types: [created]
-
-jobs:
-  trigger-jenkins:
-    if: github.event.comment.user.login != 'scylladbbot' && contains(github.event.comment.body, '@scylladbbot') && contains(github.event.comment.body, 'trigger-ci')
-    runs-on: ubuntu-latest
-    steps:
-      - name: Trigger Scylla-CI-Route Jenkins Job
-        env:
-          JENKINS_USER: ${{ secrets.JENKINS_USERNAME }}
-          JENKINS_API_TOKEN: ${{ secrets.JENKINS_TOKEN }}
-          JENKINS_URL: "https://jenkins.scylladb.com"
-        run: |
-          PR_NUMBER=${{ github.event.issue.number }}
-          PR_REPO_NAME=${{ github.event.repository.full_name }}
-          curl -X POST "$JENKINS_URL/job/releng/job/Scylla-CI-Route/buildWithParameters?PR_NUMBER=$PR_NUMBER&PR_REPO_NAME=$PR_REPO_NAME" \
-          --user "$JENKINS_USER:$JENKINS_API_TOKEN" --fail -i -v

.github/workflows/urgent_issue_reminder.yml

@@ -2,7 +2,7 @@ name: Urgent Issue Reminder
 
 on:
   schedule:
-    - cron: '10 8 * * *' # Runs daily at 8 AM
+    - cron: '10 8 * * 1' # Runs every Monday at 8 AM
 
 jobs:
   reminder:

.gitignore

@@ -35,6 +35,3 @@ compile_commands.json
 .envrc
 clang_build
 .idea/
-nuke
-rust/target
-

.gitmodules

@@ -9,6 +9,9 @@
 [submodule "abseil"]
 	path = abseil
 	url = ../abseil-cpp
+[submodule "scylla-tools"]
+	path = tools/java
+	url = ../scylla-tools-java
 [submodule "scylla-python3"]
 	path = tools/python3
 	url = ../scylla-python3

CMakeLists.txt

@@ -49,7 +49,7 @@ include(limit_jobs)
 set(CMAKE_CXX_STANDARD "23" CACHE INTERNAL "")
 set(CMAKE_CXX_EXTENSIONS ON CACHE INTERNAL "")
 set(CMAKE_CXX_SCAN_FOR_MODULES OFF CACHE INTERNAL "")
-set(CMAKE_VISIBILITY_INLINES_HIDDEN ON)
+set(CMAKE_CXX_VISIBILITY_PRESET hidden)
 
 if(is_multi_config)
     find_package(Seastar)
@@ -90,13 +90,13 @@ if(is_multi_config)
     add_dependencies(Seastar::seastar_testing Seastar)
 else()
     set(Seastar_TESTING ON CACHE BOOL "" FORCE)
-    set(Seastar_API_LEVEL 9 CACHE STRING "" FORCE)
+    set(Seastar_API_LEVEL 7 CACHE STRING "" FORCE)
     set(Seastar_DEPRECATED_OSTREAM_FORMATTERS OFF CACHE BOOL "" FORCE)
     set(Seastar_APPS ON CACHE BOOL "" FORCE)
     set(Seastar_EXCLUDE_APPS_FROM_ALL ON CACHE BOOL "" FORCE)
     set(Seastar_EXCLUDE_TESTS_FROM_ALL ON CACHE BOOL "" FORCE)
     set(Seastar_IO_URING ON CACHE BOOL "" FORCE)
-    set(Seastar_SCHEDULING_GROUPS_COUNT 21 CACHE STRING "" FORCE)
+    set(Seastar_SCHEDULING_GROUPS_COUNT 19 CACHE STRING "" FORCE)
     set(Seastar_UNUSED_RESULT_ERROR ON CACHE BOOL "" FORCE)
     add_subdirectory(seastar)
     target_compile_definitions (seastar
@@ -163,6 +163,14 @@ file(MAKE_DIRECTORY "${scylla_gen_build_dir}")
 include(add_version_library)
 generate_scylla_version()
 
+add_library(scylla-zstd STATIC
+    zstd.cc)
+target_link_libraries(scylla-zstd
+  PRIVATE
+    db
+    Seastar::seastar
+    zstd::libzstd)
+
 add_library(scylla-main STATIC)
 target_sources(scylla-main
   PRIVATE
@@ -170,23 +178,32 @@ target_sources(scylla-main
     bytes.cc
     client_data.cc
     clocks-impl.cc
-    sstable_dict_autotrainer.cc
+    collection_mutation.cc
+    compress.cc
+    converting_mutation_partition_applier.cc
+    counters.cc
+    direct_failure_detector/failure_detector.cc
+    duration.cc
     exceptions/exceptions.cc
+    frozen_schema.cc
+    generic_server.cc
     debug.cc
     init.cc
-    keys/keys.cc
+    keys.cc
+    multishard_mutation_query.cc
     mutation_query.cc
     node_ops/task_manager_module.cc
     partition_slice_builder.cc
-    query/query.cc
+    querier.cc
+    query.cc
     query_ranges_to_vnodes.cc
-    query/query-result-set.cc
+    query-result-set.cc
     tombstone_gc_options.cc
     tombstone_gc.cc
     reader_concurrency_semaphore.cc
     reader_concurrency_semaphore_group.cc
+    schema_mutations.cc
     serializer.cc
-    service/direct_failure_detector/failure_detector.cc
     sstables_loader.cc
     table_helper.cc
     tasks/task_handler.cc
@@ -197,6 +214,7 @@ target_sources(scylla-main
     vint-serialization.cc)
 target_link_libraries(scylla-main
   PRIVATE
+    "$<LINK_LIBRARY:WHOLE_ARCHIVE,scylla-zstd>"
     db
     absl::headers
     absl::btree
@@ -262,6 +280,7 @@ add_subdirectory(mutation)
 add_subdirectory(mutation_writer)
 add_subdirectory(node_ops)
 add_subdirectory(readers)
+add_subdirectory(redis)
 add_subdirectory(replica)
 add_subdirectory(raft)
 add_subdirectory(repair)
@@ -276,7 +295,6 @@ add_subdirectory(tracing)
 add_subdirectory(transport)
 add_subdirectory(types)
 add_subdirectory(utils)
-add_subdirectory(vector_search)
 add_version_library(scylla_version
     release.cc)
 
@@ -306,6 +324,7 @@ set(scylla_libs
     mutation_writer
     raft
     readers
+    redis
     repair
     replica
     schema
@@ -318,8 +337,7 @@ set(scylla_libs
     tracing
     transport
     types
-    utils
-    vector_search)
+    utils)
 target_link_libraries(scylla PRIVATE
     ${scylla_libs})
 
@@ -353,6 +371,3 @@ endif()
 if(Scylla_BUILD_INSTRUMENTED)
   add_subdirectory(pgo)
 endif()
-
-add_executable(patchelf
-  tools/patchelf.cc)

CONTRIBUTING.md

@@ -12,7 +12,7 @@ Please use the [issue tracker](https://github.com/scylladb/scylla/issues/) to re
 
 ## Contributing code to Scylla
 
-Before you can contribute code to Scylla for the first time, you should sign the [Contributor License Agreement](https://www.scylladb.com/open-source/contributor-agreement/) and send the signed form to cla@scylladb.com. You can then submit your changes as patches to the [scylladb-dev mailing list](https://groups.google.com/forum/#!forum/scylladb-dev) or as a pull request to the [Scylla project on github](https://github.com/scylladb/scylla).
+Before you can contribute code to Scylla for the first time, you should sign the [Contributor License Agreement](https://www.scylladb.com/open-source/contributor-agreement/) and send the signed form cla@scylladb.com. You can then submit your changes as patches to the [scylladb-dev mailing list](https://groups.google.com/forum/#!forum/scylladb-dev) or as a pull request to the [Scylla project on github](https://github.com/scylladb/scylla).
 If you need help formatting or sending patches, [check out these instructions](https://github.com/scylladb/scylla/wiki/Formatting-and-sending-patches).
 
 The Scylla C++ source code uses the [Seastar coding style](https://github.com/scylladb/seastar/blob/master/coding-style.md) so please adhere to that in your patches. Note that Scylla code is written with `using namespace seastar`, so should not explicitly add the `seastar::` prefix to Seastar symbols. You will usually not need to add `using namespace seastar` to new source files, because most Scylla header files have `#include "seastarx.hh"`, which does this.

HACKING.md

@@ -43,7 +43,7 @@ $ ./tools/toolchain/dbuild ninja build/release/scylla
 $ ./tools/toolchain/dbuild ./build/release/scylla --developer-mode 1
 ```
 
-Note: do not mix environments - either perform all your work with dbuild, or natively on the host.
+Note: do not mix environemtns - either perform all your work with dbuild, or natively on the host.
 Note2: you can get to an interactive shell within dbuild by running it without any parameters:
 ```bash
 $ ./tools/toolchain/dbuild
@@ -91,7 +91,7 @@ You can also specify a single mode. For example
 $ ninja-build release
 ```
 
-Will build everything in release mode. The valid modes are
+Will build everytihng in release mode. The valid modes are
 
 * Debug: Enables [AddressSanitizer](https://github.com/google/sanitizers/wiki/AddressSanitizer)
   and other sanity checks. It has no optimizations, which allows for debugging with tools like
@@ -220,9 +220,28 @@ On a development machine, one might run Scylla as
 $ SCYLLA_HOME=$HOME/scylla build/release/scylla --overprovisioned --developer-mode=yes
 ```
 
-To interact with scylla it is recommended to build our version of
-cqlsh. It is available at
-https://github.com/scylladb/scylla-cqlsh and is available as a submodule.
+To interact with scylla it is recommended to build our versions of
+cqlsh and nodetool. They are available at
+https://github.com/scylladb/scylla-tools-java and can be built with
+
+```bash
+$ sudo ./install-dependencies.sh
+$ ant jar
+```
+
+cqlsh should work out of the box, but nodetool depends on a running
+scylla-jmx (https://github.com/scylladb/scylla-jmx). It can be build
+with
+
+```bash
+$ mvn package
+```
+
+and must be started with
+
+```bash
+$ ./scripts/scylla-jmx
+```
 
 ### Branches and tags
 
@@ -361,7 +380,7 @@ avoid that the gold linker can be told to create an index with
 
 More info at https://gcc.gnu.org/wiki/DebugFission.
 
-Both options can be enabled by passing `--split-dwarf` to configure.py.
+Both options can be enable by passing `--split-dwarf` to configure.py.
 
 Note that distcc is *not* compatible with it, but icecream
 (https://github.com/icecc/icecream) is.
@@ -370,7 +389,7 @@ Note that distcc is *not* compatible with it, but icecream
 
 Sometimes Scylla development is closely tied with a feature being developed in Seastar. It can be useful to compile Scylla with a particular check-out of Seastar.
 
-One way to do this is to create a local remote for the Seastar submodule in the Scylla repository:
+One way to do this it to create a local remote for the Seastar submodule in the Scylla repository:
 
 ```bash
 $ cd $HOME/src/scylla

NOTICE.txt

@@ -1,6 +1,9 @@
 This project includes code developed by the Apache Software Foundation (http://www.apache.org/),
 especially Apache Cassandra.
 
+It includes files from https://github.com/antonblanchard/crc32-vpmsum (author Anton Blanchard <anton@au.ibm.com>, IBM).
+These files are located in utils/arch/powerpc/crc32-vpmsum. Their license may be found in licenses/LICENSE-crc32-vpmsum.TXT.
+
 It includes modified code from https://gitbox.apache.org/repos/asf?p=cassandra-dtest.git (owned by The Apache Software Foundation)
 
 It includes modified tests from https://github.com/etcd-io/etcd.git (owned by The etcd Authors)

README.md

@@ -18,7 +18,7 @@ Scylla is fairly fussy about its build environment, requiring very recent
 versions of the C++23 compiler and of many libraries to build. The document
 [HACKING.md](HACKING.md) includes detailed information on building and
 developing Scylla, but to get Scylla building quickly on (almost) any build
-machine, Scylla offers a [frozen toolchain](tools/toolchain/README.md).
+machine, Scylla offers a [frozen toolchain](tools/toolchain/README.md),
 This is a pre-configured Docker image which includes recent versions of all
 the required compilers, libraries and build tools. Using the frozen toolchain
 allows you to avoid changing anything in your build machine to meet Scylla's

SCYLLA-VERSION-GEN

@@ -78,7 +78,7 @@ fi
 
 # Default scylla product/version tags
 PRODUCT=scylla
-VERSION=2026.1.0-dev
+VERSION=2025.2.0-dev
 
 if test -f version
 then

alternator/CMakeLists.txt

@@ -17,7 +17,6 @@ target_sources(alternator
     streams.cc
     consumed_capacity.cc
     ttl.cc
-    parsed_expression_cache.cc
     ${cql_grammar_srcs})
 target_include_directories(alternator
   PUBLIC

alternator/auth.cc

@@ -11,6 +11,7 @@
 #include "utils/log.hh"
 #include <string>
 #include <string_view>
+#include "bytes.hh"
 #include "alternator/auth.hh"
 #include <fmt/format.h>
 #include "auth/password_authenticator.hh"

alternator/consumed_capacity.cc

@@ -24,7 +24,7 @@ static constexpr uint64_t KB = 1024ULL;
 static constexpr uint64_t RCU_BLOCK_SIZE_LENGTH = 4*KB;
 static constexpr uint64_t WCU_BLOCK_SIZE_LENGTH = 1*KB;
 
-bool consumed_capacity_counter::should_add_capacity(const rjson::value& request) {
+static bool should_add_capacity(const rjson::value& request) {
     const rjson::value* return_consumed = rjson::find(request, "ReturnConsumedCapacity");
     if (!return_consumed) {
         return false;
@@ -62,22 +62,15 @@ static uint64_t calculate_half_units(uint64_t unit_block_size, uint64_t total_by
 rcu_consumed_capacity_counter::rcu_consumed_capacity_counter(const rjson::value& request, bool is_quorum) :
         consumed_capacity_counter(should_add_capacity(request)),_is_quorum(is_quorum) {
 }
-uint64_t rcu_consumed_capacity_counter::get_half_units(uint64_t total_bytes, bool is_quorum) noexcept {
-    return calculate_half_units(RCU_BLOCK_SIZE_LENGTH, total_bytes, is_quorum);
-}
 
 uint64_t rcu_consumed_capacity_counter::get_half_units() const noexcept {
-    return get_half_units(_total_bytes, _is_quorum);
+    return calculate_half_units(RCU_BLOCK_SIZE_LENGTH, _total_bytes, _is_quorum);
 }
 
 uint64_t wcu_consumed_capacity_counter::get_half_units() const noexcept {
     return calculate_half_units(WCU_BLOCK_SIZE_LENGTH, _total_bytes, true);
 }
 
-uint64_t wcu_consumed_capacity_counter::get_units(uint64_t total_bytes) noexcept {
-    return calculate_half_units(WCU_BLOCK_SIZE_LENGTH, total_bytes, true) * HALF_UNIT_MULTIPLIER;
-}
-
 wcu_consumed_capacity_counter::wcu_consumed_capacity_counter(const rjson::value& request) :
         consumed_capacity_counter(should_add_capacity(request)) {
 }

alternator/consumed_capacity.hh

@@ -42,25 +42,21 @@ public:
      */
     virtual uint64_t get_half_units() const noexcept = 0;
     uint64_t _total_bytes = 0;
-    static bool should_add_capacity(const rjson::value& request);
 protected:
     bool _should_add_to_reponse = false;
 };
 
 class rcu_consumed_capacity_counter : public consumed_capacity_counter {
+    virtual uint64_t get_half_units() const noexcept;
     bool _is_quorum = false;
 public:
     rcu_consumed_capacity_counter(const rjson::value& request, bool is_quorum);
-    rcu_consumed_capacity_counter(): consumed_capacity_counter(false), _is_quorum(false){}
-    virtual uint64_t get_half_units() const noexcept;
-    static uint64_t get_half_units(uint64_t total_bytes, bool is_quorum) noexcept;
 };
 
 class wcu_consumed_capacity_counter : public consumed_capacity_counter {
     virtual uint64_t get_half_units() const noexcept;
 public:
     wcu_consumed_capacity_counter(const rjson::value& request);
-    static uint64_t get_units(uint64_t total_bytes) noexcept;
 };
 
 }

alternator/controller.cc

@@ -136,7 +136,6 @@ future<> controller::start_server() {
                 [this, addr, alternator_port, alternator_https_port, creds = std::move(creds)] (server& server) mutable {
             return server.init(addr, alternator_port, alternator_https_port, creds,
                     _config.alternator_enforce_authorization,
-                    _config.alternator_max_users_query_size_in_trace_output,
                     &_memory_limiter.local().get_semaphore(),
                     _config.max_concurrent_requests_per_shard);
         }).handle_exception([this, addr, alternator_port, alternator_https_port] (std::exception_ptr ep) {
@@ -168,8 +167,4 @@ future<> controller::request_stop_server() {
     });
 }
 
-future<utils::chunked_vector<client_data>> controller::get_client_data() {
-    return _server.local().get_client_data();
-}
-
 }

alternator/controller.hh

@@ -11,7 +11,7 @@
 #include <seastar/core/sharded.hh>
 #include <seastar/core/smp.hh>
 
-#include "transport/protocol_server.hh"
+#include "protocol_server.hh"
 
 namespace service {
 class storage_proxy;
@@ -90,10 +90,6 @@ public:
     virtual future<> start_server() override;
     virtual future<> stop_server() override;
     virtual future<> request_stop_server() override;
-    // This virtual function is called (on each shard separately) when the
-    // virtual table "system.clients" is read. It is expected to generate a
-    // list of clients connected to this server (on this shard).
-    virtual future<utils::chunked_vector<client_data>> get_client_data() override;
 };
 
 }

alternator/error.hh

@@ -94,9 +94,6 @@ public:
     static api_error internal(std::string msg) {
         return api_error("InternalServerError", std::move(msg), http::reply::status_type::internal_server_error);
     }
-    static api_error payload_too_large(std::string msg) {
-        return api_error("PayloadTooLarge", std::move(msg), status_type::payload_too_large);
-    }
 
     // Provide the "std::exception" interface, to make it easier to print this
     // exception in log messages. Note that this function is *not* used to

alternator/executor.hh

@@ -10,8 +10,8 @@
 
 #include <seastar/core/future.hh>
 #include "seastarx.hh"
+#include <seastar/json/json_elements.hh>
 #include <seastar/core/sharded.hh>
-#include <seastar/util/noncopyable_function.hh>
 
 #include "service/migration_manager.hh"
 #include "service/client_state.hh"
@@ -58,6 +58,33 @@ namespace alternator {
 
 class rmw_operation;
 
+struct make_jsonable : public json::jsonable {
+    rjson::value _value;
+public:
+    explicit make_jsonable(rjson::value&& value);
+    std::string to_json() const override;
+};
+
+/**
+ * Make return type for serializing the object "streamed",
+ * i.e. direct to HTTP output stream. Note: only useful for
+ * (very) large objects as there are overhead issues with this
+ * as well, but for massive lists of return objects this can
+ * help avoid large allocations/many re-allocs
+ */
+json::json_return_type make_streamed(rjson::value&&);
+
+struct json_string : public json::jsonable {
+    std::string _value;
+public:
+    explicit json_string(std::string&& value);
+    std::string to_json() const override;
+};
+
+namespace parsed {
+class path;
+};
+
 schema_ptr get_table(service::storage_proxy& proxy, const rjson::value& request);
 bool is_alternator_keyspace(const sstring& ks_name);
 // Wraps the db::get_tags_of_table and throws if the table is missing the tags extension.
@@ -128,9 +155,6 @@ using attrs_to_get_node = attribute_path_map_node<std::monostate>;
 // optional means we should get all attributes, not specific ones.
 using attrs_to_get = attribute_path_map<std::monostate>;
 
-namespace parsed {
-class expression_cache;
-}
 
 class executor : public peering_sharded_service<executor> {
     gms::gossiper& _gossiper;
@@ -143,27 +167,10 @@ class executor : public peering_sharded_service<executor> {
     // forwarding Alternator request between shards - if necessary for LWT.
     smp_service_group _ssg;
 
-    std::unique_ptr<parsed::expression_cache> _parsed_expression_cache;
-
 public:
     using client_state = service::client_state;
-    // request_return_type is the return type of the executor methods, which
-    // can be one of:
-    // 1. A string, which is the response body for the request.
-    // 2. A body_writer, an asynchronous function (returning future<>) that
-    //    takes an output_stream and writes the response body into it.
-    // 3. An api_error, which is an error response that should be returned to
-    //    the client.
-    // The body_writer is used for streaming responses, where the response body
-    // is written in chunks to the output_stream. This allows for efficient
-    // handling of large responses without needing to allocate a large buffer
-    // in memory.
-    using body_writer = noncopyable_function<future<>(output_stream<char>&&)>;
-    using request_return_type = std::variant<std::string, body_writer, api_error>;
+    using request_return_type = std::variant<json::json_return_type, api_error>;
     stats _stats;
-    // The metric_groups object holds this stat object's metrics registered
-    // as long as the stats object is alive.
-    seastar::metrics::metric_groups _metrics;
     static constexpr auto ATTRS_COLUMN_NAME = ":attrs";
     static constexpr auto KEYSPACE_NAME_PREFIX = "alternator_";
     static constexpr std::string_view INTERNAL_TABLE_PREFIX = ".scylla.alternator.";
@@ -175,7 +182,6 @@ public:
              cdc::metadata& cdc_metadata,
              smp_service_group ssg,
              utils::updateable_value<uint32_t> default_timeout_in_ms);
-    ~executor();
 
     future<request_return_type> create_table(client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value request);
     future<request_return_type> describe_table(client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value request);
@@ -203,23 +209,26 @@ public:
     future<request_return_type> describe_continuous_backups(client_state& client_state, service_permit permit, rjson::value request);
 
     future<> start();
-    future<> stop();
+    future<> stop() {
+        // disconnect from the value source, but keep the value unchanged.
+        s_default_timeout_in_ms = utils::updateable_value<uint32_t>{s_default_timeout_in_ms()};
+        return make_ready_future<>();
+    }
 
     static sstring table_name(const schema&);
     static db::timeout_clock::time_point default_timeout();
 private:
     static thread_local utils::updateable_value<uint32_t> s_default_timeout_in_ms;
 public:
-    static schema_ptr find_table(service::storage_proxy&, std::string_view table_name);
     static schema_ptr find_table(service::storage_proxy&, const rjson::value& request);
 
 private:
     friend class rmw_operation;
 
-    static void describe_key_schema(rjson::value& parent, const schema&, std::unordered_map<std::string,std::string> * = nullptr, const std::map<sstring, sstring> *tags = nullptr);
+    static void describe_key_schema(rjson::value& parent, const schema&, std::unordered_map<std::string,std::string> * = nullptr);
 
 public:
-    static void describe_key_schema(rjson::value& parent, const schema& schema, std::unordered_map<std::string,std::string>&, const std::map<sstring, sstring> *tags = nullptr);
+    static void describe_key_schema(rjson::value& parent, const schema& schema, std::unordered_map<std::string,std::string>&);
 
     static std::optional<rjson::value> describe_single_item(schema_ptr,
         const query::partition_slice&,
@@ -228,15 +237,11 @@ public:
         const std::optional<attrs_to_get>&,
         uint64_t* = nullptr);
 
-    // Converts a multi-row selection result to JSON compatible with DynamoDB.
-    // For each row, this method calls item_callback, which takes the size of
-    // the item as the parameter.
     static future<std::vector<rjson::value>> describe_multi_item(schema_ptr schema,
         const query::partition_slice&& slice,
         shared_ptr<cql3::selection::selection> selection,
         foreign_ptr<lw_shared_ptr<query::result>> query_result,
-        shared_ptr<const std::optional<attrs_to_get>> attrs_to_get,
-        noncopyable_function<void(uint64_t)> item_callback = {});
+        shared_ptr<const std::optional<attrs_to_get>> attrs_to_get);
 
     static void describe_single_item(const cql3::selection::selection&,
         const std::vector<managed_bytes_opt>&,
@@ -245,7 +250,7 @@ public:
         uint64_t* item_length_in_bytes = nullptr,
         bool = false);
 
-    static bool add_stream_options(const rjson::value& stream_spec, schema_builder&, service::storage_proxy& sp);
+    static void add_stream_options(const rjson::value& stream_spec, schema_builder&, service::storage_proxy& sp);
     static void supplement_table_info(rjson::value& descr, const schema& schema, service::storage_proxy& sp);
     static void supplement_table_stream_info(rjson::value& descr, const schema& schema, const service::storage_proxy& sp);
 };
@@ -266,13 +271,4 @@ bool is_big(const rjson::value& val, int big_size = 100'000);
 // appropriate user-readable api_error::access_denied is thrown.
 future<> verify_permission(bool enforce_authorization, const service::client_state&, const schema_ptr&, auth::permission);
 
-/**
- * Make return type for serializing the object "streamed",
- * i.e. direct to HTTP output stream. Note: only useful for
- * (very) large objects as there are overhead issues with this
- * as well, but for massive lists of return objects this can
- * help avoid large allocations/many re-allocs
- */
-executor::body_writer make_streamed(rjson::value&&);
-
 }

alternator/expressions.cc

@@ -165,9 +165,7 @@ static std::optional<std::string> resolve_path_component(const std::string& colu
                     fmt::format("ExpressionAttributeNames missing entry '{}' required by expression", column_name));
         }
         used_attribute_names.emplace(column_name);
-        auto result = std::string(rjson::to_string_view(*value));
-        validate_attr_name_length("", result.size(), false, "ExpressionAttributeNames contains invalid value: ");
-        return result;
+        return std::string(rjson::to_string_view(*value));
     }
     return std::nullopt;
 }
@@ -739,26 +737,6 @@ rjson::value calculate_value(const parsed::set_rhs& rhs,
     return rjson::null_value();
 }
 
-void validate_attr_name_length(std::string_view supplementary_context, size_t attr_name_length, bool is_key, std::string_view error_msg_prefix) {
-    constexpr const size_t DYNAMODB_KEY_ATTR_NAME_SIZE_MAX = 255;
-    constexpr const size_t DYNAMODB_NONKEY_ATTR_NAME_SIZE_MAX = 65535;
-
-    const size_t max_length = is_key ? DYNAMODB_KEY_ATTR_NAME_SIZE_MAX : DYNAMODB_NONKEY_ATTR_NAME_SIZE_MAX;
-    if (attr_name_length > max_length) {
-        std::string error_msg;
-        if (!error_msg_prefix.empty()) {
-            error_msg += error_msg_prefix;
-        }
-        if (!supplementary_context.empty()) {
-            error_msg += "in ";
-            error_msg += supplementary_context;
-            error_msg += " - ";
-        }
-        error_msg += fmt::format("Attribute name is too large, must be less than {} bytes", std::to_string(max_length + 1));
-        throw api_error::validation(error_msg);
-    }
-}
-
 } // namespace alternator
 
 auto fmt::formatter<alternator::parsed::path>::format(const alternator::parsed::path& p, fmt::format_context& ctx) const

alternator/expressions.g

@@ -91,18 +91,6 @@ options {
         throw expressions_syntax_error(format("{} at char {}", err,
             ex->get_charPositionInLine()));
     }
-
-    // ANTLR3 tries to recover missing tokens - it tries to finish parsing
-    // and create valid objects, as if the missing token was there.
-    // But it has a bug and leaks these tokens.
-    // We override offending method and handle abandoned pointers.
-    std::vector<std::unique_ptr<TokenType>> _missing_tokens;
-    TokenType* getMissingSymbol(IntStreamType* istream, ExceptionBaseType* e,
-                                ANTLR_UINT32 expectedTokenType, BitsetListType* follow) {
-        auto token = BaseType::getMissingSymbol(istream, e, expectedTokenType, follow);
-        _missing_tokens.emplace_back(token);
-        return token;
-    }
 }
 @lexer::context {
     void displayRecognitionError(ANTLR_UINT8** token_names, ExceptionBaseType* ex) {
@@ -196,13 +184,7 @@ path_component: NAME | NAMEREF;
 path returns [parsed::path p]:
     root=path_component           { $p.set_root($root.text); }
     (   '.' name=path_component   { $p.add_dot($name.text); }
-      | '[' INTEGER ']'           {
-                try {
-                    $p.add_index(std::stoi($INTEGER.text));
-                } catch(std::out_of_range&) {
-                    throw expressions_syntax_error("list index out of integer range");
-                }
-            }
+      | '[' INTEGER ']'           { $p.add_index(std::stoi($INTEGER.text)); }
     )*;
 
 /* See comment above why the "depth" counter was needed here */
@@ -248,7 +230,7 @@ update_expression_clause returns [parsed::update_expression e]:
 // Note the "EOF" token at the end of the update expression. We want to the
 //  parser to match the entire string given to it - not just its beginning!
 update_expression returns [parsed::update_expression e]:
-    (update_expression_clause { e.append($update_expression_clause.e); })+ EOF;
+    (update_expression_clause { e.append($update_expression_clause.e); })* EOF;
 
 projection_expression returns [std::vector<parsed::path> v]:
     p=path      { $v.push_back(std::move($p.p)); }
@@ -275,13 +257,6 @@ primitive_condition returns [parsed::primitive_condition c]:
          (',' v=value[0] { $c.add_value(std::move($v.v)); })*
          ')'
       )?
-      {
-          // Post-parse check to reject non-function single values
-          if ($c._op == parsed::primitive_condition::type::VALUE &&
-              !$c._values.front().is_func()) {
-              throw expressions_syntax_error("Single value must be a function");
-          }
-      }
     ;
 
 // The following rules for parsing boolean expressions are verbose and

alternator/expressions.hh

@@ -18,8 +18,6 @@
 
 #include "expressions_types.hh"
 #include "utils/rjson.hh"
-#include "utils/updateable_value.hh"
-#include "stats.hh"
 
 namespace alternator {
 
@@ -28,26 +26,6 @@ public:
     using runtime_error::runtime_error;
 };
 
-namespace parsed {
-class expression_cache_impl;
-class expression_cache {
-    std::unique_ptr<expression_cache_impl> _impl;
-public:
-    struct config {
-        utils::updateable_value<uint32_t> max_cache_entries;
-    };
-    expression_cache(config cfg, stats& stats);
-    ~expression_cache();
-    // stop background tasks, if any
-    future<> stop();
-
-    update_expression parse_update_expression(std::string_view query);
-    std::vector<path> parse_projection_expression(std::string_view query);
-    condition_expression parse_condition_expression(std::string_view query, const char* caller);
-};
-} // namespace parsed
-
-// Preferably use parsed::expression_cache instance instead of this free functions.
 parsed::update_expression parse_update_expression(std::string_view query);
 std::vector<parsed::path> parse_projection_expression(std::string_view query);
 parsed::condition_expression parse_condition_expression(std::string_view query, const char* caller);
@@ -113,7 +91,5 @@ rjson::value calculate_value(const parsed::value& v,
 rjson::value calculate_value(const parsed::set_rhs& rhs,
         const rjson::value* previous_item);
 
-void validate_attr_name_length(std::string_view supplementary_context, size_t attr_name_length, bool is_key, std::string_view error_msg_prefix = {});
-
 
 } /* namespace alternator */

alternator/expressions_types.hh

@@ -209,7 +209,9 @@ public:
 //    function is supported).
 // 2. Ternary operator - v1 BETWEEN v2 and v3 (means v1 >= v2 AND v1 <= v3).
 // 3. N-ary operator - v1 IN ( v2, v3, ... )
-// 4. A single function call (attribute_exists etc.).
+// 4. A single function call (attribute_exists etc.). The parser actually
+//    accepts a more general "value" here but later stages reject a value
+//    which is not a function call (because DynamoDB does it too).
 class primitive_condition {
 public:
     enum class type {

alternator/extract_from_attrs.hh

@@ -13,7 +13,7 @@
 
 #include "utils/rjson.hh"
 #include "serialization.hh"
-#include "schema/column_computation.hh"
+#include "column_computation.hh"
 #include "db/view/regular_column_transformation.hh"
 
 namespace alternator {

alternator/parsed_expression_cache.cc

@@ -1,109 +0,0 @@
-/*
- * Copyright 2025-present ScyllaDB
- */
-
-/*
- * SPDX-License-Identifier: LicenseRef-ScyllaDB-Source-Available-1.0
- */
-
-#include "expressions.hh"
-#include "utils/log.hh"
-
-#include "utils/lru_string_map.hh"
-#include <variant>
-
-static logging::logger logger_("parsed-expression-cache");
-
-namespace alternator::parsed {
-
-struct expression_cache_impl {
-    stats& _stats;
-
-    using cached_expressions_types = std::variant<
-        update_expression,
-        condition_expression,
-        std::vector<path>
-    >;
-    sized_lru_string_map<cached_expressions_types> _cached_entries;
-    utils::observable<uint32_t>::observer _max_cache_entries_observer;
-
-    expression_cache_impl(expression_cache::config cfg, stats& stats);
-
-    // to define the specialized return type of `get_or_create()`
-    template <typename Func, typename... Args>
-    using ParseResult = std::invoke_result_t<Func, std::string_view, Args...>;
-
-    // Caching layer for parsed expressions
-    // The expression type is determined by the type of the parsing function passed as a parameter,
-    // and the return type is exactly the same as the return type of this parsing function.
-    // StatsType is used only to update appropriate statistics - currently it is aligned with the expression type,
-    // but it could be extended in the future if needed, e.g. split per operation.
-    template <stats::expression_types StatsType, typename Func, typename... Args>
-    ParseResult<Func, Args...> get_or_create(std::string_view query, Func&& parse_func, Args&&... other_args) {
-        if (_cached_entries.disabled()) {
-            return parse_func(query, std::forward<Args>(other_args)...);
-        }
-        if (!_cached_entries.sanity_check()) {
-            _stats.expression_cache.requests[StatsType].misses++;
-            return parse_func(query, std::forward<Args>(other_args)...);
-        }
-        auto value = _cached_entries.find(query);
-        if (value) {
-            logger_.trace("Cache hit for query: {}", query);
-            _stats.expression_cache.requests[StatsType].hits++;
-            try {
-                return std::get<ParseResult<Func, Args...>>(value->get());
-            } catch (const std::bad_variant_access&) {
-                // User can reach this code, by sending the same query string as a different expression type.
-                // In practice valid queries are different enough to not collide.
-                // Entries in cache are only valid queries.
-                // This request will fail at parsing below.
-                // If, by any chance this is a valid query, it will be updated below with the new value.
-                logger_.trace("Cache hit for '{}', but type mismatch.", query);
-                _stats.expression_cache.requests[StatsType].hits--;
-            }
-        } else {
-            logger_.trace("Cache miss for query: {}", query);
-        }
-        ParseResult<Func, Args...> expr = parse_func(query, std::forward<Args>(other_args)...);
-        // Invalid query will throw here ^
-
-        _stats.expression_cache.requests[StatsType].misses++;
-        if (value) [[unlikely]] {
-            value->get() = cached_expressions_types{expr};
-        } else {
-            _cached_entries.insert(query, cached_expressions_types{expr});
-        }
-        return expr;
-    }
-};
-
-expression_cache_impl::expression_cache_impl(expression_cache::config cfg, stats& stats) : 
-    _stats(stats), _cached_entries(logger_, _stats.expression_cache.evictions),
-    _max_cache_entries_observer(cfg.max_cache_entries.observe([this] (uint32_t max_value) {
-        _cached_entries.set_max_size(max_value);
-    })) {
-    _cached_entries.set_max_size(cfg.max_cache_entries());
-}
-
-expression_cache::expression_cache(expression_cache::config cfg, stats& stats) : 
-    _impl(std::make_unique<expression_cache_impl>(std::move(cfg), stats)) {
-}
-expression_cache::~expression_cache() = default;
-future<> expression_cache::stop() {
-    return _impl->_cached_entries.stop();
-}
-
-update_expression expression_cache::parse_update_expression(std::string_view query) {
-    return _impl->get_or_create<stats::expression_types::UPDATE_EXPRESSION>(query, alternator::parse_update_expression);
-}
-
-std::vector<path> expression_cache::parse_projection_expression(std::string_view query) {
-    return _impl->get_or_create<stats::expression_types::PROJECTION_EXPRESSION>(query, alternator::parse_projection_expression);
-}
-
-condition_expression expression_cache::parse_condition_expression(std::string_view query, const char* caller) {
-    return _impl->get_or_create<stats::expression_types::CONDITION_EXPRESSION>(query, alternator::parse_condition_expression, caller);
-}
-
-} // namespace alternator::parsed

alternator/rmw_operation.hh

@@ -8,16 +8,13 @@
 
 #pragma once
 
-#include "cdc/cdc_options.hh"
-#include "cdc/log.hh"
 #include "seastarx.hh"
 #include "service/paxos/cas_request.hh"
-#include "service/cas_shard.hh"
 #include "utils/rjson.hh"
 #include "consumed_capacity.hh"
 #include "executor.hh"
 #include "tracing/trace_state.hh"
-#include "keys/keys.hh"
+#include "keys.hh"
 
 namespace alternator {
 
@@ -58,7 +55,7 @@ public:
     static write_isolation get_write_isolation_for_schema(schema_ptr schema);
 
     static write_isolation default_write_isolation;
-
+public:
     static void set_default_write_isolation(std::string_view mode);
 
 protected:
@@ -109,27 +106,21 @@ public:
     // violating this). We mark apply() "const" to let the compiler validate
     // this for us. The output-only field _return_attributes is marked
     // "mutable" above so that apply() can still write to it.
-    virtual std::optional<mutation> apply(std::unique_ptr<rjson::value> previous_item, api::timestamp_type ts, cdc::per_request_options& cdc_opts) const = 0;
+    virtual std::optional<mutation> apply(std::unique_ptr<rjson::value> previous_item, api::timestamp_type ts) const = 0;
     // Convert the above apply() into the signature needed by cas_request:
-    virtual std::optional<mutation> apply(foreign_ptr<lw_shared_ptr<query::result>> qr, const query::partition_slice& slice, api::timestamp_type ts, cdc::per_request_options& cdc_opts) override;
+    virtual std::optional<mutation> apply(foreign_ptr<lw_shared_ptr<query::result>> qr, const query::partition_slice& slice, api::timestamp_type ts) override;
     virtual ~rmw_operation() = default;
-    const wcu_consumed_capacity_counter& consumed_capacity() const noexcept { return _consumed_capacity; }
     schema_ptr schema() const { return _schema; }
     const rjson::value& request() const { return _request; }
     rjson::value&& move_request() && { return std::move(_request); }
     future<executor::request_return_type> execute(service::storage_proxy& proxy,
-            std::optional<service::cas_shard> cas_shard,
             service::client_state& client_state,
             tracing::trace_state_ptr trace_state,
             service_permit permit,
             bool needs_read_before_write,
-            stats& global_stats,
-            stats& per_table_stats,
+            stats& stats,
             uint64_t& wcu_total);
-    std::optional<service::cas_shard> shard_for_execute(bool needs_read_before_write);
-
-private:
-    inline bool should_fill_preimage() const { return _schema->cdc_options().enabled(); }
+    std::optional<shard_id> shard_for_execute(bool needs_read_before_write);
 };
 
 } // namespace alternator

alternator/serialization.cc

@@ -11,8 +11,8 @@
 #include "utils/log.hh"
 #include "serialization.hh"
 #include "error.hh"
-#include "types/concrete_types.hh"
-#include "types/json_utils.hh"
+#include "concrete_types.hh"
+#include "cql3/type_json.hh"
 #include "mutation/position_in_partition.hh"
 
 static logging::logger slogger("alternator-serialization");

alternator/serialization.hh

@@ -13,7 +13,7 @@
 #include <optional>
 #include "types/types.hh"
 #include "schema/schema_fwd.hh"
-#include "keys/keys.hh"
+#include "keys.hh"
 #include "utils/rjson.hh"
 #include "utils/big_decimal.hh"
 

alternator/server.cc

@@ -13,6 +13,7 @@
 #include <seastar/http/function_handlers.hh>
 #include <seastar/http/short_streams.hh>
 #include <seastar/core/coroutine.hh>
+#include <seastar/json/json_elements.hh>
 #include <seastar/util/defer.hh>
 #include <seastar/util/short_streams.hh>
 #include "seastarx.hh"
@@ -30,7 +31,6 @@
 #include "gms/gossiper.hh"
 #include "utils/overloaded_functor.hh"
 #include "utils/aws_sigv4.hh"
-#include "client_data.hh"
 
 static logging::logger slogger("alternator-server");
 
@@ -100,13 +100,6 @@ static void handle_CORS(const request& req, reply& rep, bool preflight) {
 // the user directly. Other exceptions are unexpected, and reported as
 // Internal Server Error.
 class api_handler : public handler_base {
-    // Although the the DynamoDB API responses are JSON, additional
-    // conventions apply to these responses. For this reason, DynamoDB uses
-    // the content type "application/x-amz-json-1.0" instead of the standard
-    // "application/json". Some other AWS services use later versions instead
-    // of "1.0", but DynamoDB currently uses "1.0". Note that this content
-    // type applies to all replies, both success and error.
-    static constexpr const char* REPLY_CONTENT_TYPE = "application/x-amz-json-1.0";
 public:
     api_handler(const std::function<future<executor::request_return_type>(std::unique_ptr<request> req)>& _handle) : _f_handle(
          [this, _handle](std::unique_ptr<request> req, std::unique_ptr<reply> rep) {
@@ -131,19 +124,22 @@ public:
              }
              auto res = resf.get();
              std::visit(overloaded_functor {
-                [&] (std::string&& str) {
-                    // Note that despite the move, there is a copy here -
-                    // as str is std::string and rep->_content is sstring.
-                    rep->_content = std::move(str);
-                    rep->set_content_type(REPLY_CONTENT_TYPE);
-                },
-                [&] (executor::body_writer&& body_writer) {
-                    rep->write_body(REPLY_CONTENT_TYPE, std::move(body_writer));
-                },
-                [&] (const api_error& err) {
-                    generate_error_reply(*rep, err);
-                }
-             }, std::move(res));
+                 [&] (const json::json_return_type& json_return_value) {
+                     slogger.trace("api_handler success case");
+                     if (json_return_value._body_writer) {
+                         // Unfortunately, write_body() forces us to choose
+                         // from a fixed and irrelevant list of "mime-types"
+                         // at this point. But we'll override it with the
+                         // one (application/x-amz-json-1.0) below.
+                         rep->write_body("json", std::move(json_return_value._body_writer));
+                     } else {
+                         rep->_content += json_return_value._res;
+                     }
+                 },
+                 [&] (const api_error& err) {
+                     generate_error_reply(*rep, err);
+                 }
+             }, res);
 
              return make_ready_future<std::unique_ptr<reply>>(std::move(rep));
          });
@@ -155,6 +151,7 @@ public:
         handle_CORS(*req, *rep, false);
         return _f_handle(std::move(req), std::move(rep)).then(
                 [](std::unique_ptr<reply> rep) {
+                    rep->set_mime_type("application/x-amz-json-1.0");
                     rep->done();
                     return make_ready_future<std::unique_ptr<reply>>(std::move(rep));
                 });
@@ -170,7 +167,6 @@ protected:
         rjson::add(results, "message", err._msg);
         rep._content = rjson::print(std::move(results));
         rep._status = err._http_code;
-        rep.set_content_type(REPLY_CONTENT_TYPE);
         slogger.trace("api_handler error case: {}", rep._content);
     }
 
@@ -232,8 +228,9 @@ protected:
         // If the rack does not exist, we return an empty list - not an error.
         sstring query_rack = req->get_query_param("rack");
         for (auto& id : local_dc_nodes) {
+            auto ip = _gossiper.get_address_map().get(id);
             if (!query_rack.empty()) {
-                auto rack = _gossiper.get_application_state_value(id, gms::application_state::RACK);
+                auto rack = _gossiper.get_application_state_value(ip, gms::application_state::RACK);
                 if (rack != query_rack) {
                     continue;
                 }
@@ -241,10 +238,10 @@ protected:
             // Note that it's not enough for the node to be is_alive() - a
             // node joining the cluster is also "alive" but not responsive to
             // requests. We alive *and* normal. See #19694, #21538.
-            if (_gossiper.is_alive(id) && _gossiper.is_normal(id)) {
+            if (_gossiper.is_alive(ip) && _gossiper.is_normal(ip)) {
                 // Use the gossiped broadcast_rpc_address if available instead
                 // of the internal IP address "ip". See discussion in #18711.
-                rjson::push_back(results, rjson::from_string(_gossiper.get_rpc_address(id)));
+                rjson::push_back(results, rjson::from_string(_gossiper.get_rpc_address(ip)));
             }
         }
         rep->set_status(reply::status_type::ok);
@@ -378,82 +375,35 @@ static tracing::trace_state_ptr create_tracing_session(tracing::tracing& tracing
     return tracing_instance.create_session(tracing::trace_type::QUERY, props);
 }
 
-// A helper class to represent a potentially truncated view of a chunked_content.
-// If the content is short enough and single chunked, it just holds a view into the content.
-// Otherwise it will be copied into an internal buffer, possibly truncated (depending on maximum allowed size passed in),
-// and the view will point into that buffer.
-// `as_view()` method will return the view.
-// `take_as_sstring()` will either move out the internal buffer (if any), or create a new sstring from the view.
-// You should consider `as_view()` valid as long both the original chunked_content and the truncated_content object are alive.
-class truncated_content {
-    std::string_view _view;
-    sstring _content_maybe;
-
-    void copy_from_content(const chunked_content& content) {
-        size_t offset = 0;
-        for(auto &tmp : content) {
-            size_t to_copy = std::min(tmp.size(), _content_maybe.size() - offset);
-            std::copy(tmp.get(), tmp.get() + to_copy, _content_maybe.data() + offset);
-            offset += to_copy;
-            if (offset >= _content_maybe.size()) {
-                break;
-            }
-        }
+// truncated_content_view() prints a potentially long chunked_content for
+// debugging purposes. In the common case when the content is not excessively
+// long, it just returns a view into the given content, without any copying.
+// But when the content is very long, it is truncated after some arbitrary
+// max_len (or one chunk, whichever comes first), with "<truncated>" added at
+// the end. To do this modification to the string, we need to create a new
+// std::string, so the caller must pass us a reference to one, "buf", where
+// we can store the content. The returned view is only alive for as long this
+// buf is kept alive.
+static std::string_view truncated_content_view(const chunked_content& content, std::string& buf) {
+    constexpr size_t max_len = 1024;
+    if (content.empty()) {
+        return std::string_view();
+    } else if (content.size() == 1 && content.begin()->size() <= max_len) {
+        return std::string_view(content.begin()->get(), content.begin()->size());
+    } else {
+        buf = std::string(content.begin()->get(), std::min(content.begin()->size(), max_len)) + "<truncated>";
+        return std::string_view(buf);
     }
-public:
-    truncated_content(const chunked_content& content, size_t max_len = std::numeric_limits<size_t>::max()) {
-        if (content.empty()) return;
-        if (content.size() == 1 && content.begin()->size() <= max_len) {
-            _view = std::string_view(content.begin()->get(), content.begin()->size());
-            return;
-        }
-
-        constexpr std::string_view truncated_text = "<truncated>";
-        size_t content_size = 0;
-        for(auto &tmp : content) {
-            content_size += tmp.size();
-        }
-        if (content_size <= max_len) {
-            _content_maybe = sstring{ sstring::initialized_later{}, content_size };
-            copy_from_content(content);
-        }
-        else {
-            _content_maybe = sstring{ sstring::initialized_later{}, max_len + truncated_text.size() };
-            copy_from_content(content);
-            std::copy(truncated_text.begin(), truncated_text.end(), _content_maybe.data() + _content_maybe.size() - truncated_text.size());
-        }
-        _view = std::string_view(_content_maybe);
-    }
-
-    std::string_view as_view() const { return _view; }
-    sstring take_as_sstring() && {
-        if (_content_maybe.empty() && !_view.empty()) {
-            return sstring{_view};
-        }
-        return std::move(_content_maybe);
-    }
-};
-
-// `truncated_content_view` will produce an object representing a view to a passed content
-// possibly truncated at some length. The value returned is used in two ways:
-// - to print it in logs (use `as_view()` method for this)
-// - to pass it to tracing object, where it will be stored and used later
-//   (use `take_as_sstring()` method as this produces a copy in form of a sstring)
-// `truncated_content` delays constructing `sstring` object until it's actually needed.
-// `truncated_content` is valid as long as passed `content` is alive.
-// if the content is truncated, `<truncated>` will be appended at the maximum size limit
-// and total size will be `max_users_query_size_in_trace_output() + strlen("<truncated>")`.
-static truncated_content truncated_content_view(const chunked_content& content, size_t max_size) {
-    return truncated_content{content, max_size};
 }
 
-static tracing::trace_state_ptr maybe_trace_query(service::client_state& client_state, std::string_view username, std::string_view op, const chunked_content& query, size_t max_users_query_size_in_trace_output) {
+static tracing::trace_state_ptr maybe_trace_query(service::client_state& client_state, std::string_view username, std::string_view op, const chunked_content& query) {
     tracing::trace_state_ptr trace_state;
     tracing::tracing& tracing_instance = tracing::tracing::get_local_tracing_instance();
     if (tracing_instance.trace_next_query() || tracing_instance.slow_query_tracing_enabled()) {
         trace_state = create_tracing_session(tracing_instance);
+        std::string buf;
         tracing::add_session_param(trace_state, "alternator_op", op);
-        tracing::add_query(trace_state, truncated_content_view(query, max_users_query_size_in_trace_output).take_as_sstring());
+        tracing::add_query(trace_state, truncated_content_view(query, buf));
         tracing::begin(trace_state, seastar::format("Alternator {}", op), client_state.get_client_address());
         if (!username.empty()) {
             tracing::set_username(trace_state, auth::authenticated_user(username));
@@ -462,92 +412,30 @@ static tracing::trace_state_ptr maybe_trace_query(service::client_state& client_
     return trace_state;
 }
 
-// This read_entire_stream() is similar to Seastar's read_entire_stream()
-// which reads the given content_stream until its end into non-contiguous
-// memory. The difference is that this implementation takes an extra length
-// limit, and throws an error if we read more than this limit.
-// This length-limited variant would not have been needed if Seastar's HTTP
-// server's set_content_length_limit() worked in every case, but unfortunately
-// it does not - it only works if the request has a Content-Length header (see
-// issue #8196). In contrast this function can limit the request's length no
-// matter how it's encoded. We need this limit to protect Alternator from
-// oversized requests that can deplete memory.
-static future<chunked_content>
-read_entire_stream(input_stream<char>& inp, size_t length_limit) {
-    chunked_content ret;
-    // We try to read length_limit + 1 bytes, so that we can throw an
-    // exception if we managed to read more than length_limit.
-    ssize_t remain = length_limit + 1;
-    do {
-        temporary_buffer<char> buf = co_await inp.read_up_to(remain);
-        if (buf.empty()) {
-            break;
-        }
-        remain -= buf.size();
-        ret.push_back(std::move(buf));
-    } while (remain > 0);
-    // If we read the full length_limit + 1 bytes, we went over the limit:
-    if (remain <= 0) {
-        // By throwing here an error, we may send a reply (the error message)
-        // without having read the full request body. Seastar's httpd will
-        // realize that we have not read the entire content stream, and
-        // correctly mark the connection unreusable, i.e., close it.
-        // This means we are currently exposed to issue #12166 caused by
-        // Seastar issue 1325), where the client may get an RST instead of
-        // a FIN, and may rarely get a "Connection reset by peer" before
-        // reading the error we send.
-        throw api_error::payload_too_large(fmt::format("Request content length limit of {} bytes exceeded", length_limit));
-    }
-    co_return ret;
-}
-
 future<executor::request_return_type> server::handle_api_request(std::unique_ptr<request> req) {
     _executor._stats.total_operations++;
     sstring target = req->get_header("X-Amz-Target");
     // target is DynamoDB API version followed by a dot '.' and operation type (e.g. CreateTable)
     auto dot = target.find('.');
     std::string_view op = (dot == sstring::npos) ? std::string_view() : std::string_view(target).substr(dot+1);
-    if (req->content_length > request_content_length_limit) {
-        // If we have a Content-Length header and know the request will be too
-        // long, we don't need to wait for read_entire_stream() below to
-        // discover it. And we definitely mustn't try to get_units() below for
-        // for such a size.
-        co_return api_error::payload_too_large(fmt::format("Request content length limit of {} bytes exceeded", request_content_length_limit));
-    }
     // JSON parsing can allocate up to roughly 2x the size of the raw
     // document, + a couple of bytes for maintenance.
-    // If the Content-Length of the request is not available, we assume
-    // the largest possible request (request_content_length_limit, i.e., 16 MB)
-    // and after reading the request we return_units() the excess.
-    size_t mem_estimate = (req->content_length ? req->content_length : request_content_length_limit) * 2 + 8000;
+    // TODO: consider the case where req->content_length is missing. Maybe
+    // we need to take the content_length_limit and return some of the units
+    // when we finish read_content_and_verify_signature?
+    size_t mem_estimate = req->content_length * 2 + 8000;
     auto units_fut = get_units(*_memory_limiter, mem_estimate);
     if (_memory_limiter->waiters()) {
         ++_executor._stats.requests_blocked_memory;
     }
     auto units = co_await std::move(units_fut);
     SCYLLA_ASSERT(req->content_stream);
-    chunked_content content = co_await read_entire_stream(*req->content_stream, request_content_length_limit);
-    // If the request had no Content-Length, we reserved too many units
-    // so need to return some
-    if (req->content_length == 0) {
-        size_t content_length = 0;
-        for (const auto& chunk : content) {
-            content_length += chunk.size();
-        }
-        size_t new_mem_estimate = content_length * 2 + 8000;
-        units.return_units(mem_estimate - new_mem_estimate);
-    }
+    chunked_content content = co_await util::read_entire_stream(*req->content_stream);
     auto username = co_await verify_signature(*req, content);
-    // As long as the system_clients_entry object is alive, this request will
-    // be visible in the "system.clients" virtual table. When requested, this
-    // entry will be formatted by server::ongoing_request::make_client_data().
-    auto system_clients_entry = _ongoing_requests.emplace(
-        req->get_client_address(), req->get_header("User-Agent"),
-        username, current_scheduling_group(),
-        req->get_protocol_name() == "https");
 
     if (slogger.is_enabled(log_level::trace)) {
-        slogger.trace("Request: {} {} {}", op, truncated_content_view(content, _max_users_query_size_in_trace_output).as_view(), req->_headers);
+        std::string buf;
+        slogger.trace("Request: {} {} {}", op, truncated_content_view(content, buf), req->_headers);
     }
     auto callback_it = _callbacks.find(op);
     if (callback_it == _callbacks.end()) {
@@ -567,7 +455,7 @@ future<executor::request_return_type> server::handle_api_request(std::unique_ptr
     }
     co_await client_state.maybe_update_per_service_level_params();
 
-    tracing::trace_state_ptr trace_state = maybe_trace_query(client_state, username, op, content, _max_users_query_size_in_trace_output.get());
+    tracing::trace_state_ptr trace_state = maybe_trace_query(client_state, username, op, content);
     tracing::trace(trace_state, "{}", op);
 
     auto user = client_state.user();
@@ -575,9 +463,6 @@ future<executor::request_return_type> server::handle_api_request(std::unique_ptr
             client_state = std::move(client_state), trace_state = std::move(trace_state),
             units = std::move(units), req = std::move(req)] () mutable -> future<executor::request_return_type> {
                 rjson::value json_request = co_await _json_parser.parse(std::move(content));
-                if (!json_request.IsObject()) {
-                    co_return api_error::validation("Request content must be an object");
-                }
                 co_return co_await callback(_executor, client_state, trace_state,
                     make_service_permit(std::move(units)), std::move(json_request), std::move(req));
     };
@@ -619,9 +504,8 @@ server::server(executor& exec, service::storage_proxy& proxy, gms::gossiper& gos
         , _sl_controller(sl_controller)
         , _key_cache(1024, 1min, slogger)
         , _enforce_authorization(false)
-        , _max_users_query_size_in_trace_output(1024)
         , _enabled_servers{}
-        , _pending_requests("alternator::server::pending_requests")
+        , _pending_requests{}
         , _timeout_config(_proxy.data_dictionary().get_config())
       , _callbacks{
         {"CreateTable", [] (executor& e, executor::client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value json_request, std::unique_ptr<request> req) {
@@ -700,12 +584,10 @@ server::server(executor& exec, service::storage_proxy& proxy, gms::gossiper& gos
 }
 
 future<> server::init(net::inet_address addr, std::optional<uint16_t> port, std::optional<uint16_t> https_port, std::optional<tls::credentials_builder> creds,
-        utils::updateable_value<bool> enforce_authorization, utils::updateable_value<uint64_t> max_users_query_size_in_trace_output,
-        semaphore* memory_limiter, utils::updateable_value<uint32_t> max_concurrent_requests) {
+        utils::updateable_value<bool> enforce_authorization, semaphore* memory_limiter, utils::updateable_value<uint32_t> max_concurrent_requests) {
     _memory_limiter = memory_limiter;
     _enforce_authorization = std::move(enforce_authorization);
     _max_concurrent_requests = std::move(max_concurrent_requests);
-    _max_users_query_size_in_trace_output = std::move(max_users_query_size_in_trace_output);
     if (!port && !https_port) {
         return make_exception_future<>(std::runtime_error("Either regular port or TLS port"
                 " must be specified in order to init an alternator HTTP server instance"));
@@ -715,12 +597,14 @@ future<> server::init(net::inet_address addr, std::optional<uint16_t> port, std:
 
         if (port) {
             set_routes(_http_server._routes);
+            _http_server.set_content_length_limit(server::content_length_limit);
             _http_server.set_content_streaming(true);
             _http_server.listen(socket_address{addr, *port}).get();
             _enabled_servers.push_back(std::ref(_http_server));
         }
         if (https_port) {
             set_routes(_https_server._routes);
+            _https_server.set_content_length_limit(server::content_length_limit);
             _https_server.set_content_streaming(true);
 
             if (this_shard_id() == 0) {
@@ -795,37 +679,6 @@ future<> server::json_parser::stop() {
     return std::move(_run_parse_json_thread);
 }
 
-// Convert an entry in the server's list of ongoing Alternator requests
-// (_ongoing_requests) into a client_data object. This client_data object
-// will then be used to produce a row for the "system.clients" virtual table.
-client_data server::ongoing_request::make_client_data() const {
-    client_data cd;
-    cd.ct = client_type::alternator;
-    cd.ip = _client_address.addr();
-    cd.port = _client_address.port();
-    cd.shard_id = this_shard_id();
-    cd.connection_stage = client_connection_stage::established;
-    cd.username = _username;
-    cd.scheduling_group_name = _scheduling_group.name();
-    cd.ssl_enabled = _is_https;
-    // For now, we save the full User-Agent header as the "driver name"
-    // and keep "driver_version" unset.
-    cd.driver_name = _user_agent;
-    // Leave "protocol_version" unset, it has no meaning in Alternator.
-    // Leave "hostname", "ssl_protocol" and "ssl_cipher_suite" unset.
-    // As reported in issue #9216, we never set these fields in CQL
-    // either (see cql_server::connection::make_client_data()).
-    return cd;
-}
-
-future<utils::chunked_vector<client_data>> server::get_client_data() {
-    utils::chunked_vector<client_data> ret;
-    co_await _ongoing_requests.for_each_gently([&ret] (const ongoing_request& r) {
-        ret.emplace_back(r.make_client_data());
-    });
-    co_return ret;
-}
-
 const char* api_error::what() const noexcept {
     if (_what_string.empty()) {
         _what_string = fmt::format("{} {}: {}", std::to_underlying(_http_code), _type, _msg);

alternator/server.hh

@@ -9,7 +9,6 @@
 #pragma once
 
 #include "alternator/executor.hh"
-#include "utils/scoped_item_list.hh"
 #include <seastar/core/future.hh>
 #include <seastar/core/condition-variable.hh>
 #include <seastar/http/httpd.hh>
@@ -21,18 +20,12 @@
 #include "utils/updateable_value.hh"
 #include <seastar/core/units.hh>
 
-struct client_data;
-
 namespace alternator {
 
 using chunked_content = rjson::chunked_content;
 
 class server : public peering_sharded_service<server> {
-    // The maximum size of a request body that Alternator will accept,
-    // in bytes. This is a safety measure to prevent Alternator from
-    // running out of memory when a client sends a very large request.
-    // DynamoDB also has the same limit set to 16 MB.
-    static constexpr size_t request_content_length_limit = 16*MB;
+    static constexpr size_t content_length_limit = 16*MB;
     using alternator_callback = std::function<future<executor::request_return_type>(executor&, executor::client_state&,
             tracing::trace_state_ptr, service_permit, rjson::value, std::unique_ptr<http::request>)>;
     using alternator_callbacks_map = std::unordered_map<std::string_view, alternator_callback>;
@@ -47,9 +40,8 @@ class server : public peering_sharded_service<server> {
 
     key_cache _key_cache;
     utils::updateable_value<bool> _enforce_authorization;
-    utils::updateable_value<uint64_t> _max_users_query_size_in_trace_output;
     utils::small_vector<std::reference_wrapper<seastar::httpd::http_server>, 2> _enabled_servers;
-    named_gate _pending_requests;
+    gate _pending_requests;
     // In some places we will need a CQL updateable_timeout_config object even
     // though it isn't really relevant for Alternator which defines its own
     // timeouts separately. We can create this object only once.
@@ -82,31 +74,12 @@ class server : public peering_sharded_service<server> {
     };
     json_parser _json_parser;
 
-    // The server maintains a list of ongoing requests, that are being handled
-    // by handle_api_request(). It uses this list in get_client_data(), which
-    // is called when reading the "system.clients" virtual table.
-    struct ongoing_request {
-        socket_address _client_address;
-        sstring _user_agent;
-        sstring _username;
-        scheduling_group _scheduling_group;
-        bool _is_https;
-        client_data make_client_data() const;
-    };
-    utils::scoped_item_list<ongoing_request> _ongoing_requests;
-
 public:
     server(executor& executor, service::storage_proxy& proxy, gms::gossiper& gossiper, auth::service& service, qos::service_level_controller& sl_controller);
 
     future<> init(net::inet_address addr, std::optional<uint16_t> port, std::optional<uint16_t> https_port, std::optional<tls::credentials_builder> creds,
-            utils::updateable_value<bool> enforce_authorization, utils::updateable_value<uint64_t> max_users_query_size_in_trace_output,
-            semaphore* memory_limiter, utils::updateable_value<uint32_t> max_concurrent_requests);
+            utils::updateable_value<bool> enforce_authorization, semaphore* memory_limiter, utils::updateable_value<uint32_t> max_concurrent_requests);
     future<> stop();
-    // get_client_data() is called (on each shard separately) when the virtual
-    // table "system.clients" is read. It is expected to generate a list of
-    // clients connected to this server (on this shard). This function is
-    // called by alternator::controller::get_client_data().
-    future<utils::chunked_vector<client_data>> get_client_data();
 private:
     void set_routes(seastar::httpd::routes& r);
     // If verification succeeds, returns the authenticated user's username

alternator/stats.cc

@@ -14,58 +14,28 @@
 namespace alternator {
 
 const char* ALTERNATOR_METRICS = "alternator";
-static seastar::metrics::histogram estimated_histogram_to_metrics(const utils::estimated_histogram& histogram) {
-    seastar::metrics::histogram res;
-    res.buckets.resize(histogram.bucket_offsets.size());
-    uint64_t cumulative_count = 0;
-    res.sample_count = histogram._count;
-    res.sample_sum = histogram._sample_sum;
-    for (size_t i = 0; i < res.buckets.size(); i++) {
-        auto& v = res.buckets[i];
-        v.upper_bound = histogram.bucket_offsets[i];
-        cumulative_count += histogram.buckets[i];
-        v.count = cumulative_count;
-    }
-    return res;
-}
-
-static seastar::metrics::label column_family_label("cf");
-static seastar::metrics::label keyspace_label("ks");
-
-
-static void register_metrics_with_optional_table(seastar::metrics::metric_groups& metrics, const stats& stats, const sstring& ks, const sstring& table) {
 
+stats::stats() : api_operations{} {
     // Register the
     seastar::metrics::label op("op");
-    bool has_table = table.length();
-    std::vector<seastar::metrics::label> aggregate_labels;
-    std::vector<seastar::metrics::label_instance> labels = {alternator_label};
-    sstring group_name = (has_table)? "alternator_table" : "alternator";
-    if (has_table) {
-        labels.push_back(column_family_label(table));
-        labels.push_back(keyspace_label(ks));
-        aggregate_labels.push_back(seastar::metrics::shard_label);
-    }
-    metrics.add_group(group_name, {
-#define OPERATION(name, CamelCaseName) \
-                seastar::metrics::make_total_operations("operation", stats.api_operations.name, \
-                        seastar::metrics::description("number of operations via Alternator API"), labels)(basic_level)(op(CamelCaseName)).aggregate(aggregate_labels).set_skip_when_empty(),
-#define OPERATION_LATENCY(name, CamelCaseName) \
-		metrics.add_group(group_name, { \
-                seastar::metrics::make_histogram("op_latency", \
-                        seastar::metrics::description("Latency histogram of an operation via Alternator API"), labels, [&stats]{return to_metrics_histogram(stats.api_operations.name.histogram());})(op(CamelCaseName))(basic_level).aggregate({seastar::metrics::shard_label}).set_skip_when_empty()}); \
-            if (!has_table) {\
-            	metrics.add_group("alternator", { \
-				seastar::metrics::make_summary("op_latency_summary", \
-						                        seastar::metrics::description("Latency summary of an operation via Alternator API"), [&stats]{return to_metrics_summary(stats.api_operations.name.summary());})(op(CamelCaseName))(basic_level)(alternator_label).set_skip_when_empty()}); \
-            }
 
+    _metrics.add_group("alternator", {
+#define OPERATION(name, CamelCaseName) \
+                seastar::metrics::make_total_operations("operation", api_operations.name, \
+                        seastar::metrics::description("number of operations via Alternator API"), {op(CamelCaseName), alternator_label, basic_level}).set_skip_when_empty(),
+#define OPERATION_LATENCY(name, CamelCaseName) \
+                seastar::metrics::make_histogram("op_latency", \
+                        seastar::metrics::description("Latency histogram of an operation via Alternator API"), {op(CamelCaseName), alternator_label, basic_level}, [this]{return to_metrics_histogram(api_operations.name.histogram());}).aggregate({seastar::metrics::shard_label}).set_skip_when_empty(), \
+				seastar::metrics::make_summary("op_latency_summary", \
+						                        seastar::metrics::description("Latency summary of an operation via Alternator API"), [this]{return to_metrics_summary(api_operations.name.summary());})(op(CamelCaseName))(basic_level)(alternator_label).set_skip_when_empty(),
             OPERATION(batch_get_item, "BatchGetItem")
             OPERATION(batch_write_item, "BatchWriteItem")
             OPERATION(create_backup, "CreateBackup")
             OPERATION(create_global_table, "CreateGlobalTable")
+            OPERATION(create_table, "CreateTable")
             OPERATION(delete_backup, "DeleteBackup")
             OPERATION(delete_item, "DeleteItem")
+            OPERATION(delete_table, "DeleteTable")
             OPERATION(describe_backup, "DescribeBackup")
             OPERATION(describe_continuous_backups, "DescribeContinuousBackups")
             OPERATION(describe_endpoints, "DescribeEndpoints")
@@ -94,107 +64,55 @@ static void register_metrics_with_optional_table(seastar::metrics::metric_groups
             OPERATION(update_item, "UpdateItem")
             OPERATION(update_table, "UpdateTable")
             OPERATION(update_time_to_live, "UpdateTimeToLive")
+            OPERATION_LATENCY(put_item_latency, "PutItem")
+            OPERATION_LATENCY(get_item_latency, "GetItem")
+            OPERATION_LATENCY(delete_item_latency, "DeleteItem")
+            OPERATION_LATENCY(update_item_latency, "UpdateItem")
+            OPERATION_LATENCY(batch_write_item_latency, "BatchWriteItem")
+            OPERATION_LATENCY(batch_get_item_latency, "BatchGetItem")
             OPERATION(list_streams, "ListStreams")
             OPERATION(describe_stream, "DescribeStream")
             OPERATION(get_shard_iterator, "GetShardIterator")
             OPERATION(get_records, "GetRecords")
+            OPERATION_LATENCY(get_records_latency, "GetRecords")
     });
-    OPERATION_LATENCY(put_item_latency, "PutItem")
-    OPERATION_LATENCY(get_item_latency, "GetItem")
-    OPERATION_LATENCY(delete_item_latency, "DeleteItem")
-    OPERATION_LATENCY(update_item_latency, "UpdateItem")
-    OPERATION_LATENCY(batch_write_item_latency, "BatchWriteItem")
-    OPERATION_LATENCY(batch_get_item_latency, "BatchGetItem")
-    OPERATION_LATENCY(get_records_latency, "GetRecords")
-    if (!has_table) {
-        // Create and delete operations are not applicable to a per-table metrics
-        // only register it for the global metrics
-        metrics.add_group("alternator", {
-            OPERATION(create_table, "CreateTable")
-            OPERATION(delete_table, "DeleteTable")
-
-        });
-    }
-    metrics.add_group(group_name, {
-            seastar::metrics::make_total_operations("unsupported_operations", stats.unsupported_operations,
-                    seastar::metrics::description("number of unsupported operations via Alternator API"), labels).set_skip_when_empty(),
-            seastar::metrics::make_total_operations("total_operations", stats.total_operations,
-                    seastar::metrics::description("number of total operations via Alternator API"), labels)(basic_level).aggregate(aggregate_labels).set_skip_when_empty(),
-            seastar::metrics::make_total_operations("reads_before_write", stats.reads_before_write,
-                    seastar::metrics::description("number of performed read-before-write operations"), labels).aggregate(aggregate_labels).set_skip_when_empty(),
-            seastar::metrics::make_total_operations("write_using_lwt", stats.write_using_lwt,
-                    seastar::metrics::description("number of writes that used LWT"), labels).aggregate(aggregate_labels).set_skip_when_empty(),
-            seastar::metrics::make_total_operations("shard_bounce_for_lwt", stats.shard_bounce_for_lwt,
-                    seastar::metrics::description("number writes that had to be bounced from this shard because of LWT requirements"), labels).aggregate(aggregate_labels).set_skip_when_empty(),
-            seastar::metrics::make_total_operations("requests_blocked_memory", stats.requests_blocked_memory,
-                    seastar::metrics::description("Counts a number of requests blocked due to memory pressure."), labels).aggregate(aggregate_labels).set_skip_when_empty(),
-            seastar::metrics::make_total_operations("requests_shed", stats.requests_shed,
-                    seastar::metrics::description("Counts a number of requests shed due to overload."), labels).aggregate(aggregate_labels).set_skip_when_empty(),
-            seastar::metrics::make_total_operations("filtered_rows_read_total", stats.cql_stats.filtered_rows_read_total,
-                    seastar::metrics::description("number of rows read during filtering operations"), labels).aggregate(aggregate_labels).set_skip_when_empty(),
-            seastar::metrics::make_total_operations("filtered_rows_matched_total", stats.cql_stats.filtered_rows_matched_total,
-                    seastar::metrics::description("number of rows read and matched during filtering operations"), labels).aggregate(aggregate_labels).set_skip_when_empty(),
-            seastar::metrics::make_counter("rcu_total", [&stats]{return 0.5 * stats.rcu_half_units_total;},
-                    seastar::metrics::description("total number of consumed read units"), labels).aggregate(aggregate_labels).set_skip_when_empty(),
-            seastar::metrics::make_counter("wcu_total", stats.wcu_total[stats::wcu_types::PUT_ITEM],
-                    seastar::metrics::description("total number of consumed write units"), labels)(op("PutItem")).aggregate(aggregate_labels).set_skip_when_empty(),
-            seastar::metrics::make_counter("wcu_total", stats.wcu_total[stats::wcu_types::DELETE_ITEM],
-                    seastar::metrics::description("total number of consumed write units"), labels)(op("DeleteItem")).aggregate(aggregate_labels).set_skip_when_empty(),
-            seastar::metrics::make_counter("wcu_total", stats.wcu_total[stats::wcu_types::UPDATE_ITEM],
-                    seastar::metrics::description("total number of consumed write units"), labels)(op("UpdateItem")).aggregate(aggregate_labels).set_skip_when_empty(),
-            seastar::metrics::make_counter("wcu_total", stats.wcu_total[stats::wcu_types::INDEX],
-                    seastar::metrics::description("total number of consumed write units"), labels)(op("Index")).aggregate(aggregate_labels).set_skip_when_empty(),
-            seastar::metrics::make_total_operations("filtered_rows_dropped_total", [&stats] { return stats.cql_stats.filtered_rows_read_total - stats.cql_stats.filtered_rows_matched_total; },
-                    seastar::metrics::description("number of rows read and dropped during filtering operations"), labels).aggregate(aggregate_labels).set_skip_when_empty(),
-            seastar::metrics::make_counter("batch_item_count", seastar::metrics::description("The total number of items processed across all batches"), labels,
-                    stats.api_operations.batch_write_item_batch_total)(op("BatchWriteItem")).aggregate(aggregate_labels).set_skip_when_empty(),
-            seastar::metrics::make_counter("batch_item_count", seastar::metrics::description("The total number of items processed across all batches"), labels,
-                    stats.api_operations.batch_get_item_batch_total)(op("BatchGetItem")).aggregate(aggregate_labels).set_skip_when_empty(),
-            seastar::metrics::make_histogram("batch_item_count_histogram", seastar::metrics::description("Histogram of the number of items in a batch request"), labels,
-                    [&stats]{ return estimated_histogram_to_metrics(stats.api_operations.batch_get_item_histogram);})(op("BatchGetItem")).aggregate({seastar::metrics::shard_label}).set_skip_when_empty(),
-            seastar::metrics::make_histogram("batch_item_count_histogram", seastar::metrics::description("Histogram of the number of items in a batch request"), labels,
-                    [&stats]{ return estimated_histogram_to_metrics(stats.api_operations.batch_write_item_histogram);})(op("BatchWriteItem")).aggregate({seastar::metrics::shard_label}).set_skip_when_empty(),
-            seastar::metrics::make_histogram("operation_size_kb", seastar::metrics::description("Histogram of item sizes involved in a request"), labels,
-                    [&stats]{ return estimated_histogram_to_metrics(stats.operation_sizes.get_item_op_size_kb);})(op("GetItem")).aggregate({seastar::metrics::shard_label}).set_skip_when_empty(),
-            seastar::metrics::make_histogram("operation_size_kb", seastar::metrics::description("Histogram of item sizes involved in a request"), labels,
-                    [&stats]{ return estimated_histogram_to_metrics(stats.operation_sizes.put_item_op_size_kb);})(op("PutItem")).aggregate({seastar::metrics::shard_label}).set_skip_when_empty(),
-            seastar::metrics::make_histogram("operation_size_kb", seastar::metrics::description("Histogram of item sizes involved in a request"), labels,
-                    [&stats]{ return estimated_histogram_to_metrics(stats.operation_sizes.delete_item_op_size_kb);})(op("DeleteItem")).aggregate({seastar::metrics::shard_label}).set_skip_when_empty(),
-            seastar::metrics::make_histogram("operation_size_kb", seastar::metrics::description("Histogram of item sizes involved in a request"), labels,
-                    [&stats]{ return estimated_histogram_to_metrics(stats.operation_sizes.update_item_op_size_kb);})(op("UpdateItem")).aggregate({seastar::metrics::shard_label}).set_skip_when_empty(),
-            seastar::metrics::make_histogram("operation_size_kb", seastar::metrics::description("Histogram of item sizes involved in a request"), labels,
-                    [&stats]{ return estimated_histogram_to_metrics(stats.operation_sizes.batch_get_item_op_size_kb);})(op("BatchGetItem")).aggregate({seastar::metrics::shard_label}).set_skip_when_empty(),
-            seastar::metrics::make_histogram("operation_size_kb", seastar::metrics::description("Histogram of item sizes involved in a request"), labels,
-                    [&stats]{ return estimated_histogram_to_metrics(stats.operation_sizes.batch_write_item_op_size_kb);})(op("BatchWriteItem")).aggregate({seastar::metrics::shard_label}).set_skip_when_empty(),
-    });
-
-    seastar::metrics::label expression_label("expression");
-    metrics.add_group(group_name, {
-            seastar::metrics::make_total_operations("expression_cache_evictions", stats.expression_cache.evictions,
-                    seastar::metrics::description("Counts number of entries evicted from expressions cache"), labels).aggregate(aggregate_labels).set_skip_when_empty(),
-
-            seastar::metrics::make_total_operations("expression_cache_hits", stats.expression_cache.requests[stats::expression_types::UPDATE_EXPRESSION].hits,
-                    seastar::metrics::description("Counts number of hits of cached expressions"), labels)(expression_label("UpdateExpression")).aggregate(aggregate_labels).set_skip_when_empty(),
-            seastar::metrics::make_total_operations("expression_cache_misses", stats.expression_cache.requests[stats::expression_types::UPDATE_EXPRESSION].misses,
-                    seastar::metrics::description("Counts number of misses of cached expressions"), labels)(expression_label("UpdateExpression")).aggregate(aggregate_labels).set_skip_when_empty(),
-            
-            seastar::metrics::make_total_operations("expression_cache_hits", stats.expression_cache.requests[stats::expression_types::CONDITION_EXPRESSION].hits,
-                    seastar::metrics::description("Counts number of hits of cached expressions"), labels)(expression_label("ConditionExpression")).aggregate(aggregate_labels).set_skip_when_empty(),
-            seastar::metrics::make_total_operations("expression_cache_misses", stats.expression_cache.requests[stats::expression_types::CONDITION_EXPRESSION].misses,
-                    seastar::metrics::description("Counts number of misses of cached expressions"), labels)(expression_label("ConditionExpression")).aggregate(aggregate_labels).set_skip_when_empty(),
-
-            seastar::metrics::make_total_operations("expression_cache_hits", stats.expression_cache.requests[stats::expression_types::PROJECTION_EXPRESSION].hits,
-                    seastar::metrics::description("Counts number of hits of cached expressions"), labels)(expression_label("ProjectionExpression")).aggregate(aggregate_labels).set_skip_when_empty(),
-            seastar::metrics::make_total_operations("expression_cache_misses", stats.expression_cache.requests[stats::expression_types::PROJECTION_EXPRESSION].misses,
-                    seastar::metrics::description("Counts number of misses of cached expressions"), labels)(expression_label("ProjectionExpression")).aggregate(aggregate_labels).set_skip_when_empty()
+    _metrics.add_group("alternator", {
+            seastar::metrics::make_total_operations("unsupported_operations", unsupported_operations,
+                    seastar::metrics::description("number of unsupported operations via Alternator API"))(alternator_label).set_skip_when_empty(),
+            seastar::metrics::make_total_operations("total_operations", total_operations,
+                    seastar::metrics::description("number of total operations via Alternator API"))(basic_level)(alternator_label).set_skip_when_empty(),
+            seastar::metrics::make_total_operations("reads_before_write", reads_before_write,
+                    seastar::metrics::description("number of performed read-before-write operations"))(alternator_label).set_skip_when_empty(),
+            seastar::metrics::make_total_operations("write_using_lwt", write_using_lwt,
+                    seastar::metrics::description("number of writes that used LWT"))(alternator_label).set_skip_when_empty(),
+            seastar::metrics::make_total_operations("shard_bounce_for_lwt", shard_bounce_for_lwt,
+                    seastar::metrics::description("number writes that had to be bounced from this shard because of LWT requirements"))(alternator_label).set_skip_when_empty(),
+            seastar::metrics::make_total_operations("requests_blocked_memory", requests_blocked_memory,
+                    seastar::metrics::description("Counts a number of requests blocked due to memory pressure."))(alternator_label).set_skip_when_empty(),
+            seastar::metrics::make_total_operations("requests_shed", requests_shed,
+                    seastar::metrics::description("Counts a number of requests shed due to overload."))(alternator_label).set_skip_when_empty(),
+            seastar::metrics::make_total_operations("filtered_rows_read_total", cql_stats.filtered_rows_read_total,
+                    seastar::metrics::description("number of rows read during filtering operations"))(alternator_label).set_skip_when_empty(),
+            seastar::metrics::make_total_operations("filtered_rows_matched_total", cql_stats.filtered_rows_matched_total,
+                    seastar::metrics::description("number of rows read and matched during filtering operations")),
+            seastar::metrics::make_counter("rcu_total", rcu_total,
+                    seastar::metrics::description("total number of consumed read units, counted as half units"))(alternator_label).set_skip_when_empty(),
+            seastar::metrics::make_counter("wcu_total", wcu_total[wcu_types::PUT_ITEM],
+                    seastar::metrics::description("total number of consumed write units, counted as half units"),{op("PutItem")})(alternator_label).set_skip_when_empty(),
+            seastar::metrics::make_counter("wcu_total", wcu_total[wcu_types::DELETE_ITEM],
+                    seastar::metrics::description("total number of consumed write units, counted as half units"),{op("DeleteItem")})(alternator_label).set_skip_when_empty(),
+            seastar::metrics::make_counter("wcu_total", wcu_total[wcu_types::UPDATE_ITEM],
+                    seastar::metrics::description("total number of consumed write units, counted as half units"),{op("UpdateItem")})(alternator_label).set_skip_when_empty(),
+            seastar::metrics::make_counter("wcu_total", wcu_total[wcu_types::INDEX],
+                    seastar::metrics::description("total number of consumed write units, counted as half units"),{op("Index")})(alternator_label).set_skip_when_empty(),
+            seastar::metrics::make_total_operations("filtered_rows_dropped_total", [this] { return cql_stats.filtered_rows_read_total - cql_stats.filtered_rows_matched_total; },
+                    seastar::metrics::description("number of rows read and dropped during filtering operations"))(alternator_label).set_skip_when_empty(),
+            seastar::metrics::make_counter("batch_item_count", seastar::metrics::description("The total number of items processed across all batches"),{op("BatchWriteItem")},
+                    api_operations.batch_write_item_batch_total)(alternator_label).set_skip_when_empty(),
+            seastar::metrics::make_counter("batch_item_count", seastar::metrics::description("The total number of items processed across all batches"),{op("BatchGetItem")},
+                    api_operations.batch_get_item_batch_total)(alternator_label).set_skip_when_empty(),
     });
 }
 
-void register_metrics(seastar::metrics::metric_groups& metrics, const stats& stats) {
-    register_metrics_with_optional_table(metrics, stats, "", "");
-}
-table_stats::table_stats(const sstring& ks, const sstring& table) {
-    _stats = make_lw_shared<stats>();
-    register_metrics_with_optional_table(_metrics, *_stats, ks, table);
-}
+
 }

alternator/stats.hh

@@ -12,7 +12,6 @@
 
 #include <seastar/core/metrics_registration.hh>
 #include "utils/histogram.hh"
-#include "utils/estimated_histogram.hh"
 #include "cql3/stats.hh"
 
 namespace alternator {
@@ -22,6 +21,7 @@ namespace alternator {
 // visible by the metrics REST API, with the "alternator" prefix.
 class stats {
 public:
+    stats();
     // Count of DynamoDB API operations by types
     struct {
         uint64_t batch_get_item = 0;
@@ -75,36 +75,7 @@ public:
         utils::timed_rate_moving_average_summary_and_histogram batch_write_item_latency;
         utils::timed_rate_moving_average_summary_and_histogram batch_get_item_latency;
         utils::timed_rate_moving_average_summary_and_histogram get_records_latency;
-
-        utils::estimated_histogram batch_get_item_histogram{22}; // a histogram that covers the range 1 - 100
-        utils::estimated_histogram batch_write_item_histogram{22}; // a histogram that covers the range 1 - 100
     } api_operations;
-    // Operation size metrics
-    struct {
-        // Item size statistics collected per table and aggregated per node.
-        // Each histogram covers the range 0 - 446. Resolves #25143.
-        // A size is the retrieved item's size.
-        utils::estimated_histogram get_item_op_size_kb{30};
-        // A size is the maximum of the new item's size and the old item's size.
-        utils::estimated_histogram put_item_op_size_kb{30};
-        // A size is the deleted item's size. If the deleted item's size is
-        // unknown (i.e. read-before-write wasn't necessary and it wasn't
-        // forced by a configuration option), it won't be recorded on the
-        // histogram.
-        utils::estimated_histogram delete_item_op_size_kb{30};
-        // A size is the maximum of existing item's size and the estimated size
-        // of the update. This will be changed to the maximum of the existing item's
-        // size and the new item's size in a subsequent PR.
-        utils::estimated_histogram update_item_op_size_kb{30};
-
-        // A size is the sum of the sizes of all items per table. This means
-        // that a single BatchGetItem / BatchWriteItem updates the histogram
-        // for each table that it has items in.
-        // The sizes are the retrieved items' sizes grouped per table.
-        utils::estimated_histogram batch_get_item_op_size_kb{30};
-        // The sizes are the the written items' sizes grouped per table.
-        utils::estimated_histogram batch_write_item_op_size_kb{30};
-    } operation_sizes;
     // Miscellaneous event counters
     uint64_t total_operations = 0;
     uint64_t unsupported_operations = 0;
@@ -113,7 +84,7 @@ public:
     uint64_t shard_bounce_for_lwt = 0;
     uint64_t requests_blocked_memory = 0;
     uint64_t requests_shed = 0;
-    uint64_t rcu_half_units_total = 0;
+    uint64_t rcu_total = 0;
     // wcu can results from put, update, delete and index
     // Index related will be done on top of the operation it comes with
     enum wcu_types {
@@ -127,33 +98,10 @@ public:
     uint64_t wcu_total[NUM_TYPES] = {0};
     // CQL-derived stats
     cql3::cql_stats cql_stats;
-
-    // Enumeration of expression types only for stats
-    // if needed it can be extended e.g. per operation 
-    enum expression_types {
-        UPDATE_EXPRESSION,
-        CONDITION_EXPRESSION,
-        PROJECTION_EXPRESSION,
-        NUM_EXPRESSION_TYPES
-    };
-    struct {
-        struct {
-            uint64_t hits = 0;
-            uint64_t misses = 0;
-        } requests[NUM_EXPRESSION_TYPES];
-        uint64_t evictions = 0;
-    } expression_cache;
-};
-
-struct table_stats {
-    table_stats(const sstring& ks, const sstring& table);
+private:
+    // The metric_groups object holds this stat object's metrics registered
+    // as long as the stats object is alive.
     seastar::metrics::metric_groups _metrics;
-    lw_shared_ptr<stats> _stats;
 };
-void register_metrics(seastar::metrics::metric_groups& metrics, const stats& stats);
-
-inline uint64_t bytes_to_kb_ceil(uint64_t bytes) {
-    return (bytes + 1023) / 1024;
-}
 
 }

alternator/streams.cc

@@ -13,6 +13,7 @@
 
 #include <seastar/json/formatter.hh>
 
+#include "auth/permission.hh"
 #include "db/config.hh"
 
 #include "cdc/log.hh"
@@ -31,7 +32,6 @@
 
 #include "executor.hh"
 #include "data_dictionary/data_dictionary.hh"
-#include "utils/rjson.hh"
 
 /**
  * Base template type to implement  rapidjson::internal::TypeHelper<...>:s
@@ -126,7 +126,7 @@ public:
     }
 };
 
-} // namespace alternator
+}
 
 template<typename ValueType>
 struct rapidjson::internal::TypeHelper<ValueType, alternator::stream_arn>
@@ -217,7 +217,7 @@ future<alternator::executor::request_return_type> alternator::executor::list_str
         rjson::add(ret, "LastEvaluatedStreamArn", *last);
     }
 
-    return make_ready_future<executor::request_return_type>(rjson::print(std::move(ret)));
+    return make_ready_future<executor::request_return_type>(make_jsonable(std::move(ret)));
 }
 
 struct shard_id {
@@ -296,7 +296,7 @@ sequence_number::sequence_number(std::string_view v)
     }())
 {}
 
-} // namespace alternator
+}
 
 template<typename ValueType>
 struct rapidjson::internal::TypeHelper<ValueType, alternator::shard_id>
@@ -356,7 +356,7 @@ static stream_view_type cdc_options_to_steam_view_type(const cdc::options& opts)
     return type;
 }
 
-} // namespace alternator
+}
 
 template<typename ValueType>
 struct rapidjson::internal::TypeHelper<ValueType, alternator::stream_view_type>
@@ -475,10 +475,10 @@ future<executor::request_return_type> executor::describe_stream(client_state& cl
         } else {
             status = "ENABLED";
         }
-    }
+    } 
 
     auto ttl = std::chrono::seconds(opts.ttl());
-
+    
     rjson::add(stream_desc, "StreamStatus", rjson::from_string(status));
 
     stream_view_type type = cdc_options_to_steam_view_type(opts);
@@ -491,7 +491,7 @@ future<executor::request_return_type> executor::describe_stream(client_state& cl
 
     if (!opts.enabled()) {
         rjson::add(ret, "StreamDescription", std::move(stream_desc));
-        return make_ready_future<executor::request_return_type>(rjson::print(std::move(ret)));
+        return make_ready_future<executor::request_return_type>(make_jsonable(std::move(ret)));
     }
 
     // TODO: label
@@ -617,7 +617,7 @@ future<executor::request_return_type> executor::describe_stream(client_state& cl
         rjson::add(stream_desc, "Shards", std::move(shards));
         rjson::add(ret, "StreamDescription", std::move(stream_desc));
             
-        return make_ready_future<executor::request_return_type>(rjson::print(std::move(ret)));
+        return make_ready_future<executor::request_return_type>(make_jsonable(std::move(ret)));
     });
 }
 
@@ -714,7 +714,7 @@ future<executor::request_return_type> executor::get_shard_iterator(client_state&
 
     auto type = rjson::get<shard_iterator_type>(request, "ShardIteratorType");
     auto seq_num = rjson::get_opt<sequence_number>(request, "SequenceNumber");
-
+    
     if (type < shard_iterator_type::TRIM_HORIZON && !seq_num) {
         throw api_error::validation("Missing required parameter \"SequenceNumber\"");
     }
@@ -724,7 +724,7 @@ future<executor::request_return_type> executor::get_shard_iterator(client_state&
 
     auto stream_arn = rjson::get<alternator::stream_arn>(request, "StreamArn");
     auto db = _proxy.data_dictionary();
-
+    
     schema_ptr schema = nullptr;
     std::optional<shard_id> sid;
 
@@ -770,7 +770,7 @@ future<executor::request_return_type> executor::get_shard_iterator(client_state&
     auto ret = rjson::empty_object();
     rjson::add(ret, "ShardIterator", iter);
 
-    return make_ready_future<executor::request_return_type>(rjson::print(std::move(ret)));
+    return make_ready_future<executor::request_return_type>(make_jsonable(std::move(ret)));
 }
 
 struct event_id {
@@ -789,7 +789,7 @@ struct event_id {
         return os;
     }
 };
-} // namespace alternator
+}
 
 template<typename ValueType>
 struct rapidjson::internal::TypeHelper<ValueType, alternator::event_id>
@@ -808,9 +808,6 @@ future<executor::request_return_type> executor::get_records(client_state& client
     if (limit < 1) {
         throw api_error::validation("Limit must be 1 or more");
     }
-    if (limit > 1000) {
-        throw api_error::validation("Limit must be less than or equal to 1000");
-    }
 
     auto db = _proxy.data_dictionary();
     schema_ptr schema, base;
@@ -871,12 +868,10 @@ future<executor::request_return_type> executor::get_records(client_state& client
 
     std::transform(pks.begin(), pks.end(), std::back_inserter(columns), [](auto& c) { return &c; });
     std::transform(cks.begin(), cks.end(), std::back_inserter(columns), [](auto& c) { return &c; });
-    auto regular_column_start_idx = columns.size();
-    auto regular_column_filter = std::views::filter([](const column_definition& cdef) { return cdef.name() == op_column_name || cdef.name() == eor_column_name || !cdc::is_cdc_metacolumn_name(cdef.name_as_text()); });
-    std::ranges::transform(schema->regular_columns() | regular_column_filter, std::back_inserter(columns), [](auto& c) { return &c; });
 
-    auto regular_columns = std::ranges::subrange(columns.begin() + regular_column_start_idx, columns.end())
-        | std::views::transform(&column_definition::id)
+    auto regular_columns = schema->regular_columns()
+        | std::views::filter([](const column_definition& cdef) { return cdef.name() == op_column_name || cdef.name() == eor_column_name || !cdc::is_cdc_metacolumn_name(cdef.name_as_text()); })
+        | std::views::transform([&] (const column_definition& cdef) { columns.emplace_back(&cdef); return cdef.id; })
         | std::ranges::to<query::column_id_vector>()
     ;
 
@@ -927,7 +922,6 @@ future<executor::request_return_type> executor::get_records(client_state& client
         std::optional<utils::UUID> timestamp;
         auto dynamodb = rjson::empty_object();
         auto record = rjson::empty_object();
-        const auto dc_name = _proxy.get_token_metadata_ptr()->get_topology().get_datacenter();
 
         using op_utype = std::underlying_type_t<cdc::operation>;
 
@@ -937,10 +931,9 @@ future<executor::request_return_type> executor::get_records(client_state& client
                 dynamodb = rjson::empty_object();
             }
             if (!record.ObjectEmpty()) {
-                rjson::add(record, "awsRegion", rjson::from_string(dc_name));
+                // TODO: awsRegion?
                 rjson::add(record, "eventID", event_id(iter.shard.id, *timestamp));
                 rjson::add(record, "eventSource", "scylladb:alternator");
-                rjson::add(record, "eventVersion", "1.1");
                 rjson::push_back(records, std::move(record));
                 record = rjson::empty_object();
                 --limit;
@@ -959,7 +952,7 @@ future<executor::request_return_type> executor::get_records(client_state& client
                 rjson::add(dynamodb, "ApproximateCreationDateTime", utils::UUID_gen::unix_timestamp_in_sec(ts).count());
                 rjson::add(dynamodb, "SequenceNumber", sequence_number(ts));
                 rjson::add(dynamodb, "StreamViewType", type);
-                // TODO: SizeBytes
+                //TODO: SizeInBytes
             }
 
             /**
@@ -999,16 +992,6 @@ future<executor::request_return_type> executor::get_records(client_state& client
             case cdc::operation::insert:
                 rjson::add(record, "eventName", "INSERT");
                 break;
-            case cdc::operation::service_row_delete:
-            case cdc::operation::service_partition_delete:
-            {
-                auto user_identity = rjson::empty_object();
-                rjson::add(user_identity, "Type", "Service");
-                rjson::add(user_identity, "PrincipalId", "dynamodb.amazonaws.com");
-                rjson::add(record, "userIdentity", std::move(user_identity));
-                rjson::add(record, "eventName", "REMOVE");
-                break;
-            }
             default:
                 rjson::add(record, "eventName", "REMOVE");
                 break;
@@ -1035,7 +1018,7 @@ future<executor::request_return_type> executor::get_records(client_state& client
             // will notice end end of shard and not return NextShardIterator.
             rjson::add(ret, "NextShardIterator", next_iter);
             _stats.api_operations.get_records_latency.mark(std::chrono::steady_clock::now() - start_time);
-            return make_ready_future<executor::request_return_type>(rjson::print(std::move(ret)));
+            return make_ready_future<executor::request_return_type>(make_jsonable(std::move(ret)));
         }
 
         // ugh. figure out if we are and end-of-shard
@@ -1061,12 +1044,12 @@ future<executor::request_return_type> executor::get_records(client_state& client
             if (is_big(ret)) {
                 return make_ready_future<executor::request_return_type>(make_streamed(std::move(ret)));
             }
-            return make_ready_future<executor::request_return_type>(rjson::print(std::move(ret)));
+            return make_ready_future<executor::request_return_type>(make_jsonable(std::move(ret)));
         });
     });
 }
 
-bool executor::add_stream_options(const rjson::value& stream_specification, schema_builder& builder, service::storage_proxy& sp) {
+void executor::add_stream_options(const rjson::value& stream_specification, schema_builder& builder, service::storage_proxy& sp) {
     auto stream_enabled = rjson::find(stream_specification, "StreamEnabled");
     if (!stream_enabled || !stream_enabled->IsBool()) {
         throw api_error::validation("StreamSpecification needs boolean StreamEnabled");
@@ -1100,12 +1083,10 @@ bool executor::add_stream_options(const rjson::value& stream_specification, sche
                 break;
         }
         builder.with_cdc_options(opts);
-        return true;
     } else {
         cdc::options opts;
         opts.enabled(false);
         builder.with_cdc_options(opts);
-        return false;
     }
 }
 
@@ -1134,4 +1115,4 @@ void executor::supplement_table_stream_info(rjson::value& descr, const schema& s
     }
 }
 
-} // namespace alternator
+}

alternator/ttl.cc

@@ -17,7 +17,6 @@
 #include <seastar/core/lowres_clock.hh>
 #include <seastar/coroutine/maybe_yield.hh>
 
-#include "cdc/log.hh"
 #include "exceptions/exceptions.hh"
 #include "gms/gossiper.hh"
 #include "gms/inet_address.hh"
@@ -28,7 +27,7 @@
 #include "replica/database.hh"
 #include "service/client_state.hh"
 #include "service_permit.hh"
-#include "mutation/timestamp.hh"
+#include "timestamp.hh"
 #include "service/storage_proxy.hh"
 #include "service/pager/paging_state.hh"
 #include "service/pager/query_pagers.hh"
@@ -57,14 +56,14 @@ static logging::logger tlogger("alternator_ttl");
 
 namespace alternator {
 
-// We write the expiration-time attribute enabled on a table in a
+// We write the expiration-time attribute enabled on a table using a
 // tag TTL_TAG_KEY.
 // Currently, the *value* of this tag is simply the name of the attribute,
 // and the expiration scanner interprets it as an Alternator attribute name -
 // It can refer to a real column or if that doesn't exist, to a member of
 // the ":attrs" map column. Although this is designed for Alternator, it may
 // be good enough for CQL as well (there, the ":attrs" column won't exist).
-extern const sstring TTL_TAG_KEY;
+static const sstring TTL_TAG_KEY("system:ttl_attribute");
 
 future<executor::request_return_type> executor::update_time_to_live(client_state& client_state, service_permit permit, rjson::value request) {
     _stats.api_operations.update_time_to_live++;
@@ -82,6 +81,11 @@ future<executor::request_return_type> executor::update_time_to_live(client_state
         co_return api_error::validation("UpdateTimeToLive requires boolean Enabled");
     }
     bool enabled = v->GetBool();
+    // Alternator TTL doesn't yet work when the table uses tablets (#16567)
+    if (enabled && _proxy.local_db().find_keyspace(schema->ks_name()).get_replication_strategy().uses_tablets()) {
+        co_return api_error::validation("TTL not yet supported on a table using tablets (issue #16567). "
+            "Create a table with the tag 'experimental:initial_tablets' set to 'none' to use vnodes.");
+    }
     v = rjson::find(*spec, "AttributeName");
     if (!v || !v->IsString()) {
         co_return api_error::validation("UpdateTimeToLive requires string AttributeName");
@@ -119,7 +123,7 @@ future<executor::request_return_type> executor::update_time_to_live(client_state
     // basically identical to the request's
     rjson::value response = rjson::empty_object();
     rjson::add(response, "TimeToLiveSpecification", std::move(*spec));
-    co_return rjson::print(std::move(response));
+    co_return make_jsonable(std::move(response));
 }
 
 future<executor::request_return_type> executor::describe_time_to_live(client_state& client_state, service_permit permit, rjson::value request) {
@@ -136,7 +140,7 @@ future<executor::request_return_type> executor::describe_time_to_live(client_sta
     }
     rjson::value response = rjson::empty_object();
     rjson::add(response, "TimeToLiveDescription", std::move(desc));
-    co_return rjson::print(std::move(response));
+    co_return make_jsonable(std::move(response));
 }
 
 // expiration_service is a sharded service responsible for cleaning up expired
@@ -287,18 +291,13 @@ static future<> expire_item(service::storage_proxy& proxy,
         auto ck = clustering_key::from_exploded(exploded_ck);
         m.partition().clustered_row(*schema, ck).apply(tombstone(ts, gc_clock::now()));
     }
-    utils::chunked_vector<mutation> mutations;
+    std::vector<mutation> mutations;
     mutations.push_back(std::move(m));
     return proxy.mutate(std::move(mutations),
         db::consistency_level::LOCAL_QUORUM,
         executor::default_timeout(), // FIXME - which timeout?
         qs.get_trace_state(), qs.get_permit(),
-        db::allow_per_partition_rate_limit::no,
-        false,
-        cdc::per_request_options{
-            .is_system_originated = true,
-        }
-    );
+        db::allow_per_partition_rate_limit::no);
 }
 
 static size_t random_offset(size_t min, size_t max) {
@@ -316,10 +315,8 @@ static size_t random_offset(size_t min, size_t max) {
 // this range's primary node is down. For this we need to return not just
 // a list of this node's secondary ranges - but also the primary owner of
 // each of those ranges.
-//
-// The function is to be used with vnodes only
 static future<std::vector<std::pair<dht::token_range, locator::host_id>>> get_secondary_ranges(
-        const locator::effective_replication_map* erm,
+        const locator::effective_replication_map_ptr& erm,
         locator::host_id ep) {
     const auto& tm = *erm->get_token_metadata_ptr();
     const auto& sorted_tokens = tm.sorted_tokens();
@@ -330,7 +327,6 @@ static future<std::vector<std::pair<dht::token_range, locator::host_id>>> get_se
     auto prev_tok = sorted_tokens.back();
     for (const auto& tok : sorted_tokens) {
         co_await coroutine::maybe_yield();
-        // FIXME: pass is_vnode=true to get_natural_replicas since the token is in tm.sorted_tokens()
         host_id_vector_replica_set eps = erm->get_natural_replicas(tok);
         if (eps.size() <= 1 || eps[1] != ep) {
             prev_tok = tok;
@@ -400,7 +396,7 @@ class ranges_holder_primary {
     dht::token_range_vector _token_ranges;
 public:
     explicit ranges_holder_primary(dht::token_range_vector token_ranges) : _token_ranges(std::move(token_ranges)) {}
-    static future<ranges_holder_primary> make(const locator::vnode_effective_replication_map* erm, locator::host_id ep) {
+    static future<ranges_holder_primary> make(const locator::vnode_effective_replication_map_ptr& erm, locator::host_id ep) {
         co_return ranges_holder_primary(co_await erm->get_primary_ranges(ep));
     }
     std::size_t size() const { return _token_ranges.size(); }
@@ -420,7 +416,7 @@ public:
     explicit ranges_holder_secondary(std::vector<std::pair<dht::token_range, locator::host_id>> token_ranges, const gms::gossiper& g)
         : _token_ranges(std::move(token_ranges))
         , _gossiper(g) {}
-    static future<ranges_holder_secondary> make(const locator::vnode_effective_replication_map* erm, locator::host_id ep, const gms::gossiper& g) {
+    static future<ranges_holder_secondary> make(const locator::effective_replication_map_ptr& erm, locator::host_id ep, const gms::gossiper& g) {
         co_return ranges_holder_secondary(co_await get_secondary_ranges(erm, ep), g);
     }
     std::size_t size() const { return _token_ranges.size(); }
@@ -433,8 +429,6 @@ public:
     }
 };
 
-// The token_ranges_owned_by_this_shard class is only used for vnodes, where the vnodes give a partition range for the entire node
-// and such range still needs to be divided between the shards.
 template<class primary_or_secondary_t>
 class token_ranges_owned_by_this_shard {
     schema_ptr _s;
@@ -528,7 +522,7 @@ struct scan_ranges_context {
         // should be possible (and a must for issue #7751!).
         lw_shared_ptr<service::pager::paging_state> paging_state = nullptr;
         auto regular_columns =
-            s->regular_columns() | std::views::transform(&column_definition::id)
+            s->regular_columns() | std::views::transform([] (const column_definition& cdef) { return cdef.id; })
             | std::ranges::to<query::column_id_vector>();
         selection = cql3::selection::selection::wildcard(s);
         query::partition_slice::option_set opts = selection->get_query_options();
@@ -661,17 +655,6 @@ static future<> scan_table_ranges(
     }
 }
 
-static future<> scan_tablet(locator::tablet_id tablet, service::storage_proxy& proxy, abort_source& abort_source, named_semaphore& page_sem,
-            expiration_service::stats& expiration_stats, const scan_ranges_context& scan_ctx, const locator::tablet_map& tablet_map) {
-    auto tablet_token_range = tablet_map.get_token_range(tablet);
-    dht::ring_position tablet_start(tablet_token_range.start()->value(), dht::ring_position::token_bound::start),
-                       tablet_end(tablet_token_range.end()->value(), dht::ring_position::token_bound::end);
-    auto partition_range = dht::partition_range::make(std::move(tablet_start), std::move(tablet_end));
-    // Note that because of issue #9167 we need to run a separate query on each partition range, and can't pass
-    // several of them into one partition_range_vector that is passed to scan_table_ranges().
-    return scan_table_ranges(proxy, scan_ctx, {partition_range}, abort_source, page_sem, expiration_stats);
-}
-
 // scan_table() scans, in one table, data "owned" by this shard, looking for
 // expired items and deleting them.
 // We consider each node to "own" its primary token ranges, i.e., the tokens
@@ -747,69 +730,34 @@ static future<bool> scan_table(
     expiration_stats.scan_table++;
     // FIXME: need to pace the scan, not do it all at once.
     scan_ranges_context scan_ctx{s, proxy, std::move(column_name), std::move(member)};
-
-    if (s->table().uses_tablets()) {
-        locator::effective_replication_map_ptr erm = s->table().get_effective_replication_map();
-        auto my_host_id = erm->get_topology().my_host_id();
-        const auto &tablet_map = erm->get_token_metadata().tablets().get_tablet_map(s->id());
-        for (std::optional tablet = tablet_map.first_tablet(); tablet; tablet = tablet_map.next_tablet(*tablet)) {
-            auto tablet_primary_replica = tablet_map.get_primary_replica(*tablet);
-            // check if this is the primary replica for the current tablet
-            if (tablet_primary_replica.host == my_host_id && tablet_primary_replica.shard == this_shard_id()) {
-                co_await scan_tablet(*tablet, proxy, abort_source, page_sem, expiration_stats, scan_ctx, tablet_map);
-            } else if(erm->get_replication_factor() > 1) {
-                // Check if this is the secondary replica for the current tablet
-                // and if the primary replica is down which means we will take over this work.
-                // If each node only scans its own primary ranges, then when any node is
-                // down part of the token range will not get scanned. This can be viewed
-                // as acceptable (when the comes back online, it will resume its scan),
-                // but as noted in issue #9787, we can allow more prompt expiration
-                // by tasking another node to take over scanning of the dead node's primary
-                // ranges. What we do here is that this node will also check expiration
-                // on its *secondary* ranges - but only those whose primary owner is down.
-                auto tablet_secondary_replica = tablet_map.get_secondary_replica(*tablet); // throws if no secondary replica
-                if (tablet_secondary_replica.host == my_host_id && tablet_secondary_replica.shard == this_shard_id()) {
-                    if (!gossiper.is_alive(tablet_primary_replica.host)) {
-                        co_await scan_tablet(*tablet, proxy, abort_source, page_sem, expiration_stats, scan_ctx, tablet_map);
-                    }
-                }
-            }
-        }
-    } else {  // VNodes
-        locator::static_effective_replication_map_ptr ermp =
-                db.real_database().find_keyspace(s->ks_name()).get_static_effective_replication_map();
-        auto* erm = ermp->maybe_as_vnode_effective_replication_map();
-        if (!erm) {
-            on_internal_error(tlogger, format("Keyspace {} is local", s->ks_name()));
-        }
-        auto my_host_id = erm->get_topology().my_host_id();
-        token_ranges_owned_by_this_shard my_ranges(s, co_await ranges_holder_primary::make(erm, my_host_id));
-        while (std::optional<dht::partition_range> range = my_ranges.next_partition_range()) {
-            // Note that because of issue #9167 we need to run a separate
-            // query on each partition range, and can't pass several of
-            // them into one partition_range_vector.
-            dht::partition_range_vector partition_ranges;
-            partition_ranges.push_back(std::move(*range));
-            // FIXME: if scanning a single range fails, including network errors,
-            // we fail the entire scan (and rescan from the beginning). Need to
-            // reconsider this. Saving the scan position might be a good enough
-            // solution for this problem.
-            co_await scan_table_ranges(proxy, scan_ctx, std::move(partition_ranges), abort_source, page_sem, expiration_stats);
-        }
-        // If each node only scans its own primary ranges, then when any node is
-        // down part of the token range will not get scanned. This can be viewed
-        // as acceptable (when the comes back online, it will resume its scan),
-        // but as noted in issue #9787, we can allow more prompt expiration
-        // by tasking another node to take over scanning of the dead node's primary
-        // ranges. What we do here is that this node will also check expiration
-        // on its *secondary* ranges - but only those whose primary owner is down.
-        token_ranges_owned_by_this_shard my_secondary_ranges(s, co_await ranges_holder_secondary::make(erm, my_host_id, gossiper));
-        while (std::optional<dht::partition_range> range = my_secondary_ranges.next_partition_range()) {
-            expiration_stats.secondary_ranges_scanned++;
-            dht::partition_range_vector partition_ranges;
-            partition_ranges.push_back(std::move(*range));
-            co_await scan_table_ranges(proxy, scan_ctx, std::move(partition_ranges), abort_source, page_sem, expiration_stats);
-        }
+    auto erm = db.real_database().find_keyspace(s->ks_name()).get_vnode_effective_replication_map();
+    auto my_host_id = erm->get_topology().my_host_id();
+    token_ranges_owned_by_this_shard my_ranges(s, co_await ranges_holder_primary::make(erm, my_host_id));
+    while (std::optional<dht::partition_range> range = my_ranges.next_partition_range()) {
+        // Note that because of issue #9167 we need to run a separate
+        // query on each partition range, and can't pass several of
+        // them into one partition_range_vector.
+        dht::partition_range_vector partition_ranges;
+        partition_ranges.push_back(std::move(*range));
+        // FIXME: if scanning a single range fails, including network errors,
+        // we fail the entire scan (and rescan from the beginning). Need to
+        // reconsider this. Saving the scan position might be a good enough
+        // solution for this problem.
+        co_await scan_table_ranges(proxy, scan_ctx, std::move(partition_ranges), abort_source, page_sem, expiration_stats);
+    }
+    // If each node only scans its own primary ranges, then when any node is
+    // down part of the token range will not get scanned. This can be viewed
+    // as acceptable (when the comes back online, it will resume its scan),
+    // but as noted in issue #9787, we can allow more prompt expiration
+    // by tasking another node to take over scanning of the dead node's primary
+    // ranges. What we do here is that this node will also check expiration
+    // on its *secondary* ranges - but only those whose primary owner is down.
+    token_ranges_owned_by_this_shard my_secondary_ranges(s, co_await ranges_holder_secondary::make(erm, my_host_id, gossiper));
+    while (std::optional<dht::partition_range> range = my_secondary_ranges.next_partition_range()) {
+        expiration_stats.secondary_ranges_scanned++;
+        dht::partition_range_vector partition_ranges;
+        partition_ranges.push_back(std::move(*range));
+        co_await scan_table_ranges(proxy, scan_ctx, std::move(partition_ranges), abort_source, page_sem, expiration_stats);
     }
     co_return true;
 }

api/api-doc/compaction_manager.json

@@ -246,24 +246,6 @@
             }
          }
       },
-      "sstableinfo":{
-         "id":"sstableinfo",
-         "description":"Compacted sstable information",
-         "properties":{
-            "generation":{
-               "type": "string",
-               "description":"Generation of the sstable"
-            },
-            "origin":{
-               "type":"string",
-               "description":"Origin of the sstable"
-            },
-            "size":{
-               "type":"long",
-               "description":"Size of the sstable"
-            }
-         }
-      },
       "compaction_info" :{
           "id": "compaction_info",
           "description":"A key value mapping",
@@ -345,10 +327,6 @@
                "type":"string",
                "description":"The UUID"
             },
-            "shard_id":{
-               "type":"int",
-               "description":"The shard id the compaction was executed on"
-            },
             "cf":{
                "type":"string",
                "description":"The column family name"
@@ -357,17 +335,9 @@
                "type":"string",
                "description":"The keyspace name"
             },
-            "compaction_type":{
-               "type":"string",
-               "description":"Type of compaction"
-            },
-            "started_at":{
-               "type":"long",
-               "description":"The time compaction started"
-            },
             "compacted_at":{
                "type":"long",
-               "description":"The time compaction completed"
+               "description":"The time of compaction"
             },
             "bytes_in":{
                "type":"long",
@@ -383,32 +353,6 @@
                   "type":"row_merged"
                },
                "description":"The merged rows"
-            },
-            "sstables_in": {
-               "type":"array",
-               "items":{
-                  "type":"sstableinfo"
-               },
-               "description":"List of input sstables for compaction"
-            },
-            "sstables_out": {
-               "type":"array",
-               "items":{
-                  "type":"sstableinfo"
-               },
-               "description":"List of output sstables from compaction"
-            },
-            "total_tombstone_purge_attempt":{
-               "type":"long",
-               "description":"Total number of tombstone purge attempts"
-            },
-            "total_tombstone_purge_failure_due_to_overlapping_with_memtable":{
-               "type":"long",
-               "description":"Number of tombstone purge failures due to data overlapping with memtables"
-            },
-            "total_tombstone_purge_failure_due_to_overlapping_with_uncompacting_sstable":{
-               "type":"long",
-               "description":"Number of tombstone purge failures due to data overlapping with non-compacting sstables"
             }
         }
       }

api/api-doc/gossiper.json

@@ -136,6 +136,14 @@
                      "allowMultiple":false,
                      "type":"string",
                      "paramType":"path"
+                  },
+                  {
+                     "name":"unsafe",
+                     "description":"Set to True to perform an unsafe assassination",
+                     "required":false,
+                     "allowMultiple":false,
+                     "type":"boolean",
+                     "paramType":"query"
                   }
                ]
             }

api/api-doc/storage_service.json

@@ -594,50 +594,6 @@
             }
          ]
       },
-      {
-         "path": "/storage_service/natural_endpoints/v2/{keyspace}",
-         "operations": [
-            {
-               "method": "GET",
-               "summary":"This method returns the N endpoints that are responsible for storing the specified key i.e for replication. the endpoint responsible for this key",
-               "type": "array",
-               "items": {
-                  "type": "string"
-               },
-               "nickname": "get_natural_endpoints_v2",
-               "produces": [
-                  "application/json"
-               ],
-               "parameters": [
-                  {
-                     "name": "keyspace",
-                     "description": "The keyspace to query about.",
-                     "required": true,
-                     "allowMultiple": false,
-                     "type": "string",
-                     "paramType": "path"
-                  },
-                  {
-                     "name": "cf",
-                     "description": "Column family name.",
-                     "required": true,
-                     "allowMultiple": false,
-                     "type": "string",
-                     "paramType": "query"
-                  },
-                  {
-                     "name": "key_component",
-                     "description": "Each component of the key for which we need to find the endpoint (e.g. ?key_component=part1&key_component=part2).",
-                     "required": true,
-                     "allowMultiple": true,
-                     "type": "string",
-                     "paramType": "query"
-                  }
-               ]
-            }
-         ]
-      },
-
       {
          "path":"/storage_service/cdc_streams_check_and_repair",
          "operations":[
@@ -1144,14 +1100,6 @@
                      "allowMultiple":false,
                      "type":"string",
                      "paramType":"query"
-                  },
-                  {
-                     "name": "drop_unfixable_sstables",
-                     "description": "When set to true, drop unfixable sstables. Applies only to scrub mode SEGREGATE.",
-                     "required":false,
-                     "allowMultiple":false,
-                     "type":"boolean",
-                     "paramType":"query"
                   }
                ]
             }
@@ -2196,31 +2144,6 @@
                      "allowMultiple":false,
                      "type":"string",
                      "paramType":"query"
-                  },
-                  {
-                     "name":"skip_cleanup",
-                     "description":"Don't cleanup keys from loaded sstables. Invalid if load_and_stream is true",
-                     "required":false,
-                     "allowMultiple":false,
-                     "type":"string",
-                     "paramType":"query"
-                  },
-                  {
-                     "name":"skip_reshape",
-                     "description":"Don't reshape the loaded sstables. Invalid if load_and_stream is true",
-                     "required":false,
-                     "allowMultiple":false,
-                     "type":"string",
-                     "paramType":"query"
-                  },
-                  {
-                     "name":"scope",
-                     "description":"Defines the set of nodes to which mutations can be streamed",
-                     "required":false,
-                     "allowMultiple":false,
-                     "type":"string",
-                     "paramType":"query",
-                     "enum": ["all", "dc", "rack", "node"]
                   }
                ]
             }
@@ -2973,14 +2896,6 @@
                      "allowMultiple":false,
                      "type":"string",
                      "paramType":"query"
-                  },
-                  {
-                     "name":"incremental_mode",
-                     "description":"Set the incremental repair mode. Can be 'disabled', 'incremental', or 'full'. 'incremental': The incremental repair logic is enabled. Unrepaired sstables will be included for repair. Repaired sstables will be skipped. The incremental repair states will be updated after repair. 'full': The incremental repair logic is enabled. Both repaired and unrepaired sstables will be included for repair. The incremental repair states will be updated after repair. 'disabled': The incremental repair logic is disabled completely. The incremental repair states, e.g., repaired_at in sstables and sstables_repaired_at in the system.tablets table, will not be updated after repair. When the option is not provided, it defaults to incremental mode.",
-                     "required":false,
-                     "allowMultiple":false,
-                     "type":"string",
-                     "paramType":"query"
                   }
                ]
             }
@@ -3112,73 +3027,6 @@
             }
          ]
       },
-      {
-         "path":"/storage_service/retrain_dict",
-         "operations":[
-            {
-               "method":"POST",
-               "summary":"Retrain the SSTable compression dictionary for the target table.",
-               "type":"void",
-               "nickname":"retrain_dict",
-               "produces":[
-                  "application/json"
-               ],
-               "parameters":[
-                  {
-                     "name":"keyspace",
-                     "description":"Name of the keyspace containing the target table.",
-                     "required":true,
-                     "allowMultiple":false,
-                     "type":"string",
-                     "paramType":"query"
-                  },
-                  {
-                     "name":"cf",
-                     "description":"Name of the target table.",
-                     "required":true,
-                     "allowMultiple":false,
-                     "type":"string",
-                     "paramType":"query"
-                  }
-               ]
-            }
-         ]
-      },
-      {
-         "path":"/storage_service/estimate_compression_ratios",
-         "operations":[
-            {
-               "method":"GET",
-               "summary":"Compute an estimated compression ratio for SSTables of the given table, for various compression configurations.",
-               "type":"array",
-               "items":{
-                  "type":"compression_config_result"
-               },
-               "nickname":"estimate_compression_ratios",
-               "produces":[
-                  "application/json"
-               ],
-               "parameters":[
-                  {
-                     "name":"keyspace",
-                     "description":"Name of the keyspace containing the target table.",
-                     "required":true,
-                     "allowMultiple":false,
-                     "type":"string",
-                     "paramType":"query"
-                  },
-                  {
-                     "name":"cf",
-                     "description":"Name of the target table.",
-                     "required":true,
-                     "allowMultiple":false,
-                     "type":"string",
-                     "paramType":"query"
-                  }
-               ]
-            }
-         ]
-      },
       {
          "path":"/storage_service/raft_topology/reload",
          "operations":[
@@ -3221,54 +3069,6 @@
                ]
             }
          ]
-      },
-      {
-         "path":"/storage_service/raft_topology/cmd_rpc_status",
-         "operations":[
-            {
-               "method":"GET",
-               "summary":"Get information about currently running topology cmd rpc",
-               "type":"string",
-               "nickname":"raft_topology_get_cmd_status",
-               "produces":[
-                  "application/json"
-               ],
-               "parameters":[
-               ]
-            }
-         ]
-      },
-      {
-         "path":"/storage_service/drop_quarantined_sstables",
-         "operations":[
-            {
-               "method":"POST",
-               "summary":"Drops all quarantined sstables in all keyspaces or specified keyspace and tables",
-               "type":"void",
-               "nickname":"drop_quarantined_sstables",
-               "produces":[
-                  "application/json"
-               ],
-               "parameters":[
-                  {
-                     "name":"keyspace",
-                     "description":"The keyspace name to drop quarantined sstables from.",
-                     "required":false,
-                     "allowMultiple":false,
-                     "type":"string",
-                     "paramType":"query"
-                  },
-                  {
-                     "name":"tables",
-                     "description":"Comma-separated table names to drop quarantined sstables from.",
-                     "required":false,
-                     "allowMultiple":false,
-                     "type":"string",
-                     "paramType":"query"
-                  }
-               ]
-            }
-         ]
       }
    ],
    "models":{
@@ -3405,11 +3205,11 @@
          "properties":{
             "start_token":{
                "type":"string",
-               "description":"The range start token (exclusive)"
+               "description":"The range start token"
             },
             "end_token":{
                "type":"string",
-               "description":"The range end token (inclusive)"
+               "description":"The range start token"
             },
             "endpoints":{
                "type":"array",
@@ -3482,7 +3282,7 @@
             "version":{
                "type":"string",
                "enum":[
-                  "ka", "la", "mc", "md", "me", "ms"
+                  "ka", "la", "mc", "md", "me"
                ],
                "description":"SSTable version"
             },
@@ -3528,32 +3328,6 @@
                 "type":"string"
             }
         }
-      },
-      "compression_config_result":{
-         "id":"compression_config_result",
-         "description":"Compression ratio estimation result for one config",
-         "properties":{
-            "level":{
-               "type":"long",
-               "description":"The used value of `compression_level`"
-            },
-            "chunk_length_in_kb":{
-               "type":"long",
-               "description":"The used value of `chunk_length_in_kb`"
-            },
-            "dict":{
-               "type":"string",
-               "description":"The used dictionary: `none`, `past` (== current), or `future`"
-            },
-            "sstable_compression":{
-               "type":"string",
-               "description":"The used compressor name (aka `sstable_compression`)"
-            },
-            "ratio":{
-               "type":"float",
-               "description":"The resulting compression ratio (estimated on a random sample of files)"
-            }
-         }
       }
    }
 }

api/api-doc/task_manager.json

@@ -447,7 +447,7 @@
                "description":"The number of units completed so far"
             },
             "children_ids":{
-               "type":"chunked_array",
+               "type":"array",
                "items":{
                   "type":"task_identity"
                },

api/api.cc

@@ -137,6 +137,14 @@ future<> unset_load_meter(http_context& ctx) {
     return ctx.http_server.set_routes([&ctx] (routes& r) { unset_load_meter(ctx, r); });
 }
 
+future<> set_format_selector(http_context& ctx, db::sstables_format_selector& sel) {
+    return ctx.http_server.set_routes([&ctx, &sel] (routes& r) { set_format_selector(ctx, r, sel); });
+}
+
+future<> unset_format_selector(http_context& ctx) {
+    return ctx.http_server.set_routes([&ctx] (routes& r) { unset_format_selector(ctx, r); });
+}
+
 future<> set_server_sstables_loader(http_context& ctx, sharded<sstables_loader>& sst_loader) {
     return ctx.http_server.set_routes([&ctx, &sst_loader] (routes& r) { set_sstables_loader(ctx, r, sst_loader); });
 }
@@ -216,22 +224,15 @@ future<> unset_server_gossip(http_context& ctx) {
     });
 }
 
-future<> set_server_column_family(http_context& ctx, sharded<replica::database>& db) {
-    co_await register_api(ctx, "column_family",
-                "The column family API", [&db] (http_context& ctx, routes& r) {
-                    set_column_family(ctx, r, db);
-                });
-    co_await register_api(ctx, "cache_service",
-            "The cache service API", [&db] (http_context& ctx, routes& r) {
-                    set_cache_service(ctx, db, r);
+future<> set_server_column_family(http_context& ctx, sharded<db::system_keyspace>& sys_ks) {
+    return register_api(ctx, "column_family",
+                "The column family API", [&sys_ks] (http_context& ctx, routes& r) {
+                    set_column_family(ctx, r, sys_ks);
                 });
 }
 
 future<> unset_server_column_family(http_context& ctx) {
-    return ctx.http_server.set_routes([&ctx] (routes& r) {
-        unset_column_family(ctx, r);
-        unset_cache_service(ctx, r);
-    });
+    return ctx.http_server.set_routes([&ctx] (routes& r) { unset_column_family(ctx, r); });
 }
 
 future<> set_server_messaging_service(http_context& ctx, sharded<netw::messaging_service>& ms) {
@@ -263,6 +264,15 @@ future<> unset_server_stream_manager(http_context& ctx) {
     return ctx.http_server.set_routes([&ctx] (routes& r) { unset_stream_manager(ctx, r); });
 }
 
+future<> set_server_cache(http_context& ctx) {
+    return register_api(ctx, "cache_service",
+            "The cache service API", set_cache_service);
+}
+
+future<> unset_server_cache(http_context& ctx) {
+    return ctx.http_server.set_routes([&ctx] (routes& r) { unset_cache_service(ctx, r); });
+}
+
 future<> set_hinted_handoff(http_context& ctx, sharded<service::storage_proxy>& proxy, sharded<gms::gossiper>& g) {
     return register_api(ctx, "hinted_handoff",
                 "The hinted handoff API", [&proxy, &g] (http_context& ctx, routes& r) {
@@ -274,7 +284,7 @@ future<> unset_hinted_handoff(http_context& ctx) {
     return ctx.http_server.set_routes([&ctx] (routes& r) { unset_hinted_handoff(ctx, r); });
 }
 
-future<> set_server_compaction_manager(http_context& ctx, sharded<compaction::compaction_manager>& cm) {
+future<> set_server_compaction_manager(http_context& ctx, sharded<compaction_manager>& cm) {
     return register_api(ctx, "compaction_manager", "The Compaction manager API", [&cm] (http_context& ctx, routes& r) {
         set_compaction_manager(ctx, r, cm);
     });
@@ -381,5 +391,32 @@ future<> unset_server_raft(http_context& ctx) {
     return ctx.http_server.set_routes([&ctx] (routes& r) { unset_raft(ctx, r); });
 }
 
+void req_params::process(const request& req) {
+    // Process mandatory parameters
+    for (auto& [name, ent] : params) {
+        if (!ent.is_mandatory) {
+            continue;
+        }
+        try {
+            ent.value = req.get_path_param(name);
+        } catch (std::out_of_range&) {
+            throw httpd::bad_param_exception(fmt::format("Mandatory parameter '{}' was not provided", name));
+        }
+    }
+
+    // Process optional parameters
+    for (auto& [name, value] : req.query_parameters) {
+        try {
+            auto& ent = params.at(name);
+            if (ent.is_mandatory) {
+                throw httpd::bad_param_exception(fmt::format("Parameter '{}' is expected to be provided as part of the request url", name));
+            }
+            ent.value = value;
+        } catch (std::out_of_range&) {
+            throw httpd::bad_param_exception(fmt::format("Unsupported optional parameter '{}'", name));
+        }
+    }
+}
+
 }
 

api/api.hh

@@ -23,6 +23,17 @@
 
 namespace api {
 
+template<class T>
+std::vector<sstring> container_to_vec(const T& container) {
+    std::vector<sstring> res;
+    res.reserve(std::size(container));
+
+    for (const auto& i : container) {
+        res.push_back(fmt::to_string(i));
+    }
+    return res;
+}
+
 template<class T>
 std::vector<T> map_to_key_value(const std::map<sstring, sstring>& map) {
     std::vector<T> res;
@@ -56,6 +67,17 @@ T map_sum(T&& dest, const S& src) {
     return std::move(dest);
 }
 
+template <typename MAP>
+std::vector<sstring> map_keys(const MAP& map) {
+    std::vector<sstring> res;
+    res.reserve(std::size(map));
+
+    for (const auto& i : map) {
+        res.push_back(fmt::to_string(i.first));
+    }
+    return res;
+}
+
 /**
  * General sstring splitting function
  */
@@ -73,7 +95,7 @@ inline std::vector<sstring> split(const sstring& text, const char* separator) {
  *
  */
 template<class T, class F, class V>
-future<json::json_return_type>  sum_stats(sharded<T>& d, V F::*f) {
+future<json::json_return_type>  sum_stats(distributed<T>& d, V F::*f) {
     return d.map_reduce0([f](const T& p) {return p.get_stats().*f;}, 0,
             std::plus<V>()).then([](V val) {
         return make_ready_future<json::json_return_type>(val);
@@ -106,7 +128,7 @@ httpd::utils_json::rate_moving_average_and_histogram timer_to_json(const utils::
 }
 
 template<class T, class F>
-future<json::json_return_type>  sum_histogram_stats(sharded<T>& d, utils::timed_rate_moving_average_and_histogram F::*f) {
+future<json::json_return_type>  sum_histogram_stats(distributed<T>& d, utils::timed_rate_moving_average_and_histogram F::*f) {
 
     return d.map_reduce0([f](const T& p) {return (p.get_stats().*f).hist;}, utils::ihistogram(),
             std::plus<utils::ihistogram>()).then([](const utils::ihistogram& val) {
@@ -115,7 +137,7 @@ future<json::json_return_type>  sum_histogram_stats(sharded<T>& d, utils::timed_
 }
 
 template<class T, class F>
-future<json::json_return_type>  sum_timer_stats(sharded<T>& d, utils::timed_rate_moving_average_and_histogram F::*f) {
+future<json::json_return_type>  sum_timer_stats(distributed<T>& d, utils::timed_rate_moving_average_and_histogram F::*f) {
 
     return d.map_reduce0([f](const T& p) {return (p.get_stats().*f).rate();}, utils::rate_moving_average_and_histogram(),
             std::plus<utils::rate_moving_average_and_histogram>()).then([](const utils::rate_moving_average_and_histogram& val) {
@@ -124,7 +146,7 @@ future<json::json_return_type>  sum_timer_stats(sharded<T>& d, utils::timed_rate
 }
 
 template<class T, class F>
-future<json::json_return_type>  sum_timer_stats(sharded<T>& d, utils::timed_rate_moving_average_summary_and_histogram F::*f) {
+future<json::json_return_type>  sum_timer_stats(distributed<T>& d, utils::timed_rate_moving_average_summary_and_histogram F::*f) {
     return d.map_reduce0([f](const T& p) {return (p.get_stats().*f).rate();}, utils::rate_moving_average_and_histogram(),
             std::plus<utils::rate_moving_average_and_histogram>()).then([](const utils::rate_moving_average_and_histogram& val) {
         return make_ready_future<json::json_return_type>(timer_to_json(val));
@@ -230,6 +252,67 @@ public:
     operator T() const { return value; }
 };
 
+using mandatory = bool_class<struct mandatory_tag>;
+
+class req_params {
+public:
+    struct def {
+        std::optional<sstring> value;
+        mandatory is_mandatory = mandatory::no;
+
+        def(std::optional<sstring> value_ = std::nullopt, mandatory is_mandatory_ = mandatory::no)
+            : value(std::move(value_))
+            , is_mandatory(is_mandatory_)
+        { }
+
+        def(mandatory is_mandatory_)
+            : is_mandatory(is_mandatory_)
+        { }
+    };
+
+private:
+    std::unordered_map<sstring, def> params;
+
+public:
+    req_params(std::initializer_list<std::pair<sstring, def>> l) {
+        for (const auto& [name, ent] : l) {
+            add(std::move(name), std::move(ent));
+        }
+    }
+
+    void add(sstring name, def ent) {
+        params.emplace(std::move(name), std::move(ent));
+    }
+
+    void process(const request& req);
+
+    const std::optional<sstring>& get(const char* name) const {
+        return params.at(name).value;
+    }
+
+    template <typename T = sstring>
+    const std::optional<T> get_as(const char* name) const {
+        return get(name);
+    }
+
+    template <typename T = sstring>
+    requires std::same_as<T, bool>
+    const std::optional<bool> get_as(const char* name) const {
+        auto value = get(name);
+        if (!value) {
+            return std::nullopt;
+        }
+        std::transform(value->begin(), value->end(), value->begin(), ::tolower);
+        if (value == "true" || value == "yes" || value == "1") {
+            return true;
+        }
+        if (value == "false" || value == "no" || value == "0") {
+            return false;
+        }
+        throw boost::bad_lexical_cast{};
+    }
+};
+
 httpd::utils_json::estimated_histogram time_to_json_histogram(const utils::time_estimated_histogram& val);
 
 }

api/api_init.hh

@@ -18,9 +18,7 @@
 using request = http::request;
 using reply = http::reply;
 
-namespace compaction {
 class compaction_manager;
-}
 
 namespace service {
 
@@ -58,6 +56,7 @@ class sstables_format_selector;
 namespace view {
 class view_builder;
 }
+class system_keyspace;
 }
 namespace netw { class messaging_service; }
 class repair_service;
@@ -84,9 +83,9 @@ struct http_context {
     sstring api_dir;
     sstring api_doc;
     httpd::http_server_control http_server;
-    sharded<replica::database>& db;
+    distributed<replica::database>& db;
 
-    http_context(sharded<replica::database>& _db)
+    http_context(distributed<replica::database>& _db)
             : db(_db)
     {
     }
@@ -117,7 +116,7 @@ future<> set_server_token_metadata(http_context& ctx, sharded<locator::shared_to
 future<> unset_server_token_metadata(http_context& ctx);
 future<> set_server_gossip(http_context& ctx, sharded<gms::gossiper>& g);
 future<> unset_server_gossip(http_context& ctx);
-future<> set_server_column_family(http_context& ctx, sharded<replica::database>& db);
+future<> set_server_column_family(http_context& ctx, sharded<db::system_keyspace>& sys_ks);
 future<> unset_server_column_family(http_context& ctx);
 future<> set_server_messaging_service(http_context& ctx, sharded<netw::messaging_service>& ms);
 future<> unset_server_messaging_service(http_context& ctx);
@@ -127,7 +126,9 @@ future<> set_server_stream_manager(http_context& ctx, sharded<streaming::stream_
 future<> unset_server_stream_manager(http_context& ctx);
 future<> set_hinted_handoff(http_context& ctx, sharded<service::storage_proxy>& p, sharded<gms::gossiper>& g);
 future<> unset_hinted_handoff(http_context& ctx);
-future<> set_server_compaction_manager(http_context& ctx, sharded<compaction::compaction_manager>& cm);
+future<> set_server_cache(http_context& ctx);
+future<> unset_server_cache(http_context& ctx);
+future<> set_server_compaction_manager(http_context& ctx, sharded<compaction_manager>& cm);
 future<> unset_server_compaction_manager(http_context& ctx);
 future<> set_server_done(http_context& ctx);
 future<> set_server_task_manager(http_context& ctx, sharded<tasks::task_manager>& tm, lw_shared_ptr<db::config> cfg, sharded<gms::gossiper>& gossiper);
@@ -140,6 +141,8 @@ future<> set_server_raft(http_context&, sharded<service::raft_group_registry>&);
 future<> unset_server_raft(http_context&);
 future<> set_load_meter(http_context& ctx, service::load_meter& lm);
 future<> unset_load_meter(http_context& ctx);
+future<> set_format_selector(http_context& ctx, db::sstables_format_selector& sel);
+future<> unset_format_selector(http_context& ctx);
 future<> set_server_cql_server_test(http_context& ctx, cql_transport::controller& ctl);
 future<> unset_server_cql_server_test(http_context& ctx);
 future<> set_server_service_levels(http_context& ctx, cql_transport::controller& ctl, sharded<cql3::query_processor>& qp);

api/cache_service.cc

@@ -16,7 +16,7 @@ using namespace json;
 using namespace seastar::httpd;
 namespace cs = httpd::cache_service_json;
 
-void set_cache_service(http_context& ctx, sharded<replica::database>& db, routes& r) {
+void set_cache_service(http_context& ctx, routes& r) {
     cs::get_row_cache_save_period_in_seconds.set(r, [](std::unique_ptr<http::request> req) {
         // We never save the cache
         // Origin uses 0 for never
@@ -204,53 +204,53 @@ void set_cache_service(http_context& ctx, sharded<replica::database>& db, routes
         });
     });
 
-    cs::get_row_hits.set(r, [&db] (std::unique_ptr<http::request> req) {
-        return map_reduce_cf(db, uint64_t(0), [](const replica::column_family& cf) {
+    cs::get_row_hits.set(r, [&ctx] (std::unique_ptr<http::request> req) {
+        return map_reduce_cf(ctx, uint64_t(0), [](const replica::column_family& cf) {
             return cf.get_row_cache().stats().hits.count();
         }, std::plus<uint64_t>());
     });
 
-    cs::get_row_requests.set(r, [&db] (std::unique_ptr<http::request> req) {
-        return map_reduce_cf(db, uint64_t(0), [](const replica::column_family& cf) {
+    cs::get_row_requests.set(r, [&ctx] (std::unique_ptr<http::request> req) {
+        return map_reduce_cf(ctx, uint64_t(0), [](const replica::column_family& cf) {
             return cf.get_row_cache().stats().hits.count() + cf.get_row_cache().stats().misses.count();
         }, std::plus<uint64_t>());
     });
 
-    cs::get_row_hit_rate.set(r, [&db] (std::unique_ptr<http::request> req) {
-        return map_reduce_cf(db, ratio_holder(), [](const replica::column_family& cf) {
+    cs::get_row_hit_rate.set(r, [&ctx] (std::unique_ptr<http::request> req) {
+        return map_reduce_cf(ctx, ratio_holder(), [](const replica::column_family& cf) {
             return ratio_holder(cf.get_row_cache().stats().hits.count() + cf.get_row_cache().stats().misses.count(),
                     cf.get_row_cache().stats().hits.count());
         }, std::plus<ratio_holder>());
     });
 
-    cs::get_row_hits_moving_avrage.set(r, [&db] (std::unique_ptr<http::request> req) {
-        return map_reduce_cf_raw(db, utils::rate_moving_average(), [](const replica::column_family& cf) {
+    cs::get_row_hits_moving_avrage.set(r, [&ctx] (std::unique_ptr<http::request> req) {
+        return map_reduce_cf_raw(ctx, utils::rate_moving_average(), [](const replica::column_family& cf) {
             return cf.get_row_cache().stats().hits.rate();
         }, std::plus<utils::rate_moving_average>()).then([](const utils::rate_moving_average& m) {
             return make_ready_future<json::json_return_type>(meter_to_json(m));
         });
     });
 
-    cs::get_row_requests_moving_avrage.set(r, [&db] (std::unique_ptr<http::request> req) {
-        return map_reduce_cf_raw(db, utils::rate_moving_average(), [](const replica::column_family& cf) {
+    cs::get_row_requests_moving_avrage.set(r, [&ctx] (std::unique_ptr<http::request> req) {
+        return map_reduce_cf_raw(ctx, utils::rate_moving_average(), [](const replica::column_family& cf) {
             return cf.get_row_cache().stats().hits.rate() + cf.get_row_cache().stats().misses.rate();
         }, std::plus<utils::rate_moving_average>()).then([](const utils::rate_moving_average& m) {
             return make_ready_future<json::json_return_type>(meter_to_json(m));
         });
     });
 
-    cs::get_row_size.set(r, [&db] (std::unique_ptr<http::request> req) {
+    cs::get_row_size.set(r, [&ctx] (std::unique_ptr<http::request> req) {
         // In origin row size is the weighted size.
         // We currently do not support weights, so we use raw size in bytes instead
-        return db.map_reduce0([](replica::database& db) -> uint64_t {
+        return ctx.db.map_reduce0([](replica::database& db) -> uint64_t {
             return db.row_cache_tracker().region().occupancy().used_space();
         }, uint64_t(0), std::plus<uint64_t>()).then([](const int64_t& res) {
             return make_ready_future<json::json_return_type>(res);
         });
     });
 
-    cs::get_row_entries.set(r, [&db] (std::unique_ptr<http::request> req) {
-        return db.map_reduce0([](replica::database& db) -> uint64_t {
+    cs::get_row_entries.set(r, [&ctx] (std::unique_ptr<http::request> req) {
+        return ctx.db.map_reduce0([](replica::database& db) -> uint64_t {
             return db.row_cache_tracker().partitions();
         }, uint64_t(0), std::plus<uint64_t>()).then([](const int64_t& res) {
             return make_ready_future<json::json_return_type>(res);

api/cache_service.hh

@@ -7,20 +7,15 @@
  */
 
 #pragma once
-#include <seastar/core/sharded.hh>
 
 namespace seastar::httpd {
 class routes;
 }
 
-namespace replica {
-class database;
-}
-
 namespace api {
 
 struct http_context;
-void set_cache_service(http_context& ctx, seastar::sharded<replica::database>& db, seastar::httpd::routes& r);
+void set_cache_service(http_context& ctx, seastar::httpd::routes& r);
 void unset_cache_service(http_context& ctx, seastar::httpd::routes& r);
 
 }

api/column_family.hh

@@ -13,25 +13,25 @@
 #include <any>
 #include "api/api_init.hh"
 
+namespace db {
+class system_keyspace;
+}
+
 namespace api {
 
-void set_column_family(http_context& ctx, httpd::routes& r, sharded<replica::database>& db);
+void set_column_family(http_context& ctx, httpd::routes& r, sharded<db::system_keyspace>& sys_ks);
 void unset_column_family(http_context& ctx, httpd::routes& r);
 
 table_info parse_table_info(const sstring& name, const replica::database& db);
 
 template<class Mapper, class I, class Reducer>
-future<I> map_reduce_cf_raw(sharded<replica::database>& db, const sstring& name, I init,
+future<I> map_reduce_cf_raw(http_context& ctx, const sstring& name, I init,
         Mapper mapper, Reducer reducer) {
-    auto uuid = parse_table_info(name, db.local()).id;
-    using mapper_type = std::function<future<std::unique_ptr<std::any>>(replica::database&)>;
+    auto uuid = parse_table_info(name, ctx.db.local()).id;
+    using mapper_type = std::function<std::unique_ptr<std::any>(replica::database&)>;
     using reducer_type = std::function<std::unique_ptr<std::any>(std::unique_ptr<std::any>, std::unique_ptr<std::any>)>;
-    return db.map_reduce0(mapper_type([mapper, uuid](replica::database& db) {
-        return futurize_invoke([mapper, &db, uuid] {
-            return mapper(db.find_column_family(uuid));
-        }).then([] (auto result) {
-            return std::make_unique<std::any>(I(std::move(result)));
-        });
+    return ctx.db.map_reduce0(mapper_type([mapper, uuid](replica::database& db) {
+        return std::make_unique<std::any>(I(mapper(db.find_column_family(uuid))));
     }), std::make_unique<std::any>(std::move(init)), reducer_type([reducer = std::move(reducer)] (std::unique_ptr<std::any> a, std::unique_ptr<std::any> b) mutable {
         return std::make_unique<std::any>(I(reducer(std::any_cast<I>(std::move(*a)), std::any_cast<I>(std::move(*b)))));
     })).then([] (std::unique_ptr<std::any> r) {
@@ -41,30 +41,33 @@ future<I> map_reduce_cf_raw(sharded<replica::database>& db, const sstring& name,
 
 
 template<class Mapper, class I, class Reducer>
-future<json::json_return_type> map_reduce_cf(sharded<replica::database>& db, const sstring& name, I init,
+future<json::json_return_type> map_reduce_cf(http_context& ctx, const sstring& name, I init,
         Mapper mapper, Reducer reducer) {
-    return map_reduce_cf_raw(db, name, init, mapper, reducer).then([](const I& res) {
+    return map_reduce_cf_raw(ctx, name, init, mapper, reducer).then([](const I& res) {
         return make_ready_future<json::json_return_type>(res);
     });
 }
 
 template<class Mapper, class I, class Reducer, class Result>
-future<json::json_return_type> map_reduce_cf(sharded<replica::database>& db, const sstring& name, I init,
+future<json::json_return_type> map_reduce_cf(http_context& ctx, const sstring& name, I init,
         Mapper mapper, Reducer reducer, Result result) {
-    return map_reduce_cf_raw(db, name, init, mapper, reducer).then([result](const I& res) mutable {
+    return map_reduce_cf_raw(ctx, name, init, mapper, reducer).then([result](const I& res) mutable {
         result = res;
         return make_ready_future<json::json_return_type>(result);
     });
 }
 
+future<json::json_return_type> map_reduce_cf_time_histogram(http_context& ctx, const sstring& name, std::function<utils::time_estimated_histogram(const replica::column_family&)> f);
+
 struct map_reduce_column_families_locally {
     std::any init;
-    std::function<future<std::unique_ptr<std::any>>(replica::column_family&)> mapper;
+    std::function<std::unique_ptr<std::any>(replica::column_family&)> mapper;
     std::function<std::unique_ptr<std::any>(std::unique_ptr<std::any>, std::unique_ptr<std::any>)> reducer;
     future<std::unique_ptr<std::any>> operator()(replica::database& db) const {
         auto res = seastar::make_lw_shared<std::unique_ptr<std::any>>(std::make_unique<std::any>(init));
-        return db.get_tables_metadata().for_each_table_gently([res, this] (table_id, seastar::lw_shared_ptr<replica::table> table) -> future<> {
-            *res = reducer(std::move(*res), co_await mapper(*table.get()));
+        return db.get_tables_metadata().for_each_table_gently([res, this] (table_id, seastar::lw_shared_ptr<replica::table> table) {
+            *res = reducer(std::move(*res), mapper(*table.get()));
+            return make_ready_future();
         }).then([res] () {
             return std::move(*res);
         });
@@ -72,21 +75,17 @@ struct map_reduce_column_families_locally {
 };
 
 template<class Mapper, class I, class Reducer>
-future<I> map_reduce_cf_raw(sharded<replica::database>& db, I init,
+future<I> map_reduce_cf_raw(http_context& ctx, I init,
         Mapper mapper, Reducer reducer) {
-    using mapper_type = std::function<future<std::unique_ptr<std::any>>(replica::column_family&)>;
+    using mapper_type = std::function<std::unique_ptr<std::any>(replica::column_family&)>;
     using reducer_type = std::function<std::unique_ptr<std::any>(std::unique_ptr<std::any>, std::unique_ptr<std::any>)>;
     auto wrapped_mapper = mapper_type([mapper = std::move(mapper)] (replica::column_family& cf) mutable {
-        return futurize_invoke([&cf, mapper] {
-            return mapper(cf);
-        }).then([] (auto result) {
-            return std::make_unique<std::any>(I(std::move(result)));
-        });
+        return std::make_unique<std::any>(I(mapper(cf)));
     });
     auto wrapped_reducer = reducer_type([reducer = std::move(reducer)] (std::unique_ptr<std::any> a, std::unique_ptr<std::any> b) mutable {
         return std::make_unique<std::any>(I(reducer(std::any_cast<I>(std::move(*a)), std::any_cast<I>(std::move(*b)))));
     });
-    return db.map_reduce0(map_reduce_column_families_locally{init,
+    return ctx.db.map_reduce0(map_reduce_column_families_locally{init,
             std::move(wrapped_mapper), wrapped_reducer}, std::make_unique<std::any>(init), wrapped_reducer).then([] (std::unique_ptr<std::any> res) {
         return std::any_cast<I>(std::move(*res));
     });
@@ -94,13 +93,20 @@ future<I> map_reduce_cf_raw(sharded<replica::database>& db, I init,
 
 
 template<class Mapper, class I, class Reducer>
-future<json::json_return_type> map_reduce_cf(sharded<replica::database>& db, I init,
+future<json::json_return_type> map_reduce_cf(http_context& ctx, I init,
         Mapper mapper, Reducer reducer) {
-    return map_reduce_cf_raw(db, init, mapper, reducer).then([](const I& res) {
+    return map_reduce_cf_raw(ctx, init, mapper, reducer).then([](const I& res) {
         return make_ready_future<json::json_return_type>(res);
     });
 }
 
+future<json::json_return_type>  get_cf_stats(http_context& ctx, const sstring& name,
+        int64_t replica::column_family_stats::*f);
+
+future<json::json_return_type>  get_cf_stats(http_context& ctx,
+        int64_t replica::column_family_stats::*f);
+
+
 std::tuple<sstring, sstring> parse_fully_qualified_cf_name(sstring name);
 
 }

api/compaction_manager.cc

@@ -14,7 +14,6 @@
 #include "api/api.hh"
 #include "api/api-doc/compaction_manager.json.hh"
 #include "api/api-doc/storage_service.json.hh"
-#include "db/compaction_history_entry.hh"
 #include "db/system_keyspace.hh"
 #include "column_family.hh"
 #include "unimplemented.hh"
@@ -29,9 +28,9 @@ namespace ss = httpd::storage_service_json;
 using namespace json;
 using namespace seastar::httpd;
 
-static future<json::json_return_type> get_cm_stats(sharded<compaction::compaction_manager>& cm,
-        int64_t compaction::compaction_manager::stats::*f) {
-    return cm.map_reduce0([f](compaction::compaction_manager& cm) {
+static future<json::json_return_type> get_cm_stats(sharded<compaction_manager>& cm,
+        int64_t compaction_manager::stats::*f) {
+    return cm.map_reduce0([f](compaction_manager& cm) {
         return cm.get_stats().*f;
     }, int64_t(0), std::plus<int64_t>()).then([](const int64_t& res) {
         return make_ready_future<json::json_return_type>(res);
@@ -47,9 +46,9 @@ static std::unordered_map<std::pair<sstring, sstring>, uint64_t, utils::tuple_ha
     return std::move(a);
 }
 
-void set_compaction_manager(http_context& ctx, routes& r, sharded<compaction::compaction_manager>& cm) {
+void set_compaction_manager(http_context& ctx, routes& r, sharded<compaction_manager>& cm) {
     cm::get_compactions.set(r, [&cm] (std::unique_ptr<http::request> req) {
-        return cm.map_reduce0([](compaction::compaction_manager& cm) {
+        return cm.map_reduce0([](compaction_manager& cm) {
             std::vector<cm::summary> summaries;
 
             for (const auto& c : cm.get_compactions()) {
@@ -58,7 +57,7 @@ void set_compaction_manager(http_context& ctx, routes& r, sharded<compaction::co
                 s.ks = c.ks_name;
                 s.cf = c.cf_name;
                 s.unit = "keys";
-                s.task_type = compaction::compaction_name(c.type);
+                s.task_type = sstables::compaction_name(c.type);
                 s.completed = c.total_keys_written;
                 s.total = c.total_partitions;
                 summaries.push_back(std::move(s));
@@ -72,9 +71,10 @@ void set_compaction_manager(http_context& ctx, routes& r, sharded<compaction::co
     cm::get_pending_tasks_by_table.set(r, [&ctx] (std::unique_ptr<http::request> req) {
         return ctx.db.map_reduce0([](replica::database& db) {
             return do_with(std::unordered_map<std::pair<sstring, sstring>, uint64_t, utils::tuple_hash>(), [&db](std::unordered_map<std::pair<sstring, sstring>, uint64_t, utils::tuple_hash>& tasks) {
-                return db.get_tables_metadata().for_each_table_gently([&tasks] (table_id, lw_shared_ptr<replica::table> table) -> future<> {
+                return db.get_tables_metadata().for_each_table_gently([&tasks] (table_id, lw_shared_ptr<replica::table> table) {
                     replica::table& cf = *table.get();
-                    tasks[std::make_pair(cf.schema()->ks_name(), cf.schema()->cf_name())] = co_await cf.estimate_pending_compactions();
+                    tasks[std::make_pair(cf.schema()->ks_name(), cf.schema()->cf_name())] = cf.estimate_pending_compactions();
+                    return make_ready_future<>();
                 }).then([&tasks] {
                     return std::move(tasks);
                 });
@@ -103,20 +103,23 @@ void set_compaction_manager(http_context& ctx, routes& r, sharded<compaction::co
 
     cm::stop_compaction.set(r, [&cm] (std::unique_ptr<http::request> req) {
         auto type = req->get_query_param("type");
-        return cm.invoke_on_all([type] (compaction::compaction_manager& cm) {
+        return cm.invoke_on_all([type] (compaction_manager& cm) {
             return cm.stop_compaction(type);
         }).then([] {
             return make_ready_future<json::json_return_type>(json_void());
         });
     });
 
-    cm::stop_keyspace_compaction.set(r, [&ctx, &cm] (std::unique_ptr<http::request> req) -> future<json::json_return_type> {
-        auto [ks_name, tables] = parse_table_infos(ctx, *req, "tables");
+    cm::stop_keyspace_compaction.set(r, [&ctx] (std::unique_ptr<http::request> req) -> future<json::json_return_type> {
+        auto ks_name = validate_keyspace(ctx, req);
+        auto tables = parse_table_infos(ks_name, ctx, req->query_parameters, "tables");
         auto type = req->get_query_param("type");
-        co_await cm.invoke_on_all([&] (compaction::compaction_manager& cm) {
+        co_await ctx.db.invoke_on_all([&] (replica::database& db) {
+            auto& cm = db.get_compaction_manager();
             return parallel_for_each(tables, [&] (const table_info& ti) {
-                return cm.stop_compaction(type, [id = ti.id] (const compaction::compaction_group_view* x) {
-                    return x->schema()->id() == id;
+                auto& t = db.find_column_family(ti.id);
+                return t.parallel_foreach_table_state([&] (compaction::table_state& ts) {
+                    return cm.stop_compaction(type, &ts);
                 });
             });
         });
@@ -124,13 +127,13 @@ void set_compaction_manager(http_context& ctx, routes& r, sharded<compaction::co
     });
 
     cm::get_pending_tasks.set(r, [&ctx] (std::unique_ptr<http::request> req) {
-        return map_reduce_cf(ctx.db, int64_t(0), [](replica::column_family& cf) {
+        return map_reduce_cf(ctx, int64_t(0), [](replica::column_family& cf) {
             return cf.estimate_pending_compactions();
         }, std::plus<int64_t>());
     });
 
     cm::get_completed_tasks.set(r, [&cm] (std::unique_ptr<http::request> req) {
-        return get_cm_stats(cm, &compaction::compaction_manager::stats::completed_tasks);
+        return get_cm_stats(cm, &compaction_manager::stats::completed_tasks);
     });
 
     cm::get_total_compactions_completed.set(r, [] (std::unique_ptr<http::request> req) {
@@ -148,7 +151,7 @@ void set_compaction_manager(http_context& ctx, routes& r, sharded<compaction::co
     });
 
     cm::get_compaction_history.set(r, [&cm] (std::unique_ptr<http::request> req) {
-        noncopyable_function<future<>(output_stream<char>&&)> f = [&cm] (output_stream<char>&& out) -> future<> {
+        std::function<future<>(output_stream<char>&&)> f = [&cm] (output_stream<char>&& out) -> future<> {
             auto s = std::move(out);
             bool first = true;
             std::exception_ptr ex;
@@ -157,11 +160,8 @@ void set_compaction_manager(http_context& ctx, routes& r, sharded<compaction::co
                 co_await cm.local().get_compaction_history([&s, &first](const db::compaction_history_entry& entry) mutable -> future<> {
                         cm::history h;
                         h.id = fmt::to_string(entry.id);
-                        h.shard_id = entry.shard_id;
                         h.ks = std::move(entry.ks);
                         h.cf = std::move(entry.cf);
-                        h.compaction_type = entry.compaction_type;
-                        h.started_at = entry.started_at;
                         h.compacted_at = entry.compacted_at;
                         h.bytes_in = entry.bytes_in;
                         h.bytes_out =  entry.bytes_out;
@@ -173,24 +173,6 @@ void set_compaction_manager(http_context& ctx, routes& r, sharded<compaction::co
                             e.value = it.second;
                             h.rows_merged.push(std::move(e));
                         }
-                        for (const auto& data : entry.sstables_in) {
-                            httpd::compaction_manager_json::sstableinfo sstable;
-                            sstable.generation = fmt::to_string(data.generation),
-                            sstable.origin = data.origin,
-                            sstable.size = data.size,
-                            h.sstables_in.push(std::move(sstable));
-                        }
-                        for (const auto& data : entry.sstables_out) {
-                            httpd::compaction_manager_json::sstableinfo sstable;
-                            sstable.generation = fmt::to_string(data.generation),
-                            sstable.origin = data.origin,
-                            sstable.size = data.size,
-                            h.sstables_out.push(std::move(sstable));
-                        }
-                        h.total_tombstone_purge_attempt = entry.total_tombstone_purge_attempt;
-                        h.total_tombstone_purge_failure_due_to_overlapping_with_memtable = entry.total_tombstone_purge_failure_due_to_overlapping_with_memtable;
-                        h.total_tombstone_purge_failure_due_to_overlapping_with_uncompacting_sstable = entry.total_tombstone_purge_failure_due_to_overlapping_with_uncompacting_sstable;
-
                         if (!first) {
                             co_await s.write(", ");
                         }

api/compaction_manager.hh

@@ -13,13 +13,11 @@ namespace seastar::httpd {
 class routes;
 }
 
-namespace compaction {
 class compaction_manager;
-}
 
 namespace api {
 struct http_context;
-void set_compaction_manager(http_context& ctx, seastar::httpd::routes& r, seastar::sharded<compaction::compaction_manager>& cm);
+void set_compaction_manager(http_context& ctx, seastar::httpd::routes& r, seastar::sharded<compaction_manager>& cm);
 void unset_compaction_manager(http_context& ctx, seastar::httpd::routes& r);
 
 }

api/config.cc

@@ -23,6 +23,22 @@ using namespace seastar::httpd;
 namespace sp = httpd::storage_proxy_json;
 namespace ss = httpd::storage_service_json;
 
+template<class T>
+json::json_return_type get_json_return_type(const T& val) {
+    return json::json_return_type(val);
+}
+
+/*
+ * As commented on db::seed_provider_type is not used
+ * and probably never will.
+ *
+ * Just in case, we will return its name
+ */
+template<>
+json::json_return_type get_json_return_type(const db::seed_provider_type& val) {
+    return json::json_return_type(val.class_name);
+}
+
 std::string_view format_type(std::string_view type) {
     if (type == "int") {
         return "integer";
@@ -171,7 +187,7 @@ void set_config(std::shared_ptr < api_registry_builder20 > rb, http_context& ctx
     });
 
     ss::get_all_data_file_locations.set(r, [&cfg](const_req req) {
-        return cfg.data_file_directories();
+        return container_to_vec(cfg.data_file_directories());
     });
 
     ss::get_saved_caches_location.set(r, [&cfg](const_req req) {

api/error_injection.cc

@@ -21,10 +21,10 @@ namespace hf = httpd::error_injection_json;
 
 void set_error_injection(http_context& ctx, routes& r) {
 
-    hf::enable_injection.set(r, [](std::unique_ptr<request> req) -> future<json::json_return_type> {
+    hf::enable_injection.set(r, [](std::unique_ptr<request> req) {
         sstring injection = req->get_path_param("injection");
         bool one_shot = req->get_query_param("one_shot") == "True";
-        auto params = co_await util::read_entire_stream_contiguous(*req->content_stream);
+        auto params = req->content;
 
         const size_t max_params_size = 1024 * 1024;
         if (params.size() > max_params_size) {
@@ -39,11 +39,12 @@ void set_error_injection(http_context& ctx, routes& r) {
                 : rjson::parse_to_map<utils::error_injection_parameters>(params);
 
             auto& errinj = utils::get_local_injector();
-            co_await errinj.enable_on_all(injection, one_shot, std::move(parameters));
+            return errinj.enable_on_all(injection, one_shot, std::move(parameters)).then([] {
+                return make_ready_future<json::json_return_type>(json::json_void());
+            });
         } catch (const rjson::error& e) {
             throw httpd::bad_param_exception(format("Failed to parse injections parameters: {}", e.what()));
         }
-        co_return json::json_void();
     });
 
     hf::get_enabled_injections_on_all.set(r, [](std::unique_ptr<request> req) {

api/failure_detector.cc

@@ -22,10 +22,10 @@ void set_failure_detector(http_context& ctx, routes& r, gms::gossiper& g) {
         return g.container().invoke_on(0, [] (gms::gossiper& g) {
             std::vector<fd::endpoint_state> res;
             res.reserve(g.num_endpoints());
-            g.for_each_endpoint_state([&] (const gms::endpoint_state& eps) {
+            g.for_each_endpoint_state([&] (const gms::inet_address& addr, const gms::endpoint_state& eps) {
                 fd::endpoint_state val;
-                val.addrs = fmt::to_string(eps.get_ip());
-                val.is_alive = g.is_alive(eps.get_host_id());
+                val.addrs = fmt::to_string(addr);
+                val.is_alive = g.is_alive(addr);
                 val.generation = eps.get_heart_beat_state().get_generation().value();
                 val.version = eps.get_heart_beat_state().get_heart_beat_version().value();
                 val.update_time = eps.get_update_timestamp().time_since_epoch().count();
@@ -40,9 +40,7 @@ void set_failure_detector(http_context& ctx, routes& r, gms::gossiper& g) {
                 }
                 res.emplace_back(std::move(val));
             });
-            return make_ready_future<json::json_return_type>(json::stream_range_as_array(res, [](const fd::endpoint_state& i){
-                return i;
-            }));
+            return make_ready_future<json::json_return_type>(res);
         });
     });
 
@@ -66,15 +64,11 @@ void set_failure_detector(http_context& ctx, routes& r, gms::gossiper& g) {
 
     fd::get_simple_states.set(r, [&g] (std::unique_ptr<request> req) {
         return g.container().invoke_on(0, [] (gms::gossiper& g) {
-            std::vector<fd::mapper> nodes_status;
-            nodes_status.reserve(g.num_endpoints());
-            g.for_each_endpoint_state([&] (const gms::endpoint_state& es) {
-                fd::mapper val;
-                val.key = fmt::to_string(es.get_ip());
-                val.value = g.is_alive(es.get_host_id()) ? "UP" : "DOWN";
-                nodes_status.emplace_back(std::move(val));
+            std::map<sstring, sstring> nodes_status;
+            g.for_each_endpoint_state([&] (const gms::inet_address& node, const gms::endpoint_state&) {
+                nodes_status.emplace(fmt::to_string(node), g.is_alive(node) ? "UP" : "DOWN");
             });
-            return make_ready_future<json::json_return_type>(std::move(nodes_status));
+            return make_ready_future<json::json_return_type>(map_to_key_value<fd::mapper>(nodes_status));
         });
     });
 
@@ -87,7 +81,7 @@ void set_failure_detector(http_context& ctx, routes& r, gms::gossiper& g) {
 
     fd::get_endpoint_state.set(r, [&g] (std::unique_ptr<request> req) {
         return g.container().invoke_on(0, [req = std::move(req)] (gms::gossiper& g) {
-            auto state = g.get_endpoint_state_ptr(g.get_host_id(gms::inet_address(req->get_path_param("addr"))));
+            auto state = g.get_endpoint_state_ptr(gms::inet_address(req->get_path_param("addr")));
             if (!state) {
                 return make_ready_future<json::json_return_type>(format("unknown endpoint {}", req->get_path_param("addr")));
             }

api/gossiper.cc

@@ -21,45 +21,51 @@ using namespace json;
 void set_gossiper(http_context& ctx, routes& r, gms::gossiper& g) {
     httpd::gossiper_json::get_down_endpoint.set(r, [&g] (std::unique_ptr<request> req) -> future<json::json_return_type> {
         auto res = co_await g.get_unreachable_members_synchronized();
-        co_return json::json_return_type(res | std::views::transform([] (auto& ep) { return fmt::to_string(ep); }) | std::ranges::to<std::vector>());
+        co_return json::json_return_type(container_to_vec(res));
     });
 
 
-    httpd::gossiper_json::get_live_endpoint.set(r, [&g] (std::unique_ptr<request> req) -> future<json::json_return_type> {
-        auto res = co_await g.get_live_members_synchronized();
-        co_return json::json_return_type(res | std::views::transform([] (auto& ep) { return fmt::to_string(ep); }) | std::ranges::to<std::vector>());
+    httpd::gossiper_json::get_live_endpoint.set(r, [&g] (std::unique_ptr<request> req) {
+        return g.get_live_members_synchronized().then([] (auto res) {
+            return make_ready_future<json::json_return_type>(container_to_vec(res));
+        });
     });
 
     httpd::gossiper_json::get_endpoint_downtime.set(r, [&g] (std::unique_ptr<request> req) -> future<json::json_return_type> {
         gms::inet_address ep(req->get_path_param("addr"));
         // synchronize unreachable_members on all shards
         co_await g.get_unreachable_members_synchronized();
-        co_return g.get_endpoint_downtime(g.get_host_id(ep));
+        co_return g.get_endpoint_downtime(ep);
     });
 
     httpd::gossiper_json::get_current_generation_number.set(r, [&g] (std::unique_ptr<http::request> req) {
         gms::inet_address ep(req->get_path_param("addr"));
-        return g.get_current_generation_number(g.get_host_id(ep)).then([] (gms::generation_type res) {
+        return g.get_current_generation_number(ep).then([] (gms::generation_type res) {
             return make_ready_future<json::json_return_type>(res.value());
         });
     });
 
     httpd::gossiper_json::get_current_heart_beat_version.set(r, [&g] (std::unique_ptr<http::request> req) {
         gms::inet_address ep(req->get_path_param("addr"));
-        return g.get_current_heart_beat_version(g.get_host_id(ep)).then([] (gms::version_type res) {
+        return g.get_current_heart_beat_version(ep).then([] (gms::version_type res) {
             return make_ready_future<json::json_return_type>(res.value());
         });
     });
 
     httpd::gossiper_json::assassinate_endpoint.set(r, [&g](std::unique_ptr<http::request> req) {
-        return g.assassinate_endpoint(req->get_path_param("addr")).then([] {
+        if (req->get_query_param("unsafe") != "True") {
+            return g.assassinate_endpoint(req->get_path_param("addr")).then([] {
+                return make_ready_future<json::json_return_type>(json_void());
+            });
+        }
+        return g.unsafe_assassinate_endpoint(req->get_path_param("addr")).then([] {
             return make_ready_future<json::json_return_type>(json_void());
         });
     });
 
     httpd::gossiper_json::force_remove_endpoint.set(r, [&g](std::unique_ptr<http::request> req) {
         gms::inet_address ep(req->get_path_param("addr"));
-        return g.force_remove_endpoint(g.get_host_id(ep), gms::null_permit_id).then([] () {
+        return g.force_remove_endpoint(ep, gms::null_permit_id).then([] {
             return make_ready_future<json::json_return_type>(json_void());
         });
     });

api/messaging_service.cc

@@ -148,7 +148,7 @@ void set_messaging_service(http_context& ctx, routes& r, sharded<netw::messaging
     hf::inject_disconnect.set(r, [&ms] (std::unique_ptr<request> req) -> future<json::json_return_type> {
         auto ip = msg_addr(req->get_path_param("ip"));
         co_await ms.invoke_on_all([ip] (netw::messaging_service& ms) {
-            ms.remove_rpc_client(ip, std::nullopt);
+            ms.remove_rpc_client(ip);
         });
         co_return json::json_void();
     });

api/raft.cc

@@ -71,7 +71,7 @@ void set_raft(http_context&, httpd::routes& r, sharded<service::raft_group_regis
         co_return json_void{};
     });
     r::get_leader_host.set(r, [&raft_gr] (std::unique_ptr<http::request> req) -> future<json_return_type> {
-        if (req->get_query_param("group_id").empty()) {
+        if (!req->query_parameters.contains("group_id")) {
             const auto leader_id = co_await raft_gr.invoke_on(0, [] (service::raft_group_registry& raft_gr) {
                 auto& srv = raft_gr.group0();
                 return srv.current_leader();
@@ -100,7 +100,7 @@ void set_raft(http_context&, httpd::routes& r, sharded<service::raft_group_regis
     r::read_barrier.set(r, [&raft_gr] (std::unique_ptr<http::request> req) -> future<json_return_type> {
         auto timeout = get_request_timeout(*req);
 
-        if (req->get_query_param("group_id").empty()) {
+        if (!req->query_parameters.contains("group_id")) {
             // Read barrier on group 0 by default
             co_await raft_gr.invoke_on(0, [timeout] (service::raft_group_registry& raft_gr) -> future<> {
                 co_await raft_gr.group0_with_timeouts().read_barrier(nullptr, timeout);
@@ -131,7 +131,7 @@ void set_raft(http_context&, httpd::routes& r, sharded<service::raft_group_regis
         const auto stepdown_timeout_ticks = dur / service::raft_tick_interval;
         auto timeout_dur = raft::logical_clock::duration(stepdown_timeout_ticks);
 
-        if (req->get_query_param("group_id").empty()) {
+        if (!req->query_parameters.contains("group_id")) {
             // Stepdown on group 0 by default
             co_await raft_gr.invoke_on(0, [timeout_dur] (service::raft_group_registry& raft_gr) {
                 apilog.info("Triggering stepdown for group0");

api/service_levels.cc

@@ -11,7 +11,7 @@
 #include "cql3/query_processor.hh"
 #include "cql3/untyped_result_set.hh"
 #include "db/consistency_level_type.hh"
-#include <seastar/json/json_elements.hh>
+#include "seastar/json/json_elements.hh"
 #include "transport/controller.hh"
 #include <unordered_map>
 

api/storage_proxy.cc

@@ -39,7 +39,7 @@ utils::time_estimated_histogram timed_rate_moving_average_summary_merge(utils::t
  * @return A future that resolves to the result of the aggregation.
  */
 template<typename V, typename Reducer, typename InnerMapper>
-future<V> two_dimensional_map_reduce(sharded<service::storage_proxy>& d,
+future<V> two_dimensional_map_reduce(distributed<service::storage_proxy>& d,
         InnerMapper mapper, Reducer reducer, V initial_value) {
     return d.map_reduce0( [mapper, reducer, initial_value] (const service::storage_proxy& sp) {
         return map_reduce_scheduling_group_specific<service::storage_proxy_stats::stats>(
@@ -59,7 +59,7 @@ future<V> two_dimensional_map_reduce(sharded<service::storage_proxy>& d,
  * @return A future that resolves to the result of the aggregation.
  */
 template<typename V, typename Reducer, typename F, typename C>
-future<V> two_dimensional_map_reduce(sharded<service::storage_proxy>& d,
+future<V> two_dimensional_map_reduce(distributed<service::storage_proxy>& d,
         C F::*f, Reducer reducer, V initial_value) {
     return two_dimensional_map_reduce(d, [f] (F& stats) -> V {
         return stats.*f;
@@ -75,20 +75,20 @@ future<V> two_dimensional_map_reduce(sharded<service::storage_proxy>& d,
  *
  */
 template<typename V, typename F>
-future<json::json_return_type>  sum_stats_storage_proxy(sharded<proxy>& d, V F::*f) {
+future<json::json_return_type>  sum_stats_storage_proxy(distributed<proxy>& d, V F::*f) {
     return two_dimensional_map_reduce(d, [f] (F& stats) { return stats.*f; }, std::plus<V>(), V(0)).then([] (V val) {
         return make_ready_future<json::json_return_type>(val);
     });
 }
 
 
-static future<utils::rate_moving_average>  sum_timed_rate(sharded<proxy>& d, utils::timed_rate_moving_average service::storage_proxy_stats::stats::*f) {
+static future<utils::rate_moving_average>  sum_timed_rate(distributed<proxy>& d, utils::timed_rate_moving_average service::storage_proxy_stats::stats::*f) {
     return two_dimensional_map_reduce(d, [f] (service::storage_proxy_stats::stats& stats) {
         return (stats.*f).rate();
     }, std::plus<utils::rate_moving_average>(), utils::rate_moving_average());
 }
 
-static future<json::json_return_type>  sum_timed_rate_as_obj(sharded<proxy>& d, utils::timed_rate_moving_average service::storage_proxy_stats::stats::*f) {
+static future<json::json_return_type>  sum_timed_rate_as_obj(distributed<proxy>& d, utils::timed_rate_moving_average service::storage_proxy_stats::stats::*f) {
     return sum_timed_rate(d, f).then([](const utils::rate_moving_average& val) {
         httpd::utils_json::rate_moving_average m;
         m = val;
@@ -100,7 +100,7 @@ httpd::utils_json::rate_moving_average_and_histogram get_empty_moving_average()
     return timer_to_json(utils::rate_moving_average_and_histogram());
 }
 
-static future<json::json_return_type>  sum_timed_rate_as_long(sharded<proxy>& d, utils::timed_rate_moving_average service::storage_proxy_stats::stats::*f) {
+static future<json::json_return_type>  sum_timed_rate_as_long(distributed<proxy>& d, utils::timed_rate_moving_average service::storage_proxy_stats::stats::*f) {
     return sum_timed_rate(d, f).then([](const utils::rate_moving_average& val) {
         return make_ready_future<json::json_return_type>(val.count);
     });
@@ -152,7 +152,7 @@ static future<json::json_return_type>  total_latency(sharded<service::storage_pr
  */
 template<typename F>
 future<json::json_return_type>
-sum_histogram_stats_storage_proxy(sharded<proxy>& d,
+sum_histogram_stats_storage_proxy(distributed<proxy>& d,
         utils::timed_rate_moving_average_summary_and_histogram F::*f) {
     return two_dimensional_map_reduce(d, [f] (service::storage_proxy_stats::stats& stats) {
         return (stats.*f).hist;
@@ -172,7 +172,7 @@ sum_histogram_stats_storage_proxy(sharded<proxy>& d,
  */
 template<typename F>
 future<json::json_return_type>
-sum_timer_stats_storage_proxy(sharded<proxy>& d,
+sum_timer_stats_storage_proxy(distributed<proxy>& d,
         utils::timed_rate_moving_average_summary_and_histogram F::*f) {
 
     return two_dimensional_map_reduce(d, [f] (service::storage_proxy_stats::stats& stats) {

api/storage_service.hh

@@ -52,19 +52,17 @@ table_id validate_table(const replica::database& db, sstring ks_name, sstring ta
 // containing the description of the respective no_such_column_family error.
 // Returns a vector of all table infos given by the parameter, or
 // if the parameter is not found or is empty, returns a list of all table infos in the keyspace.
+std::vector<table_info> parse_table_infos(const sstring& ks_name, const http_context& ctx, const std::unordered_map<sstring, sstring>& query_params, sstring param_name);
 
 std::vector<table_info> parse_table_infos(const sstring& ks_name, const http_context& ctx, sstring value);
 
-std::pair<sstring, std::vector<table_info>> parse_table_infos(const http_context& ctx, const http::request& req, sstring cf_param_name = "cf");
-
 struct scrub_info {
-    compaction::compaction_type_options::scrub opts;
+    sstables::compaction_type_options::scrub opts;
     sstring keyspace;
     std::vector<sstring> column_families;
-    sstring snapshot_tag;
 };
 
-scrub_info parse_scrub_options(const http_context& ctx, std::unique_ptr<http::request> req);
+future<scrub_info> parse_scrub_options(const http_context& ctx, sharded<db::snapshot_ctl>& snap_ctl, std::unique_ptr<http::request> req);
 
 void set_storage_service(http_context& ctx, httpd::routes& r, sharded<service::storage_service>& ss, service::raft_group0_client&);
 void unset_storage_service(http_context& ctx, httpd::routes& r);
@@ -82,13 +80,6 @@ void set_snapshot(http_context& ctx, httpd::routes& r, sharded<db::snapshot_ctl>
 void unset_snapshot(http_context& ctx, httpd::routes& r);
 void set_load_meter(http_context& ctx, httpd::routes& r, service::load_meter& lm);
 void unset_load_meter(http_context& ctx, httpd::routes& r);
-seastar::future<json::json_return_type> run_toppartitions_query(db::toppartitions_query& q, bool legacy_request = false);
-
-// converts string value of boolean parameter into bool
-// maps (case insensitively)
-//     "true", "yes" and "1" into true
-//     "false", "no" and "0" into false
-// otherwise throws runtime_error
-bool validate_bool_x(const sstring& param, bool default_value);
+seastar::future<json::json_return_type> run_toppartitions_query(db::toppartitions_query& q, http_context &ctx, bool legacy_request = false);
 
 } // namespace api

api/system.cc

@@ -10,7 +10,7 @@
 #include "api/api-doc/system.json.hh"
 #include "api/api-doc/metrics.json.hh"
 #include "replica/database.hh"
-#include "sstables/sstables_manager.hh"
+#include "db/sstables-format-selector.hh"
 
 #include <rapidjson/document.h>
 #include <boost/lexical_cast.hpp>
@@ -54,8 +54,7 @@ void set_system(http_context& ctx, routes& r) {
 
     hm::set_metrics_config.set(r, [](std::unique_ptr<http::request> req) -> future<json::json_return_type> {
         rapidjson::Document doc;
-        auto content = co_await util::read_entire_stream_contiguous(*req->content_stream);
-        doc.Parse(content.c_str());
+        doc.Parse(req->content.c_str());
         if (!doc.IsArray()) {
             throw bad_param_exception("Expected a json array");
         }
@@ -88,19 +87,21 @@ void set_system(http_context& ctx, routes& r) {
                 relabels[i].expr = element["regex"].GetString();
             }
         }
-        bool failed = false;
-        co_await smp::invoke_on_all([&relabels, &failed] {
-            return metrics::set_relabel_configs(relabels).then([&failed](const metrics::metric_relabeling_result& result) {
-                if (result.metrics_relabeled_due_to_collision > 0) {
-                    failed = true;
+        return do_with(std::move(relabels), false, [](const std::vector<seastar::metrics::relabel_config>& relabels, bool& failed) {
+            return smp::invoke_on_all([&relabels, &failed] {
+                return metrics::set_relabel_configs(relabels).then([&failed](const metrics::metric_relabeling_result& result) {
+                    if (result.metrics_relabeled_due_to_collision > 0) {
+                        failed = true;
+                    }
+                    return;
+                });
+            }).then([&failed](){
+                if (failed) {
+                    throw bad_param_exception("conflicts found during relabeling");
                 }
-                return;
+                return make_ready_future<json::json_return_type>(seastar::json::json_void());
             });
         });
-        if (failed) {
-            throw bad_param_exception("conflicts found during relabeling");
-        }
-        co_return seastar::json::json_void();
     });
 
     hs::get_system_uptime.set(r, [](const_req req) {
@@ -183,13 +184,18 @@ void set_system(http_context& ctx, routes& r) {
         apilog.info("Profile dumped to {}", profile_dest);
         return make_ready_future<json::json_return_type>(json::json_return_type(json::json_void()));
     }) ;
+}
 
-    hs::get_highest_supported_sstable_version.set(r, [&ctx] (std::unique_ptr<request> req) {
-        return smp::submit_to(0, [&ctx] {
-            auto format = ctx.db.local().get_user_sstables_manager().get_highest_supported_format();
-            return make_ready_future<json::json_return_type>(seastar::to_sstring(format));
+void set_format_selector(http_context& ctx, routes& r, db::sstables_format_selector& sel) {
+    hs::get_highest_supported_sstable_version.set(r, [&sel] (std::unique_ptr<request> req) {
+        return smp::submit_to(0, [&sel] {
+            return make_ready_future<json::json_return_type>(seastar::to_sstring(sel.selected_format()));
         });
     });
 }
 
+void unset_format_selector(http_context& ctx, routes& r) {
+    hs::get_highest_supported_sstable_version.unset(r);
+}
+
 }

api/system.hh

@@ -12,9 +12,14 @@ namespace seastar::httpd {
 class routes;
 }
 
+namespace db { class sstables_format_selector; }
+
 namespace api {
 
 struct http_context;
 void set_system(http_context& ctx, seastar::httpd::routes& r);
 
+void set_format_selector(http_context& ctx, seastar::httpd::routes& r, db::sstables_format_selector& sel);
+void unset_format_selector(http_context& ctx, seastar::httpd::routes& r);
+
 }

api/task_manager.cc

@@ -6,7 +6,6 @@
  * SPDX-License-Identifier: LicenseRef-ScyllaDB-Source-Available-1.0
  */
 
-#include <seastar/core/chunked_fifo.hh>
 #include <seastar/core/coroutine.hh>
 #include <seastar/coroutine/exception.hh>
 #include <seastar/http/exception.hh>
@@ -35,9 +34,8 @@ static ::tm get_time(db_clock::time_point tp) {
 }
 
 tm::task_status make_status(tasks::task_status status, sharded<gms::gossiper>& gossiper) {
-    chunked_fifo<tm::task_identity> tis;
-    tis.reserve(status.children.size());
-    for (const auto& child : status.children) {
+    std::vector<tm::task_identity> tis{status.children.size()};
+    std::ranges::transform(status.children, tis.begin(), [&gossiper] (const auto& child) {
         tm::task_identity ident;
         gms::inet_address addr{};
         if (gossiper.local_is_initialized()) {
@@ -45,8 +43,8 @@ tm::task_status make_status(tasks::task_status status, sharded<gms::gossiper>& g
         }
         ident.task_id = child.task_id.to_sstring();
         ident.node = fmt::format("{}", addr);
-        tis.push_back(std::move(ident));
-    }
+        return ident;
+    });
 
     tm::task_status res{};
     res.id = status.task_id.to_sstring();
@@ -107,11 +105,11 @@ void set_task_manager(http_context& ctx, routes& r, sharded<tasks::task_manager>
                 throw bad_param_exception(fmt::format("{}", std::current_exception()));
             }
 
-            if (auto param = req->get_query_param("keyspace"); !param.empty()) {
-                keyspace = param;
+            if (auto it = req->query_parameters.find("keyspace"); it != req->query_parameters.end()) {
+                keyspace = it->second;
             }
-            if (auto param = req->get_query_param("table"); !param.empty()) {
-                table = param;
+            if (auto it = req->query_parameters.find("table"); it != req->query_parameters.end()) {
+                table = it->second;
             }
 
             return module->get_stats(internal, [keyspace = std::move(keyspace), table = std::move(table)] (std::string& ks, std::string& t) {
@@ -119,7 +117,7 @@ void set_task_manager(http_context& ctx, routes& r, sharded<tasks::task_manager>
             });
         });
 
-        noncopyable_function<future<>(output_stream<char>&&)> f = [r = std::move(res)] (output_stream<char>&& os) -> future<> {
+        std::function<future<>(output_stream<char>&&)> f = [r = std::move(res)] (output_stream<char>&& os) -> future<> {
             auto s = std::move(os);
             std::exception_ptr ex;
             try {
@@ -175,8 +173,8 @@ void set_task_manager(http_context& ctx, routes& r, sharded<tasks::task_manager>
         auto id = tasks::task_id{utils::UUID{req->get_path_param("task_id")}};
         tasks::task_status status;
         std::optional<std::chrono::seconds> timeout = std::nullopt;
-        if (auto param = req->get_query_param("timeout"); !param.empty()) {
-            timeout = std::chrono::seconds(boost::lexical_cast<uint32_t>(param));
+        if (auto it = req->query_parameters.find("timeout"); it != req->query_parameters.end()) {
+            timeout = std::chrono::seconds(boost::lexical_cast<uint32_t>(it->second));
         }
         try {
             auto task = tasks::task_handler{tm.local(), id};
@@ -196,7 +194,7 @@ void set_task_manager(http_context& ctx, routes& r, sharded<tasks::task_manager>
             auto task = tasks::task_handler{tm.local(), id};
             auto res = co_await task.get_status_recursively(true);
 
-            noncopyable_function<future<>(output_stream<char>&&)> f = [r = std::move(res), &gossiper] (output_stream<char>&& os) -> future<> {
+            std::function<future<>(output_stream<char>&&)> f = [r = std::move(res), &gossiper] (output_stream<char>&& os) -> future<> {
                 auto s = std::move(os);
                 auto res = std::move(r);
                 co_await s.write("[");
@@ -217,7 +215,7 @@ void set_task_manager(http_context& ctx, routes& r, sharded<tasks::task_manager>
     tm::get_and_update_ttl.set(r, [&cfg] (std::unique_ptr<http::request> req) -> future<json::json_return_type> {
         uint32_t ttl = cfg.task_ttl_seconds();
         try {
-            co_await cfg.task_ttl_seconds.set_value_on_all_shards(req->get_query_param("ttl"), utils::config_file::config_source::API);
+            co_await cfg.task_ttl_seconds.set_value_on_all_shards(req->query_parameters["ttl"], utils::config_file::config_source::API);
         } catch (...) {
             throw bad_param_exception(fmt::format("{}", std::current_exception()));
         }
@@ -232,7 +230,7 @@ void set_task_manager(http_context& ctx, routes& r, sharded<tasks::task_manager>
     tm::get_and_update_user_ttl.set(r, [&cfg] (std::unique_ptr<http::request> req) -> future<json::json_return_type> {
         uint32_t user_ttl = cfg.user_task_ttl_seconds();
         try {
-            co_await cfg.user_task_ttl_seconds.set_value_on_all_shards(req->get_query_param("user_ttl"), utils::config_file::config_source::API);
+            co_await cfg.user_task_ttl_seconds.set_value_on_all_shards(req->query_parameters["user_ttl"], utils::config_file::config_source::API);
         } catch (...) {
             throw bad_param_exception(fmt::format("{}", std::current_exception()));
         }

api/task_manager_test.cc

@@ -57,16 +57,20 @@ void set_task_manager_test(http_context& ctx, routes& r, sharded<tasks::task_man
 
     tmt::register_test_task.set(r, [&tm] (std::unique_ptr<http::request> req) -> future<json::json_return_type> {
         sharded<tasks::task_manager>& tms = tm;
-        const auto id_param = req->get_query_param("task_id");
-        auto id = !id_param.empty() ? tasks::task_id{utils::UUID{id_param}} : tasks::task_id::create_null_id();
-        const auto shard_param = req->get_query_param("shard");
-        unsigned shard = shard_param.empty() ? 0 : boost::lexical_cast<unsigned>(shard_param);
-        std::string keyspace = req->get_query_param("keyspace");
-        std::string table = req->get_query_param("table");
-        std::string entity = req->get_query_param("entity");
+        auto it = req->query_parameters.find("task_id");
+        auto id = it != req->query_parameters.end() ? tasks::task_id{utils::UUID{it->second}} : tasks::task_id::create_null_id();
+        it = req->query_parameters.find("shard");
+        unsigned shard = it != req->query_parameters.end() ? boost::lexical_cast<unsigned>(it->second) : 0;
+        it = req->query_parameters.find("keyspace");
+        std::string keyspace = it != req->query_parameters.end() ? it->second : "";
+        it = req->query_parameters.find("table");
+        std::string table = it != req->query_parameters.end() ? it->second : "";
+        it = req->query_parameters.find("entity");
+        std::string entity = it != req->query_parameters.end() ? it->second : "";
+        it = req->query_parameters.find("parent_id");
         tasks::task_info data;
-        if (auto parent_id = req->get_query_param("parent_id"); !parent_id.empty()) {
-            data.id = tasks::task_id{utils::UUID{parent_id}};
+        if (it != req->query_parameters.end()) {
+            data.id = tasks::task_id{utils::UUID{it->second}};
             auto parent_ptr = co_await tasks::task_manager::lookup_task_on_all_shards(tm, data.id);
             data.shard = parent_ptr->get_status().shard;
         }
@@ -84,7 +88,7 @@ void set_task_manager_test(http_context& ctx, routes& r, sharded<tasks::task_man
     });
 
     tmt::unregister_test_task.set(r, [&tm] (std::unique_ptr<http::request> req) -> future<json::json_return_type> {
-        auto id = tasks::task_id{utils::UUID{req->get_query_param("task_id")}};
+        auto id = tasks::task_id{utils::UUID{req->query_parameters["task_id"]}};
         try {
             co_await tasks::task_manager::invoke_on_task(tm, id, [] (tasks::task_manager::task_variant task_v, tasks::virtual_task_hint) -> future<> {
                 return std::visit(overloaded_functor{
@@ -105,8 +109,9 @@ void set_task_manager_test(http_context& ctx, routes& r, sharded<tasks::task_man
 
     tmt::finish_test_task.set(r, [&tm] (std::unique_ptr<http::request> req) -> future<json::json_return_type> {
         auto id = tasks::task_id{utils::UUID{req->get_path_param("task_id")}};
-        std::string error = req->get_query_param("error");
-        bool fail = !error.empty();
+        auto it = req->query_parameters.find("error");
+        bool fail = it != req->query_parameters.end();
+        std::string error = fail ? it->second : "";
 
         try {
             co_await tasks::task_manager::invoke_on_task(tm, id, [fail, error = std::move(error)] (tasks::task_manager::task_variant task_v, tasks::virtual_task_hint) -> future<> {

api/tasks.cc

@@ -12,7 +12,6 @@
 #include "api/api.hh"
 #include "api/storage_service.hh"
 #include "api/api-doc/tasks.json.hh"
-#include "api/api-doc/storage_service.json.hh"
 #include "compaction/compaction_manager.hh"
 #include "compaction/task_manager_module.hh"
 #include "service/storage_service.hh"
@@ -26,14 +25,14 @@ extern logging::logger apilog;
 namespace api {
 
 namespace t = httpd::tasks_json;
-namespace ss = httpd::storage_service_json;
 using namespace json;
 
 using ks_cf_func = std::function<future<json::json_return_type>(http_context&, std::unique_ptr<http::request>, sstring, std::vector<table_info>)>;
 
 static auto wrap_ks_cf(http_context &ctx, ks_cf_func f) {
     return [&ctx, f = std::move(f)](std::unique_ptr<http::request> req) {
-        auto [keyspace, table_infos] = parse_table_infos(ctx, *req);
+        auto keyspace = validate_keyspace(ctx, req);
+        auto table_infos = parse_table_infos(keyspace, ctx, req->query_parameters, "cf");
         return f(ctx, std::move(req), std::move(keyspace), std::move(table_infos));
     };
 }
@@ -41,93 +40,52 @@ static auto wrap_ks_cf(http_context &ctx, ks_cf_func f) {
 void set_tasks_compaction_module(http_context& ctx, routes& r, sharded<service::storage_service>& ss, sharded<db::snapshot_ctl>& snap_ctl) {
     t::force_keyspace_compaction_async.set(r, [&ctx](std::unique_ptr<http::request> req) -> future<json::json_return_type> {
         auto& db = ctx.db;
-        auto [ keyspace, table_infos ] = parse_table_infos(ctx, *req, "cf");
-        auto flush = validate_bool_x(req->get_query_param("flush_memtables"), true);
+        auto params = req_params({
+            std::pair("keyspace", mandatory::yes),
+            std::pair("cf", mandatory::no),
+            std::pair("flush_memtables", mandatory::no),
+        });
+        params.process(*req);
+        auto keyspace = validate_keyspace(ctx, *params.get("keyspace"));
+        auto table_infos = parse_table_infos(keyspace, ctx, params.get("cf").value_or(""));
+        auto flush = params.get_as<bool>("flush_memtables").value_or(true);
         apilog.debug("force_keyspace_compaction_async: keyspace={} tables={}, flush={}", keyspace, table_infos, flush);
 
         auto& compaction_module = db.local().get_compaction_manager().get_task_manager_module();
-        std::optional<compaction::flush_mode> fmopt;
+        std::optional<flush_mode> fmopt;
         if (!flush) {
-            fmopt = compaction::flush_mode::skip;
+            fmopt = flush_mode::skip;
         }
-        auto task = co_await compaction_module.make_and_start_task<compaction::major_keyspace_compaction_task_impl>({}, std::move(keyspace), tasks::task_id::create_null_id(), db, table_infos, fmopt);
+        auto task = co_await compaction_module.make_and_start_task<major_keyspace_compaction_task_impl>({}, std::move(keyspace), tasks::task_id::create_null_id(), db, table_infos, fmopt);
 
         co_return json::json_return_type(task->get_status().id.to_sstring());
     });
 
-    ss::force_keyspace_compaction.set(r, [&ctx](std::unique_ptr<http::request> req) -> future<json::json_return_type> {
-        auto& db = ctx.db;
-        auto [ keyspace, table_infos ] = parse_table_infos(ctx, *req, "cf");
-        auto flush = validate_bool_x(req->get_query_param("flush_memtables"), true);
-        auto consider_only_existing_data = validate_bool_x(req->get_query_param("consider_only_existing_data"), false);
-        apilog.info("force_keyspace_compaction: keyspace={} tables={}, flush={} consider_only_existing_data={}", keyspace, table_infos, flush, consider_only_existing_data);
-
-        auto& compaction_module = db.local().get_compaction_manager().get_task_manager_module();
-        std::optional<compaction::flush_mode> fmopt;
-        if (!flush && !consider_only_existing_data) {
-            fmopt = compaction::flush_mode::skip;
-        }
-        auto task = co_await compaction_module.make_and_start_task<compaction::major_keyspace_compaction_task_impl>({}, std::move(keyspace), tasks::task_id::create_null_id(), db, table_infos, fmopt, consider_only_existing_data);
-        co_await task->done();
-        co_return json_void();
-    });
-
     t::force_keyspace_cleanup_async.set(r, [&ctx, &ss](std::unique_ptr<http::request> req) -> future<json::json_return_type> {
         auto& db = ctx.db;
-        auto [keyspace, table_infos] = parse_table_infos(ctx, *req);
+        auto keyspace = validate_keyspace(ctx, req);
+        auto table_infos = parse_table_infos(keyspace, ctx, req->query_parameters, "cf");
         apilog.info("force_keyspace_cleanup_async: keyspace={} tables={}", keyspace, table_infos);
-        if (!co_await ss.local().is_vnodes_cleanup_allowed(keyspace)) {
+        if (!co_await ss.local().is_cleanup_allowed(keyspace)) {
             auto msg = "Can not perform cleanup operation when topology changes";
             apilog.warn("force_keyspace_cleanup_async: keyspace={} tables={}: {}", keyspace, table_infos, msg);
             co_await coroutine::return_exception(std::runtime_error(msg));
         }
 
         auto& compaction_module = db.local().get_compaction_manager().get_task_manager_module();
-        auto task = co_await compaction_module.make_and_start_task<compaction::cleanup_keyspace_compaction_task_impl>({}, std::move(keyspace), db, table_infos, compaction::flush_mode::all_tables, tasks::is_user_task::yes);
+        auto task = co_await compaction_module.make_and_start_task<cleanup_keyspace_compaction_task_impl>({}, std::move(keyspace), db, table_infos, flush_mode::all_tables, tasks::is_user_task::yes);
 
         co_return json::json_return_type(task->get_status().id.to_sstring());
     });
 
-    ss::force_keyspace_cleanup.set(r, [&ctx, &ss](std::unique_ptr<http::request> req) -> future<json::json_return_type> {
-        auto& db = ctx.db;
-        auto [keyspace, table_infos] = parse_table_infos(ctx, *req);
-        const auto& rs = db.local().find_keyspace(keyspace).get_replication_strategy();
-        if (rs.is_local() || !rs.is_vnode_based()) {
-            auto reason = rs.is_local() ? "require" : "support";
-            apilog.info("Keyspace {} does not {} cleanup", keyspace, reason);
-            co_return json::json_return_type(0);
-        }
-        apilog.info("force_keyspace_cleanup: keyspace={} tables={}", keyspace, table_infos);
-        if (!co_await ss.local().is_vnodes_cleanup_allowed(keyspace)) {
-            auto msg = "Can not perform cleanup operation when topology changes";
-            apilog.warn("force_keyspace_cleanup: keyspace={} tables={}: {}", keyspace, table_infos, msg);
-            co_await coroutine::return_exception(std::runtime_error(msg));
-        }
-
-        auto& compaction_module = db.local().get_compaction_manager().get_task_manager_module();
-        auto task = co_await compaction_module.make_and_start_task<compaction::cleanup_keyspace_compaction_task_impl>(
-            {}, std::move(keyspace), db, table_infos, compaction::flush_mode::all_tables, tasks::is_user_task::yes);
-        co_await task->done();
-        co_return json::json_return_type(0);
-    });
-
     t::perform_keyspace_offstrategy_compaction_async.set(r, wrap_ks_cf(ctx, [] (http_context& ctx, std::unique_ptr<http::request> req, sstring keyspace, std::vector<table_info> table_infos) -> future<json::json_return_type> {
         apilog.info("perform_keyspace_offstrategy_compaction: keyspace={} tables={}", keyspace, table_infos);
         auto& compaction_module = ctx.db.local().get_compaction_manager().get_task_manager_module();
-        auto task = co_await compaction_module.make_and_start_task<compaction::offstrategy_keyspace_compaction_task_impl>({}, std::move(keyspace), ctx.db, table_infos, nullptr);
+        auto task = co_await compaction_module.make_and_start_task<offstrategy_keyspace_compaction_task_impl>({}, std::move(keyspace), ctx.db, table_infos, nullptr);
 
         co_return json::json_return_type(task->get_status().id.to_sstring());
     }));
 
-    ss::perform_keyspace_offstrategy_compaction.set(r, wrap_ks_cf(ctx, [] (http_context& ctx, std::unique_ptr<http::request> req, sstring keyspace, std::vector<table_info> table_infos) -> future<json::json_return_type> {
-        apilog.info("perform_keyspace_offstrategy_compaction: keyspace={} tables={}", keyspace, table_infos);
-        bool res = false;
-        auto& compaction_module = ctx.db.local().get_compaction_manager().get_task_manager_module();
-        auto task = co_await compaction_module.make_and_start_task<compaction::offstrategy_keyspace_compaction_task_impl>({}, std::move(keyspace), ctx.db, table_infos, &res);
-        co_await task->done();
-        co_return json::json_return_type(res);
-    }));
-
     t::upgrade_sstables_async.set(r, wrap_ks_cf(ctx, [] (http_context& ctx, std::unique_ptr<http::request> req, sstring keyspace, std::vector<table_info> table_infos) -> future<json::json_return_type> {
         auto& db = ctx.db;
         bool exclude_current_version = req_param<bool>(*req, "exclude_current_version", false);
@@ -135,65 +93,28 @@ void set_tasks_compaction_module(http_context& ctx, routes& r, sharded<service::
         apilog.info("upgrade_sstables: keyspace={} tables={} exclude_current_version={}", keyspace, table_infos, exclude_current_version);
 
         auto& compaction_module = db.local().get_compaction_manager().get_task_manager_module();
-        auto task = co_await compaction_module.make_and_start_task<compaction::upgrade_sstables_compaction_task_impl>({}, std::move(keyspace), db, table_infos, exclude_current_version);
+        auto task = co_await compaction_module.make_and_start_task<upgrade_sstables_compaction_task_impl>({}, std::move(keyspace), db, table_infos, exclude_current_version);
 
         co_return json::json_return_type(task->get_status().id.to_sstring());
     }));
 
-    ss::upgrade_sstables.set(r, wrap_ks_cf(ctx, [] (http_context& ctx, std::unique_ptr<http::request> req, sstring keyspace, std::vector<table_info> table_infos) -> future<json::json_return_type> {
-        auto& db = ctx.db;
-        bool exclude_current_version = req_param<bool>(*req, "exclude_current_version", false);
-
-        apilog.info("upgrade_sstables: keyspace={} tables={} exclude_current_version={}", keyspace, table_infos, exclude_current_version);
-
-        auto& compaction_module = db.local().get_compaction_manager().get_task_manager_module();
-        auto task = co_await compaction_module.make_and_start_task<compaction::upgrade_sstables_compaction_task_impl>({}, std::move(keyspace), db, table_infos, exclude_current_version);
-        co_await task->done();
-        co_return json::json_return_type(0);
-    }));
-
     t::scrub_async.set(r, [&ctx, &snap_ctl] (std::unique_ptr<http::request> req) -> future<json::json_return_type> {
         auto& db = ctx.db;
-        auto info = parse_scrub_options(ctx, std::move(req));
-
-        if (!info.snapshot_tag.empty()) {
-            co_await snap_ctl.local().take_column_family_snapshot(info.keyspace, info.column_families, info.snapshot_tag, db::snapshot_ctl::skip_flush::no);
-        }
+        auto info = co_await parse_scrub_options(ctx, snap_ctl, std::move(req));
 
         auto& compaction_module = db.local().get_compaction_manager().get_task_manager_module();
-        auto task = co_await compaction_module.make_and_start_task<compaction::scrub_sstables_compaction_task_impl>({}, std::move(info.keyspace), db, std::move(info.column_families), info.opts, nullptr);
+        auto task = co_await compaction_module.make_and_start_task<scrub_sstables_compaction_task_impl>({}, std::move(info.keyspace), db, std::move(info.column_families), info.opts, nullptr);
 
         co_return json::json_return_type(task->get_status().id.to_sstring());
     });
-
-    ss::force_compaction.set(r, [&ctx] (std::unique_ptr<http::request> req) -> future<json::json_return_type> {
-        auto& db = ctx.db;
-        auto flush = validate_bool_x(req->get_query_param("flush_memtables"), true);
-        auto consider_only_existing_data = validate_bool_x(req->get_query_param("consider_only_existing_data"), false);
-        apilog.info("force_compaction: flush={} consider_only_existing_data={}", flush, consider_only_existing_data);
-
-        auto& compaction_module = db.local().get_compaction_manager().get_task_manager_module();
-        std::optional<compaction::flush_mode> fmopt;
-        if (!flush && !consider_only_existing_data) {
-            fmopt = compaction::flush_mode::skip;
-        }
-        auto task = co_await compaction_module.make_and_start_task<compaction::global_major_compaction_task_impl>({}, db, fmopt, consider_only_existing_data);
-        co_await task->done();
-        co_return json_void();
-    });
 }
 
 void unset_tasks_compaction_module(http_context& ctx, httpd::routes& r) {
     t::force_keyspace_compaction_async.unset(r);
-    ss::force_keyspace_compaction.unset(r);
     t::force_keyspace_cleanup_async.unset(r);
-    ss::force_keyspace_cleanup.unset(r);
     t::perform_keyspace_offstrategy_compaction_async.unset(r);
-    ss::perform_keyspace_offstrategy_compaction.unset(r);
     t::upgrade_sstables_async.unset(r);
-    ss::upgrade_sstables.unset(r);
     t::scrub_async.unset(r);
-    ss::force_compaction.unset(r);
 }
 
 }

api/token_metadata.cc

@@ -54,12 +54,12 @@ void set_token_metadata(http_context& ctx, routes& r, sharded<locator::shared_to
         for (const auto host_id: leaving_host_ids) {
             eps.insert(g.local().get_address_map().get(host_id));
         }
-        return eps | std::views::transform([] (auto& i) { return fmt::to_string(i); }) | std::ranges::to<std::vector>();
+        return container_to_vec(eps);
     });
 
     ss::get_moving_nodes.set(r, [](const_req req) {
         std::unordered_set<sstring> addr;
-        return addr | std::ranges::to<std::vector>();
+        return container_to_vec(addr);
     });
 
     ss::get_joining_nodes.set(r, [&tm, &g](const_req req) {
@@ -70,21 +70,15 @@ void set_token_metadata(http_context& ctx, routes& r, sharded<locator::shared_to
         for (const auto& [token, host_id]: points) {
             eps.insert(g.local().get_address_map().get(host_id));
         }
-        return eps | std::views::transform([] (auto& i) { return fmt::to_string(i); }) | std::ranges::to<std::vector>();
+        return container_to_vec(eps);
     });
 
     ss::get_host_id_map.set(r, [&tm, &g](const_req req) {
-        if (!g.local().is_enabled()) {
-            throw std::runtime_error("The gossiper is not ready yet");
-        }
-        return tm.local().get()->get_host_ids()
-            | std::views::transform([&g] (locator::host_id id) {
-                ss::mapper m;
-                m.key = fmt::to_string(g.local().get_address_map().get(id));
-                m.value = fmt::to_string(id);
-                return m;
-            })
-            | std::ranges::to<std::vector<ss::mapper>>();
+        std::vector<ss::mapper> res;
+        auto map = tm.local().get()->get_host_ids() |
+            std::views::transform([&g] (locator::host_id id) { return std::make_pair(g.local().get_address_map().get(id), id); }) |
+            std::ranges::to<std::unordered_map>();
+        return map_to_key_value(std::move(map), res);
     });
 
     static auto host_or_broadcast = [&tm](const_req req) {

audit/audit.cc

@@ -209,11 +209,6 @@ future<> audit::log(const audit_info* audit_info, service::query_state& query_st
     static const sstring anonymous_username("anonymous");
     const sstring& username = client_state.user() ? client_state.user()->name.value_or(anonymous_username) : no_username;
     socket_address client_ip = client_state.get_client_address().addr();
-    if (logger.is_enabled(logging::log_level::debug)) {
-        logger.debug("Log written: node_ip {} category {} cl {} error {} keyspace {} query '{}' client_ip {} table {} username {}",
-            node_ip, audit_info->category_string(), cl, error, audit_info->keyspace(),
-            audit_info->query(), client_ip, audit_info->table(), username);
-    }
     return futurize_invoke(std::mem_fn(&storage_helper::write), _storage_helper_ptr, audit_info, node_ip, client_ip, cl, username, error)
         .handle_exception([audit_info, node_ip, client_ip, cl, username, error] (auto ep) {
             logger.error("Unexpected exception when writing log with: node_ip {} category {} cl {} error {} keyspace {} query '{}' client_ip {} table {} username {} exception {}",
@@ -224,10 +219,6 @@ future<> audit::log(const audit_info* audit_info, service::query_state& query_st
 
 future<> audit::log_login(const sstring& username, socket_address client_ip, bool error) noexcept {
     socket_address node_ip = _token_metadata.get()->get_topology().my_address().addr();
-    if (logger.is_enabled(logging::log_level::debug)) {
-        logger.debug("Login log written: node_ip {}, client_ip {}, username {}, error {}",
-            node_ip, client_ip, username, error ? "true" : "false");
-    }
     return futurize_invoke(std::mem_fn(&storage_helper::write_login), _storage_helper_ptr, username, node_ip, client_ip, error)
         .handle_exception([username, node_ip, client_ip, error] (auto ep) {
             logger.error("Unexpected exception when writing login log with: node_ip {} client_ip {} username {} error {} exception {}",

audit/audit_cf_storage_helper.cc

@@ -16,7 +16,6 @@
 #include "cql3/statements/ks_prop_defs.hh"
 #include "service/migration_manager.hh"
 #include "service/storage_proxy.hh"
-#include "locator/abstract_replication_strategy.hh"
 
 namespace audit {
 
@@ -65,8 +64,8 @@ future<> audit_cf_storage_helper::migrate_audit_table(service::group0_guard grou
             data_dictionary::database db = _qp.db();
             cql3::statements::ks_prop_defs old_ks_prop_defs;
             auto old_ks_metadata = old_ks_prop_defs.as_ks_metadata_update(
-                    ks->metadata(), *_qp.proxy().get_token_metadata_ptr(), db.features(), db.get_config());
-            locator::replication_strategy_config_options strategy_opts;
+                    ks->metadata(), *_qp.proxy().get_token_metadata_ptr(), db.features());
+            std::map<sstring, sstring> strategy_opts;
             for (const auto &dc: _qp.proxy().get_token_metadata_ptr()->get_topology().get_datacenters())
                 strategy_opts[dc] = "3";
 
@@ -74,7 +73,6 @@ future<> audit_cf_storage_helper::migrate_audit_table(service::group0_guard grou
                                                                    "org.apache.cassandra.locator.NetworkTopologyStrategy",
                                                                    strategy_opts,
                                                                    std::nullopt, // initial_tablets
-                                                                   std::nullopt, // consistency_option
                                                                    old_ks_metadata->durable_writes(),
                                                                    old_ks_metadata->get_storage_options(),
                                                                    old_ks_metadata->tables());

audit/audit_syslog_storage_helper.cc

@@ -33,6 +33,20 @@ namespace audit {
 
 namespace {
 
+future<> syslog_send_helper(net::datagram_channel& sender,
+                            const socket_address& address,
+                            const sstring& msg) {
+    return sender.send(address, net::packet{msg.data(), msg.size()}).handle_exception([address](auto&& exception_ptr) {
+        auto error_msg = seastar::format(
+            "Syslog audit backend failed (sending a message to {} resulted in {}).",
+            address,
+            exception_ptr
+        );
+        logger.error("{}", error_msg);
+        throw audit_exception(std::move(error_msg));
+    });
+}
+
 static auto syslog_address_helper(const db::config& cfg)
 {
     return cfg.audit_unix_socket_path.is_set()
@@ -40,40 +54,11 @@ static auto syslog_address_helper(const db::config& cfg)
         : unix_domain_addr(_PATH_LOG);
 }
 
-static std::string json_escape(std::string_view str) {
-    std::string result;
-    result.reserve(str.size() * 1.2);
-    for (auto c : str) {
-        if (c == '"' || c == '\\') {
-            result.push_back('\\');
-        }
-        result.push_back(c);
-    }
-    return result;
-}
-
-}
-
-future<> audit_syslog_storage_helper::syslog_send_helper(const sstring& msg) {
-    try {
-        auto lock = co_await get_units(_semaphore, 1, std::chrono::hours(1));
-        co_await _sender.send(_syslog_address, net::packet{msg.data(), msg.size()});
-    }
-    catch (const std::exception& e) {
-        auto error_msg = seastar::format(
-            "Syslog audit backend failed (sending a message to {} resulted in {}).",
-            _syslog_address,
-            e
-        );
-        logger.error("{}", error_msg);
-        throw audit_exception(std::move(error_msg));
-    }
 }
 
 audit_syslog_storage_helper::audit_syslog_storage_helper(cql3::query_processor& qp, service::migration_manager&) :
     _syslog_address(syslog_address_helper(qp.db().get_config())),
-    _sender(make_unbound_datagram_channel(AF_UNIX)),
-    _semaphore(1) {
+    _sender(make_unbound_datagram_channel(AF_UNIX)) {
 }
 
 audit_syslog_storage_helper::~audit_syslog_storage_helper() {
@@ -88,10 +73,10 @@ audit_syslog_storage_helper::~audit_syslog_storage_helper() {
  */
 future<> audit_syslog_storage_helper::start(const db::config& cfg) {
     if (this_shard_id() != 0) {
-        co_return;
+        return make_ready_future();
     }
 
-    co_await syslog_send_helper("Initializing syslog audit backend.");
+    return syslog_send_helper(_sender, _syslog_address, "Initializing syslog audit backend.");
 }
 
 future<> audit_syslog_storage_helper::stop() {
@@ -108,7 +93,7 @@ future<> audit_syslog_storage_helper::write(const audit_info* audit_info,
     auto now = std::chrono::system_clock::to_time_t(std::chrono::system_clock::now());
     tm time;
     localtime_r(&now, &time);
-    sstring msg = seastar::format(R"(<{}>{:%h %e %T} scylla-audit: node="{}", category="{}", cl="{}", error="{}", keyspace="{}", query="{}", client_ip="{}", table="{}", username="{}")",
+    sstring msg = seastar::format("<{}>{:%h %e %T} scylla-audit: \"{}\", \"{}\", \"{}\", \"{}\", \"{}\", \"{}\", \"{}\", \"{}\", \"{}\"",
                                     LOG_NOTICE | LOG_USER,
                                     time,
                                     node_ip,
@@ -116,12 +101,12 @@ future<> audit_syslog_storage_helper::write(const audit_info* audit_info,
                                     cl,
                                     (error ? "true" : "false"),
                                     audit_info->keyspace(),
-                                    json_escape(audit_info->query()),
+                                    audit_info->query(),
                                     client_ip,
                                     audit_info->table(),
                                     username);
 
-    co_await syslog_send_helper(msg);
+    return syslog_send_helper(_sender, _syslog_address, msg);
 }
 
 future<> audit_syslog_storage_helper::write_login(const sstring& username,
@@ -132,15 +117,15 @@ future<> audit_syslog_storage_helper::write_login(const sstring& username,
     auto now = std::chrono::system_clock::to_time_t(std::chrono::system_clock::now());
     tm time;
     localtime_r(&now, &time);
-    sstring msg = seastar::format(R"(<{}>{:%h %e %T} scylla-audit: node="{}", category="AUTH", cl="", error="{}", keyspace="", query="", client_ip="{}", table="", username="{}")",
+    sstring msg = seastar::format("<{}>{:%h %e %T} scylla-audit: \"{}\", \"AUTH\", \"\", \"\", \"\", \"\", \"{}\", \"{}\", \"{}\"",
                                     LOG_NOTICE | LOG_USER,
                                     time,
                                     node_ip,
-                                    (error ? "true" : "false"),
                                     client_ip,
-                                    username);
+                                    username,
+                                    (error ? "true" : "false"));
 
-    co_await syslog_send_helper(msg.c_str());
+    co_await syslog_send_helper(_sender, _syslog_address, msg.c_str());
 }
 
 using registry = class_registrator<storage_helper, audit_syslog_storage_helper, cql3::query_processor&, service::migration_manager&>;

audit/audit_syslog_storage_helper.hh

@@ -24,9 +24,6 @@ namespace audit {
 class audit_syslog_storage_helper : public storage_helper {
     socket_address _syslog_address;
     net::datagram_channel _sender;
-    seastar::semaphore _semaphore;
-
-    future<> syslog_send_helper(const sstring& msg);
 public:
     explicit audit_syslog_storage_helper(cql3::query_processor&, service::migration_manager&);
     virtual ~audit_syslog_storage_helper();

auth/allow_all_authenticator.cc

@@ -9,7 +9,6 @@
 #include "auth/allow_all_authenticator.hh"
 
 #include "service/migration_manager.hh"
-#include "utils/alien_worker.hh"
 #include "utils/class_registrator.hh"
 
 namespace auth {
@@ -22,7 +21,6 @@ static const class_registrator<
         allow_all_authenticator,
         cql3::query_processor&,
         ::service::raft_group0_client&,
-        ::service::migration_manager&,
-        utils::alien_worker&> registration("org.apache.cassandra.auth.AllowAllAuthenticator");
+        ::service::migration_manager&> registration("org.apache.cassandra.auth.AllowAllAuthenticator");
 
 }

auth/allow_all_authenticator.hh

@@ -13,7 +13,6 @@
 #include "auth/authenticated_user.hh"
 #include "auth/authenticator.hh"
 #include "auth/common.hh"
-#include "utils/alien_worker.hh"
 
 namespace cql3 {
 class query_processor;
@@ -29,7 +28,7 @@ extern const std::string_view allow_all_authenticator_name;
 
 class allow_all_authenticator final : public authenticator {
 public:
-    allow_all_authenticator(cql3::query_processor&, ::service::raft_group0_client&, ::service::migration_manager&, utils::alien_worker&) {
+    allow_all_authenticator(cql3::query_processor&, ::service::raft_group0_client&, ::service::migration_manager&) {
     }
 
     virtual future<> start() override {

auth/certificate_authenticator.cc

@@ -33,14 +33,13 @@ static const class_registrator<auth::authenticator
     , auth::certificate_authenticator
     , cql3::query_processor&
     , ::service::raft_group0_client&
-    , ::service::migration_manager&
-    , utils::alien_worker&> cert_auth_reg(CERT_AUTH_NAME);
+    , ::service::migration_manager&> cert_auth_reg(CERT_AUTH_NAME);
 
 enum class auth::certificate_authenticator::query_source {
     subject, altname
 };
 
-auth::certificate_authenticator::certificate_authenticator(cql3::query_processor& qp, ::service::raft_group0_client&, ::service::migration_manager&, utils::alien_worker&)
+auth::certificate_authenticator::certificate_authenticator(cql3::query_processor& qp, ::service::raft_group0_client&, ::service::migration_manager&)
     : _queries([&] {
         auto& conf = qp.db().get_config();
         auto queries = conf.auth_certificate_role_queries();

auth/certificate_authenticator.hh

@@ -10,7 +10,6 @@
 #pragma once
 
 #include "auth/authenticator.hh"
-#include "utils/alien_worker.hh"
 #include <boost/regex_fwd.hpp>  // IWYU pragma: keep
 
 namespace cql3 {
@@ -32,7 +31,7 @@ class certificate_authenticator : public authenticator {
     enum class query_source;
     std::vector<std::pair<query_source, boost::regex>> _queries;
 public:
-    certificate_authenticator(cql3::query_processor&, ::service::raft_group0_client&, ::service::migration_manager&, utils::alien_worker&);
+    certificate_authenticator(cql3::query_processor&, ::service::raft_group0_client&, ::service::migration_manager&);
     ~certificate_authenticator();
 
     future<> start() override;

auth/common.cc

@@ -15,7 +15,7 @@
 
 #include "mutation/canonical_mutation.hh"
 #include "schema/schema_fwd.hh"
-#include "mutation/timestamp.hh"
+#include "timestamp.hh"
 #include "utils/assert.hh"
 #include "utils/exponential_backoff_retry.hh"
 #include "cql3/query_processor.hh"
@@ -119,14 +119,9 @@ future<> create_legacy_metadata_table_if_missing(
     return qs;
 }
 
-::service::raft_timeout get_raft_timeout() noexcept {
-    auto dur = internal_distributed_query_state().get_client_state().get_timeout_config().other_timeout;
-    return ::service::raft_timeout{.value = lowres_clock::now() + dur};
-}
-
 static future<> announce_mutations_with_guard(
         ::service::raft_group0_client& group0_client,
-        utils::chunked_vector<canonical_mutation> muts,
+        std::vector<canonical_mutation> muts,
         ::service::group0_guard group0_guard,
         seastar::abort_source& as,
         std::optional<::service::raft_timeout> timeout) {
@@ -154,7 +149,7 @@ future<> announce_mutations_with_batching(
     });
 
     size_t memory_usage = 0;
-    utils::chunked_vector<canonical_mutation> muts;
+    std::vector<canonical_mutation> muts;
 
     // guard has to be taken before we execute code in gen as
     // it can do read-before-write and we want announce_mutations
@@ -204,7 +199,7 @@ future<> announce_mutations(
             internal_distributed_query_state(),
             timestamp,
             std::move(values));
-    utils::chunked_vector<canonical_mutation> cmuts = {muts.begin(), muts.end()};
+    std::vector<canonical_mutation> cmuts = {muts.begin(), muts.end()};
     co_await announce_mutations_with_guard(group0_client, std::move(cmuts), std::move(group0_guard), as, timeout);
 }
 

auth/common.hh

@@ -17,7 +17,6 @@
 
 #include "types/types.hh"
 #include "service/raft/raft_group0_client.hh"
-#include "timeout_config.hh"
 
 using namespace std::chrono_literals;
 
@@ -78,8 +77,6 @@ future<> create_legacy_metadata_table_if_missing(
 ///
 ::service::query_state& internal_distributed_query_state() noexcept;
 
-::service::raft_timeout get_raft_timeout() noexcept;
-
 // Execute update query via group0 mechanism, mutations will be applied on all nodes.
 // Use this function when need to perform read before write on a single guard or if
 // you have more than one mutation and potentially exceed single command size limit.

auth/ldap_role_manager.cc

@@ -233,9 +233,9 @@ future<role_set> ldap_role_manager::query_granted(std::string_view grantee_name,
 }
 
 future<role_to_directly_granted_map>
-ldap_role_manager::query_all_directly_granted(::service::query_state& qs) {
+ldap_role_manager::query_all_directly_granted() {
     role_to_directly_granted_map result;
-    auto roles = co_await query_all(qs);
+    auto roles = co_await query_all();
     for (auto& role: roles) {
         auto granted_set = co_await query_granted(role, recursive_role_query::no);
         for (auto& granted: granted_set) {
@@ -247,8 +247,8 @@ ldap_role_manager::query_all_directly_granted(::service::query_state& qs) {
     co_return result;
 }
 
-future<role_set> ldap_role_manager::query_all(::service::query_state& qs) {
-    return _std_mgr.query_all(qs);
+future<role_set> ldap_role_manager::query_all() {
+    return _std_mgr.query_all();
 }
 
 future<> ldap_role_manager::create_role(std::string_view role_name) {
@@ -311,12 +311,12 @@ future<bool> ldap_role_manager::can_login(std::string_view role_name) {
 }
 
 future<std::optional<sstring>> ldap_role_manager::get_attribute(
-        std::string_view role_name, std::string_view attribute_name, ::service::query_state& qs) {
-    return _std_mgr.get_attribute(role_name, attribute_name, qs);
+        std::string_view role_name, std::string_view attribute_name) {
+    return _std_mgr.get_attribute(role_name, attribute_name);
 }
 
-future<role_manager::attribute_vals> ldap_role_manager::query_attribute_for_all(std::string_view attribute_name, ::service::query_state& qs) {
-    return _std_mgr.query_attribute_for_all(attribute_name, qs);
+future<role_manager::attribute_vals> ldap_role_manager::query_attribute_for_all(std::string_view attribute_name) {
+    return _std_mgr.query_attribute_for_all(attribute_name);
 }
 
 future<> ldap_role_manager::set_attribute(
@@ -338,7 +338,8 @@ future<std::vector<cql3::description>> ldap_role_manager::describe_role_grants()
 }
 
 future<> ldap_role_manager::ensure_superuser_is_created() {
-    return _std_mgr.ensure_superuser_is_created();
+    // ldap is responsible for users
+    co_return;
 }
 
 } // namespace auth

auth/ldap_role_manager.hh

@@ -75,9 +75,9 @@ class ldap_role_manager : public role_manager {
 
     future<role_set> query_granted(std::string_view, recursive_role_query) override;
 
-    future<role_to_directly_granted_map> query_all_directly_granted(::service::query_state&) override;
+    future<role_to_directly_granted_map> query_all_directly_granted() override;
 
-    future<role_set> query_all(::service::query_state&) override;
+    future<role_set> query_all() override;
 
     future<bool> exists(std::string_view) override;
 
@@ -85,9 +85,9 @@ class ldap_role_manager : public role_manager {
 
     future<bool> can_login(std::string_view) override;
 
-    future<std::optional<sstring>> get_attribute(std::string_view, std::string_view, ::service::query_state&) override;
+    future<std::optional<sstring>> get_attribute(std::string_view, std::string_view) override;
 
-    future<role_manager::attribute_vals> query_attribute_for_all(std::string_view, ::service::query_state&) override;
+    future<role_manager::attribute_vals> query_attribute_for_all(std::string_view) override;
 
     future<> set_attribute(std::string_view, std::string_view, std::string_view, ::service::group0_batch& mc) override;
 

auth/maintenance_socket_role_manager.cc

@@ -78,11 +78,11 @@ future<role_set> maintenance_socket_role_manager::query_granted(std::string_view
     return operation_not_supported_exception<role_set>("QUERY GRANTED");
 }
 
-future<role_to_directly_granted_map> maintenance_socket_role_manager::query_all_directly_granted(::service::query_state&) {
+future<role_to_directly_granted_map> maintenance_socket_role_manager::query_all_directly_granted() {
     return operation_not_supported_exception<role_to_directly_granted_map>("QUERY ALL DIRECTLY GRANTED");
 }
 
-future<role_set> maintenance_socket_role_manager::query_all(::service::query_state&) {
+future<role_set> maintenance_socket_role_manager::query_all() {
     return operation_not_supported_exception<role_set>("QUERY ALL");
 }
 
@@ -98,11 +98,11 @@ future<bool> maintenance_socket_role_manager::can_login(std::string_view role_na
     return make_ready_future<bool>(true);
 }
 
-future<std::optional<sstring>> maintenance_socket_role_manager::get_attribute(std::string_view role_name, std::string_view attribute_name, ::service::query_state&) {
+future<std::optional<sstring>> maintenance_socket_role_manager::get_attribute(std::string_view role_name, std::string_view attribute_name) {
     return operation_not_supported_exception<std::optional<sstring>>("GET ATTRIBUTE");
 }
 
-future<role_manager::attribute_vals> maintenance_socket_role_manager::query_attribute_for_all(std::string_view attribute_name, ::service::query_state&) {
+future<role_manager::attribute_vals> maintenance_socket_role_manager::query_attribute_for_all(std::string_view attribute_name) {
     return operation_not_supported_exception<role_manager::attribute_vals>("QUERY ATTRIBUTE");
 }
 

auth/maintenance_socket_role_manager.hh

@@ -53,9 +53,9 @@ public:
 
     virtual future<role_set> query_granted(std::string_view grantee_name, recursive_role_query) override;
 
-    virtual future<role_to_directly_granted_map> query_all_directly_granted(::service::query_state&) override;
+    virtual future<role_to_directly_granted_map> query_all_directly_granted() override;
 
-    virtual future<role_set> query_all(::service::query_state&) override;
+    virtual future<role_set> query_all() override;
 
     virtual future<bool> exists(std::string_view role_name) override;
 
@@ -63,9 +63,9 @@ public:
 
     virtual future<bool> can_login(std::string_view role_name) override;
 
-    virtual future<std::optional<sstring>> get_attribute(std::string_view role_name, std::string_view attribute_name, ::service::query_state&) override;
+    virtual future<std::optional<sstring>> get_attribute(std::string_view role_name, std::string_view attribute_name) override;
 
-    virtual future<role_manager::attribute_vals> query_attribute_for_all(std::string_view attribute_name, ::service::query_state&) override;
+    virtual future<role_manager::attribute_vals> query_attribute_for_all(std::string_view attribute_name) override;
 
     virtual future<> set_attribute(std::string_view role_name, std::string_view attribute_name, std::string_view attribute_value, ::service::group0_batch& mc) override;
 

auth/password_authenticator.cc

@@ -48,14 +48,14 @@ static const class_registrator<
         password_authenticator,
         cql3::query_processor&,
         ::service::raft_group0_client&,
-        ::service::migration_manager&,
-        utils::alien_worker&> password_auth_reg("org.apache.cassandra.auth.PasswordAuthenticator");
+        ::service::migration_manager&> password_auth_reg("org.apache.cassandra.auth.PasswordAuthenticator");
 
 static thread_local auto rng_for_salt = std::default_random_engine(std::random_device{}());
 
 static std::string_view get_config_value(std::string_view value, std::string_view def) {
     return value.empty() ? def : value;
 }
+
 std::string password_authenticator::default_superuser(const db::config& cfg) {
     return std::string(get_config_value(cfg.auth_superuser_name(), DEFAULT_USER_NAME));
 }
@@ -63,13 +63,12 @@ std::string password_authenticator::default_superuser(const db::config& cfg) {
 password_authenticator::~password_authenticator() {
 }
 
-password_authenticator::password_authenticator(cql3::query_processor& qp, ::service::raft_group0_client& g0, ::service::migration_manager& mm, utils::alien_worker& hashing_worker)
+password_authenticator::password_authenticator(cql3::query_processor& qp, ::service::raft_group0_client& g0, ::service::migration_manager& mm)
     : _qp(qp)
     , _group0_client(g0)
     , _migration_manager(mm)
     , _stopped(make_ready_future<>()) 
     , _superuser(default_superuser(qp.db().get_config()))
-    , _hashing_worker(hashing_worker)
 {}
 
 static bool has_salted_hash(const cql3::untyped_result_set_row& row) {
@@ -102,11 +101,7 @@ future<> password_authenticator::migrate_legacy_metadata() const {
         return do_for_each(*results, [this](const cql3::untyped_result_set_row& row) {
             auto username = row.get_as<sstring>("username");
             auto salted_hash = row.get_as<sstring>(SALTED_HASH);
-            static const auto query = seastar::format("UPDATE {}.{} SET {} = ? WHERE {} = ?",
-                    meta::legacy::AUTH_KS,
-                    meta::roles_table::name,
-                    SALTED_HASH,
-                    meta::roles_table::role_col_name);
+            static const auto query = update_row_query();
             return _qp.execute_internal(
                     query,
                     consistency_for_user(username),
@@ -122,111 +117,42 @@ future<> password_authenticator::migrate_legacy_metadata() const {
     });
 }
 
-future<> password_authenticator::legacy_create_default_if_missing() {
-    const auto exists = co_await legacy::default_role_row_satisfies(_qp, &has_salted_hash, _superuser);
+future<> password_authenticator::create_default_if_missing() {
+    const auto exists = co_await default_role_row_satisfies(_qp, &has_salted_hash, _superuser);
     if (exists) {
         co_return;
     }
     std::string salted_pwd(get_config_value(_qp.db().get_config().auth_superuser_salted_password(), ""));
     if (salted_pwd.empty()) {
-        salted_pwd = passwords::hash(DEFAULT_USER_PASSWORD, rng_for_salt, _scheme);
+        salted_pwd = passwords::hash(DEFAULT_USER_PASSWORD, rng_for_salt);
     }
-    const auto query = seastar::format("UPDATE {}.{} SET {} = ? WHERE {} = ?",
-            meta::legacy::AUTH_KS,
-            meta::roles_table::name,
-            SALTED_HASH,
-            meta::roles_table::role_col_name);
-    co_await _qp.execute_internal(
+    const auto query = update_row_query();
+    if (legacy_mode(_qp)) {
+        co_await _qp.execute_internal(
             query,
             db::consistency_level::QUORUM,
             internal_distributed_query_state(),
             {salted_pwd, _superuser},
             cql3::query_processor::cache_internal::no);
-    plogger.info("Created default superuser authentication record.");
-}
-
-future<> password_authenticator::maybe_create_default_password() {
-    auto needs_password = [this] () -> future<bool> {
-        const sstring query = seastar::format("SELECT * FROM {}.{} WHERE is_superuser = true ALLOW FILTERING", get_auth_ks_name(_qp), meta::roles_table::name);
-        auto results = co_await _qp.execute_internal(query,
-                db::consistency_level::LOCAL_ONE,
-                internal_distributed_query_state(), cql3::query_processor::cache_internal::yes);
-        // Don't add default password if
-        // - there is no default superuser
-        // - there is a superuser with a password.
-        bool has_default = false;
-        bool has_superuser_with_password = false;
-        for (auto& result : *results) {
-            if (result.get_as<sstring>(meta::roles_table::role_col_name) == _superuser) {
-                has_default = true;
-            }
-            if (has_salted_hash(result)) {
-                has_superuser_with_password = true;
-            }
-        }
-        co_return has_default && !has_superuser_with_password;
-    };
-    if (!co_await needs_password()) {
-        co_return;
-    }
-    // We don't want to start operation earlier to avoid quorum requirement in
-    // a common case.
-    ::service::group0_batch batch(
-            co_await _group0_client.start_operation(_as, get_raft_timeout()));
-    // Check again as the state may have changed before we took the guard (batch).
-    if (!co_await needs_password()) {
-        co_return;
-    }
-    // Set default superuser's password.
-    std::string salted_pwd(get_config_value(_qp.db().get_config().auth_superuser_salted_password(), ""));
-    if (salted_pwd.empty()) {
-        salted_pwd = passwords::hash(DEFAULT_USER_PASSWORD, rng_for_salt, _scheme);
-    }
-    const auto update_query = update_row_query();
-    co_await collect_mutations(_qp, batch, update_query, {salted_pwd, _superuser});
-    co_await std::move(batch).commit(_group0_client, _as, get_raft_timeout());
-    plogger.info("Created default superuser authentication record.");
-}
-
-future<> password_authenticator::maybe_create_default_password_with_retries() {
-    size_t retries = _migration_manager.get_concurrent_ddl_retries();
-    while (true)  {
-        try {
-            co_return co_await maybe_create_default_password();
-        } catch (const ::service::group0_concurrent_modification& ex) {
-            plogger.warn("Failed to execute maybe_create_default_password due to guard conflict.{}.", retries ? " Retrying" : " Number of retries exceeded, giving up");
-            if (retries--) {
-                continue;
-            }
-            // Log error but don't crash the whole node startup sequence.
-            plogger.error("Failed to create default superuser password due to guard conflict.");
-            co_return;
-        } catch (const ::service::raft_operation_timeout_error& ex) {
-            plogger.error("Failed to create default superuser password due to exception: {}", ex.what());
-            co_return;
-        }
+        plogger.info("Created default superuser authentication record.");
+    } else {
+        co_await announce_mutations(_qp, _group0_client, query,
+            {salted_pwd, _superuser}, _as, ::service::raft_timeout{});
+        plogger.info("Created default superuser authentication record.");
     }
 }
 
 future<> password_authenticator::start() {
     return once_among_shards([this] {
-        // Verify that at least one hashing scheme is supported.
-        passwords::detail::verify_scheme(_scheme);
-        plogger.info("Using password hashing scheme: {}", passwords::detail::prefix_for_scheme(_scheme));
-
         _stopped = do_after_system_ready(_as, [this] {
             return async([this] {
                 if (legacy_mode(_qp)) {
                     if (!_superuser_created_promise.available()) {
-                        // Counterintuitively, we mark promise as ready before any startup work
-                        // because wait_for_schema_agreement() below will block indefinitely
-                        // without cluster majority. In that case, blocking node startup
-                        // would lead to a cluster deadlock.
                         _superuser_created_promise.set_value();
                     }
                     _migration_manager.wait_for_schema_agreement(_qp.db().real_database(), db::timeout_clock::time_point::max(), &_as).get();
 
-                    if (legacy::any_nondefault_role_row_satisfies(_qp, &has_salted_hash, _superuser).get()) {
+                    if (any_nondefault_role_row_satisfies(_qp, &has_salted_hash, _superuser).get()) {
                         if (legacy_metadata_exists()) {
                             plogger.warn("Ignoring legacy authentication metadata since nondefault data already exist.");
                         }
@@ -238,34 +164,20 @@ future<> password_authenticator::start() {
                         migrate_legacy_metadata().get();
                         return;
                     }
-                    legacy_create_default_if_missing().get();
                 }
                 utils::get_local_injector().inject("password_authenticator_start_pause", utils::wait_for_message(5min)).get();
+                create_default_if_missing().get();
                 if (!legacy_mode(_qp)) {
-                    maybe_create_default_password_with_retries().get();
-                    if (!_superuser_created_promise.available()) {
-                        _superuser_created_promise.set_value();
-                    }
+                    _superuser_created_promise.set_value();
                 }
             });
         });
 
         if (legacy_mode(_qp)) {
-            static const sstring create_roles_query = fmt::format(
-                    "CREATE TABLE {}.{} ("
-                    "  {} text PRIMARY KEY,"
-                    "  can_login boolean,"
-                    "  is_superuser boolean,"
-                    "  member_of set<text>,"
-                    "  salted_hash text"
-                    ")",
-                    meta::legacy::AUTH_KS,
-                    meta::roles_table::name,
-                    meta::roles_table::role_col_name);
             return create_legacy_metadata_table_if_missing(
                     meta::roles_table::name,
                     _qp,
-                    create_roles_query,
+                    meta::roles_table::creation_query(),
                     _migration_manager);
         }
         return make_ready_future<>();
@@ -316,13 +228,7 @@ future<authenticated_user> password_authenticator::authenticate(
 
     try {
         const std::optional<sstring> salted_hash = co_await get_password_hash(username);
-        if (!salted_hash) {
-            throw exceptions::authentication_exception("Username and/or password are incorrect");
-        }
-        const bool password_match = co_await _hashing_worker.submit<bool>([password = std::move(password), salted_hash = std::move(salted_hash)]{
-            return passwords::check(password, *salted_hash);
-        });
-        if (!password_match) {
+        if (!salted_hash || !passwords::check(password, *salted_hash)) {
             throw exceptions::authentication_exception("Username and/or password are incorrect");
         }
         co_return username;
@@ -346,7 +252,7 @@ future<> password_authenticator::create(std::string_view role_name, const authen
     auto maybe_hash = options.credentials.transform([&] (const auto& creds) -> sstring {
         return std::visit(make_visitor(
                 [&] (const password_option& opt) {
-                    return passwords::hash(opt.password, rng_for_salt, _scheme);
+                    return passwords::hash(opt.password, rng_for_salt);
                 },
                 [] (const hashed_password_option& opt) {
                     return opt.hashed_password;
@@ -389,11 +295,11 @@ future<> password_authenticator::alter(std::string_view role_name, const authent
                 query,
                 consistency_for_user(role_name),
                 internal_distributed_query_state(),
-                {passwords::hash(password, rng_for_salt, _scheme), sstring(role_name)},
+                {passwords::hash(password, rng_for_salt), sstring(role_name)},
                 cql3::query_processor::cache_internal::no).discard_result();
     } else {
         co_await collect_mutations(_qp, mc, query,
-                {passwords::hash(password, rng_for_salt, _scheme), sstring(role_name)});
+                {passwords::hash(password, rng_for_salt), sstring(role_name)});
     }
 }
 

auth/password_authenticator.hh

@@ -15,9 +15,7 @@
 
 #include "db/consistency_level_type.hh"
 #include "auth/authenticator.hh"
-#include "auth/passwords.hh"
 #include "service/raft/raft_group0_client.hh"
-#include "utils/alien_worker.hh"
 
 namespace db {
     class config;
@@ -43,17 +41,14 @@ class password_authenticator : public authenticator {
     ::service::migration_manager& _migration_manager;
     future<> _stopped;
     abort_source _as;
-    std::string _superuser; // default superuser name from the config (may or may not be present in roles table)
+    std::string _superuser;
     shared_promise<> _superuser_created_promise;
-    // We used to also support bcrypt, SHA-256, and MD5 (ref. scylladb#24524).
-    constexpr static auth::passwords::scheme _scheme = passwords::scheme::sha_512;
-    utils::alien_worker& _hashing_worker;
 
 public:
     static db::consistency_level consistency_for_user(std::string_view role_name);
     static std::string default_superuser(const db::config&);
 
-    password_authenticator(cql3::query_processor&, ::service::raft_group0_client&, ::service::migration_manager&, utils::alien_worker&);
+    password_authenticator(cql3::query_processor&, ::service::raft_group0_client&, ::service::migration_manager&);
 
     ~password_authenticator();
 
@@ -94,10 +89,7 @@ private:
 
     future<> migrate_legacy_metadata() const;
 
-    future<> legacy_create_default_if_missing();
-
-    future<> maybe_create_default_password();
-    future<> maybe_create_default_password_with_retries();
+    future<> create_default_if_missing();
 
     sstring update_row_query() const;
 };

auth/passwords.cc

@@ -21,14 +21,18 @@ static thread_local crypt_data tlcrypt = {};
 
 namespace detail {
 
-void verify_scheme(scheme scheme) {
+scheme identify_best_supported_scheme() {
+    const auto all_schemes = { scheme::bcrypt_y, scheme::bcrypt_a, scheme::sha_512, scheme::sha_256, scheme::md5 };
+    // "Random", for testing schemes.
     const sstring random_part_of_salt = "aaaabbbbccccdddd";
 
-    const sstring salt = sstring(prefix_for_scheme(scheme)) + random_part_of_salt;
-    const char* e = crypt_r("fisk", salt.c_str(), &tlcrypt);
+    for (scheme c : all_schemes) {
+        const sstring salt = sstring(prefix_for_scheme(c)) + random_part_of_salt;
+        const char* e = crypt_r("fisk", salt.c_str(), &tlcrypt);
 
-    if (e && (e[0] != '*')) {
-        return;
+        if (e && (e[0] != '*')) {
+            return c;
+        }
     }
 
     throw no_supported_schemes();

auth/passwords.hh

@@ -21,11 +21,10 @@ class no_supported_schemes : public std::runtime_error {
 public:
     no_supported_schemes();
 };
+
 ///
-/// Apache Cassandra uses a library to provide the bcrypt scheme. In ScyllaDB, we use SHA-512
-/// instead of bcrypt for performance and for historical reasons (see scylladb#24524).
-/// Currently, SHA-512 is always chosen as the hashing scheme for new passwords, but the other
-/// algorithms remain supported for CREATE ROLE WITH HASHED PASSWORD and backward compatibility.
+/// Apache Cassandra uses a library to provide the bcrypt scheme. Many Linux implementations do not support bcrypt, so
+/// we support alternatives. The cost is loss of direct compatibility with Apache Cassandra system tables.
 ///
 enum class scheme {
     bcrypt_y,
@@ -52,11 +51,11 @@ sstring generate_random_salt_bytes(RandomNumberEngine& g) {
 }
 
 ///
-/// Test given hashing scheme on the current system.
+/// Test each allowed hashing scheme and report the best supported one on the current system.
 ///
-/// \throws \ref no_supported_schemes when scheme is unsupported.
+/// \throws \ref no_supported_schemes when none of the known schemes is supported.
 ///
-void verify_scheme(scheme scheme);
+scheme identify_best_supported_scheme();
 
 std::string_view prefix_for_scheme(scheme) noexcept;
 
@@ -68,7 +67,8 @@ std::string_view prefix_for_scheme(scheme) noexcept;
 /// \throws \ref no_supported_schemes when no known hashing schemes are supported on the system.
 ///
 template <typename RandomNumberEngine>
-sstring generate_salt(RandomNumberEngine& g, scheme scheme) {
+sstring generate_salt(RandomNumberEngine& g) {
+    static const scheme scheme = identify_best_supported_scheme();
     static const sstring prefix = sstring(prefix_for_scheme(scheme));
     return prefix + generate_random_salt_bytes(g);
 }
@@ -93,8 +93,8 @@ sstring hash_with_salt(const sstring& pass, const sstring& salt);
 /// \throws \ref std::system_error when the implementation-specific implementation fails to hash the cleartext.
 ///
 template <typename RandomNumberEngine>
-sstring hash(const sstring& pass, RandomNumberEngine& g, scheme scheme) {
-    return detail::hash_with_salt(pass, detail::generate_salt(g, scheme));
+sstring hash(const sstring& pass, RandomNumberEngine& g) {
+    return detail::hash_with_salt(pass, detail::generate_salt(g));
 }
 
 ///

auth/permission.cc

@@ -36,8 +36,7 @@ static const std::unordered_map<sstring, auth::permission> permission_names({
         {"MODIFY", auth::permission::MODIFY},
         {"AUTHORIZE", auth::permission::AUTHORIZE},
         {"DESCRIBE", auth::permission::DESCRIBE},
-        {"EXECUTE", auth::permission::EXECUTE},
-        {"VECTOR_SEARCH_INDEXING", auth::permission::VECTOR_SEARCH_INDEXING}});
+        {"EXECUTE", auth::permission::EXECUTE}});
 
 const sstring& auth::permissions::to_string(permission p) {
     for (auto& v : permission_names) {

auth/permission.hh

@@ -33,7 +33,6 @@ enum class permission {
     // data access
     SELECT, // required for SELECT.
     MODIFY, // required for INSERT, UPDATE, DELETE, TRUNCATE.
-    VECTOR_SEARCH_INDEXING, // required for SELECT from tables with vector indexes if SELECT permission is not granted.
 
     // permission management
     AUTHORIZE, // required for GRANT and REVOKE.
@@ -55,8 +54,7 @@ typedef enum_set<
                 permission::MODIFY,
                 permission::AUTHORIZE,
                 permission::DESCRIBE,
-                permission::EXECUTE,
-                permission::VECTOR_SEARCH_INDEXING>> permission_set;
+                permission::EXECUTE>> permission_set;
 
 bool operator<(const permission_set&, const permission_set&);
 

auth/resource.cc

@@ -41,26 +41,22 @@ static const std::unordered_map<resource_kind, std::size_t> max_parts{
         {resource_kind::functions, 2}};
 
 static permission_set applicable_permissions(const data_resource_view& dv) {
-    
-    // We only support VECTOR_SEARCH_INDEXING permission for ALL KEYSPACES.
-
-    auto set = permission_set::of<
+    if (dv.table()) {
+        return permission_set::of<
                 permission::ALTER,
                 permission::DROP,
                 permission::SELECT,
                 permission::MODIFY,
                 permission::AUTHORIZE>();
-
-    if (!dv.table()) {
-        set.add(permission_set::of<permission::CREATE>());
     }
 
-    if (!dv.table() && !dv.keyspace()) {
-        set.add(permission_set::of<permission::VECTOR_SEARCH_INDEXING>());
-    }
-
-    return set;
-        
+    return permission_set::of<
+            permission::CREATE,
+            permission::ALTER,
+            permission::DROP,
+            permission::SELECT,
+            permission::MODIFY,
+            permission::AUTHORIZE>();
 }
 
 static permission_set applicable_permissions(const role_resource_view& rv) {
@@ -197,7 +193,9 @@ service_level_resource_view::service_level_resource_view(const resource &r) {
 
 sstring encode_signature(std::string_view name, std::vector<data_type> args) {
     return seastar::format("{}[{}]", name,
-            fmt::join(args | std::views::transform(&abstract_type::name), "^"));
+            fmt::join(args | std::views::transform([] (const data_type t) {
+                return t->name();
+            }), "^"));
 }
 
 std::pair<sstring, std::vector<data_type>> decode_signature(std::string_view encoded_signature) {
@@ -223,7 +221,9 @@ std::pair<sstring, std::vector<data_type>> decode_signature(std::string_view enc
 static sstring decoded_signature_string(std::string_view encoded_signature) {
     auto [function_name, arg_types] = decode_signature(encoded_signature);
     return seastar::format("{}({})", cql3::util::maybe_quote(sstring(function_name)),
-            fmt::join(arg_types | std::views::transform(&abstract_type::cql3_type_name), ", "));
+            fmt::join(arg_types | std::views::transform([] (data_type t) {
+                return t->cql3_type_name();
+            }), ", "));
 }
 
 resource make_functions_resource(const cql3::functions::function& f) {

auth/role_manager.hh

@@ -17,17 +17,12 @@
 #include <seastar/core/format.hh>
 #include <seastar/core/sstring.hh>
 
-#include "auth/common.hh"
 #include "auth/resource.hh"
 #include "cql3/description.hh"
 #include "seastarx.hh"
 #include "exceptions/exceptions.hh"
 #include "service/raft/raft_group0_client.hh"
 
-namespace service {
-class query_state;
-};
-
 namespace auth {
 
 struct role_config final {
@@ -172,9 +167,9 @@ public:
     ///   (role2, role3)
     /// }
     ///  
-    virtual future<role_to_directly_granted_map> query_all_directly_granted(::service::query_state& = internal_distributed_query_state()) = 0;
+    virtual future<role_to_directly_granted_map> query_all_directly_granted() = 0;
 
-    virtual future<role_set> query_all(::service::query_state& = internal_distributed_query_state()) = 0;
+    virtual future<role_set> query_all() = 0;
 
     virtual future<bool> exists(std::string_view role_name) = 0;
 
@@ -191,12 +186,12 @@ public:
     ///
     /// \returns the value of the named attribute, if one is set.
     ///
-    virtual future<std::optional<sstring>> get_attribute(std::string_view role_name, std::string_view attribute_name, ::service::query_state& = internal_distributed_query_state()) = 0;
+    virtual future<std::optional<sstring>> get_attribute(std::string_view role_name, std::string_view attribute_name) = 0;
 
     ///
     /// \returns a mapping of each role's value for the named attribute, if one is set for the role.
     ///
-    virtual future<attribute_vals> query_attribute_for_all(std::string_view attribute_name, ::service::query_state& = internal_distributed_query_state()) = 0;
+    virtual future<attribute_vals> query_attribute_for_all(std::string_view attribute_name) = 0;
 
     /// Sets `attribute_name` with `attribute_value` for `role_name`.
     /// \returns an exceptional future with nonexistant_role if the role does not exist.

auth/roles-metadata.cc

@@ -18,21 +18,43 @@
 
 namespace auth {
 
-namespace legacy {
+namespace meta {
+
+namespace roles_table {
+
+std::string_view creation_query() {
+    static const sstring instance = fmt::format(
+            "CREATE TABLE {}.{} ("
+            "  {} text PRIMARY KEY,"
+            "  can_login boolean,"
+            "  is_superuser boolean,"
+            "  member_of set<text>,"
+            "  salted_hash text"
+            ")",
+            meta::legacy::AUTH_KS,
+            name,
+            role_col_name);
+
+    return instance;
+}
+
+} // namespace roles_table
+
+} // namespace meta
 
 future<bool> default_role_row_satisfies(
         cql3::query_processor& qp,
         std::function<bool(const cql3::untyped_result_set_row&)> p,
         std::optional<std::string> rolename) {
     const sstring query = seastar::format("SELECT * FROM {}.{} WHERE {} = ?",
-            auth::meta::legacy::AUTH_KS,
+            get_auth_ks_name(qp),
             meta::roles_table::name,
             meta::roles_table::role_col_name);
 
     for (auto cl : { db::consistency_level::ONE, db::consistency_level::QUORUM }) {
         auto results = co_await qp.execute_internal(query, cl
             , internal_distributed_query_state()
-            , {rolename.value_or(std::string(auth::meta::DEFAULT_SUPERUSER_NAME))}
+            , {rolename.value_or(std::string(meta::DEFAULT_SUPERUSER_NAME))}
             , cql3::query_processor::cache_internal::yes
             );
         if (!results->empty()) {
@@ -46,7 +68,7 @@ future<bool> any_nondefault_role_row_satisfies(
         cql3::query_processor& qp,
         std::function<bool(const cql3::untyped_result_set_row&)> p,
         std::optional<std::string> rolename) {
-    const sstring query = seastar::format("SELECT * FROM {}.{}", auth::meta::legacy::AUTH_KS, meta::roles_table::name);
+    const sstring query = seastar::format("SELECT * FROM {}.{}", get_auth_ks_name(qp), meta::roles_table::name);
 
     auto results = co_await qp.execute_internal(query, db::consistency_level::QUORUM
         , internal_distributed_query_state(), cql3::query_processor::cache_internal::no
@@ -63,6 +85,4 @@ future<bool> any_nondefault_role_row_satisfies(
     });
 }
 
-} // namespace legacy
-
-} // namespace auth
+}