cql3: fix documentation and avoid unnecessary copy in execute_paged_internal

- Remove duplicate comment block in header file
- Avoid copying query_state when it's provided by using a reference
  instead of a copy in the ternary expression

Co-authored-by: ptrsmrn <124208650+ptrsmrn@users.noreply.github.com>

.github/CODEOWNERS

@@ -1,5 +1,5 @@
 # AUTH
-auth/* @nuivall
+auth/* @nuivall @ptrsmrn
 
 # CACHE
 row_cache* @tgrabiec
@@ -25,11 +25,11 @@ compaction/* @raphaelsc
 transport/*
 
 # CQL QUERY LANGUAGE
-cql3/* @tgrabiec @nuivall
+cql3/* @tgrabiec @nuivall @ptrsmrn
 
 # COUNTERS
-counters* @nuivall
-tests/counter_test* @nuivall
+counters* @nuivall @ptrsmrn
+tests/counter_test* @nuivall @ptrsmrn
 
 # DOCS
 docs/* @annastuchlik @tzach

.github/copilot-instructions.md

@@ -84,14 +84,3 @@ ninja build/<mode>/scylla
 - Strive for simplicity and clarity, add complexity only when clearly justified
 - Question requests: don't blindly implement requests - evaluate trade-offs, identify issues, and suggest better alternatives when appropriate
 - Consider different approaches, weigh pros and cons, and recommend the best fit for the specific context
-
-## Test Philosophy
-- Performance matters. Tests should run as quickly as possible. Sleeps in the code are highly discouraged and should be avoided, to reduce run time and flakiness.
-- Stability matters. Tests should be stable. New tests should be executed 100 times at least to ensure they pass 100 out of 100 times. (use --repeat 100 --max-failures 1 when running it)
-- Unit tests should ideally test one thing and one thing only.
-- Tests for bug fixes should run before the fix - and show the failure and after the fix - and show they now pass.
-- Tests for bug fixes should have in their comments which bug fixes (GitHub or JIRA issue) they test.
-- Tests in debug are always slower, so if needed, reduce number of iterations, rows, data used, cycles, etc. in debug mode.
-- Tests should strive to be repeatable, and not use random input that will make their results unpredictable.
-- Tests should consume as little resources as possible. Prefer running tests on a single node if it is sufficient, for example.
-

.github/workflows/backport-pr-fixes-validation.yaml

@@ -18,7 +18,7 @@ jobs:
             
             // Regular expression pattern to check for "Fixes" prefix
             // Adjusted to dynamically insert the repository full name
-            const pattern = `Fixes:? ((?:#|${repo.replace('/', '\\/')}#|https://github\\.com/${repo.replace('/', '\\/')}/issues/)(\\d+)|(?:https://scylladb\\.atlassian\\.net/browse/)?([A-Z]+-\\d+))`;
+            const pattern = `Fixes:? (?:#|${repo.replace('/', '\\/')}#|https://github\\.com/${repo.replace('/', '\\/')}/issues/)(\\d+)`;
             const regex = new RegExp(pattern);
             
             if (!regex.test(body)) {

.github/workflows/call_backport_with_jira.yaml

@@ -1,53 +0,0 @@
-name: Backport with Jira Integration
-
-on:
-  push:
-    branches:
-      - master
-      - next-*.*
-      - branch-*.*
-  pull_request_target:
-    types: [labeled, closed]
-    branches: 
-      - master
-      - next
-      - next-*.*
-      - branch-*.*
-
-jobs:
-  backport-on-push:
-    if: github.event_name == 'push'
-    uses: scylladb/github-automation/.github/workflows/backport-with-jira.yaml@main
-    with:
-      event_type: 'push'
-      base_branch: ${{ github.ref }}
-      commits: ${{ github.event.before }}..${{ github.sha }}
-    secrets:
-      gh_token: ${{ secrets.AUTO_BACKPORT_TOKEN }}
-      jira_auth: ${{ secrets.USER_AND_KEY_FOR_JIRA_AUTOMATION }}
-
-  backport-on-label:
-    if: github.event_name == 'pull_request_target' && github.event.action == 'labeled'
-    uses: scylladb/github-automation/.github/workflows/backport-with-jira.yaml@main
-    with:
-      event_type: 'labeled'
-      base_branch: refs/heads/${{ github.event.pull_request.base.ref }}
-      pull_request_number: ${{ github.event.pull_request.number }}
-      head_commit: ${{ github.event.pull_request.base.sha }}
-      label_name: ${{ github.event.label.name }}
-      pr_state: ${{ github.event.pull_request.state }}
-    secrets:
-      gh_token: ${{ secrets.AUTO_BACKPORT_TOKEN }}
-      jira_auth: ${{ secrets.USER_AND_KEY_FOR_JIRA_AUTOMATION }}
-
-  backport-chain:
-    if: github.event_name == 'pull_request_target' && github.event.action == 'closed' && github.event.pull_request.merged == true
-    uses: scylladb/github-automation/.github/workflows/backport-with-jira.yaml@main
-    with:
-      event_type: 'chain'
-      base_branch: refs/heads/${{ github.event.pull_request.base.ref }}
-      pull_request_number: ${{ github.event.pull_request.number }}
-      pr_body: ${{ github.event.pull_request.body }}
-    secrets:
-      gh_token: ${{ secrets.AUTO_BACKPORT_TOKEN }}
-      jira_auth: ${{ secrets.USER_AND_KEY_FOR_JIRA_AUTOMATION }}

.github/workflows/call_jira_status_in_progress.yml

@@ -0,0 +1,12 @@
+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

@@ -0,0 +1,12 @@
+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

@@ -0,0 +1,12 @@
+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/call_jira_sync.yml

@@ -1,41 +0,0 @@
-name: Sync Jira Based on PR Events
-
-on:
-  pull_request_target:
-    types: [opened, ready_for_review, review_requested, labeled, unlabeled, closed]
-
-permissions:
-  contents: read
-  pull-requests: write
-  issues: write
-
-jobs:
-  jira-sync-pr-opened:
-    if: github.event.action == 'opened'
-    uses: scylladb/github-automation/.github/workflows/main_jira_sync_pr_opened.yml@main
-    secrets:
-      caller_jira_auth: ${{ secrets.USER_AND_KEY_FOR_JIRA_AUTOMATION }}
-
-  jira-sync-in-review:
-    if: github.event.action == 'ready_for_review' || github.event.action == 'review_requested'
-    uses: scylladb/github-automation/.github/workflows/main_jira_sync_in_review.yml@main
-    secrets:
-      caller_jira_auth: ${{ secrets.USER_AND_KEY_FOR_JIRA_AUTOMATION }}
-
-  jira-sync-add-label:
-    if: github.event.action == 'labeled'
-    uses: scylladb/github-automation/.github/workflows/main_jira_sync_add_label.yml@main
-    secrets:
-      caller_jira_auth: ${{ secrets.USER_AND_KEY_FOR_JIRA_AUTOMATION }}
-
-  jira-status-remove-label:
-    if: github.event.action == 'unlabeled'
-    uses: scylladb/github-automation/.github/workflows/main_jira_sync_remove_label.yml@main
-    secrets:
-      caller_jira_auth: ${{ secrets.USER_AND_KEY_FOR_JIRA_AUTOMATION }}
-
-  jira-status-pr-closed:
-    if: github.event.action == 'closed' 
-    uses: scylladb/github-automation/.github/workflows/main_jira_sync_pr_closed.yml@main
-    secrets:
-      caller_jira_auth: ${{ secrets.USER_AND_KEY_FOR_JIRA_AUTOMATION }}

.github/workflows/call_jira_sync_pr_milestone.yml

@@ -1,22 +0,0 @@
-name: Sync Jira Based on PR Milestone Events
-
-on:
-  pull_request_target:
-    types: [milestoned, demilestoned]
-
-permissions:
-  contents: read
-  pull-requests: read
-
-jobs:
-  jira-sync-milestone-set:
-    if: github.event.action == 'milestoned'
-    uses: scylladb/github-automation/.github/workflows/main_jira_sync_pr_milestone_set.yml@main
-    secrets:
-      caller_jira_auth: ${{ secrets.USER_AND_KEY_FOR_JIRA_AUTOMATION }}
-
-  jira-sync-milestone-removed:
-    if: github.event.action == 'demilestoned'
-    uses: scylladb/github-automation/.github/workflows/main_jira_sync_pr_milestone_removed.yml@main
-    secrets:
-      caller_jira_auth: ${{ secrets.USER_AND_KEY_FOR_JIRA_AUTOMATION }}

.github/workflows/call_sync_milestone_to_jira.yml

@@ -1,14 +0,0 @@
-name: Call Jira release creation for new milestone
-
-on:
-  milestone:
-    types: [created]
-
-jobs:
-  sync-milestone-to-jira:
-    uses: scylladb/github-automation/.github/workflows/main_sync_milestone_to_jira_release.yml@main
-    with:
-      # Comma-separated list of Jira project keys
-      jira_project_keys: "SCYLLADB,CUSTOMER,SMI"
-    secrets:
-      caller_jira_auth: ${{ secrets.USER_AND_KEY_FOR_JIRA_AUTOMATION }}

.github/workflows/call_validate_pr_author_email.yml

@@ -1,13 +0,0 @@
-name: validate_pr_author_email
-
-on:
-  pull_request_target:
-    types:
-      - opened
-      - synchronize
-      - reopened
-
-jobs:
-  validate_pr_author_email:
-    uses: scylladb/github-automation/.github/workflows/validate_pr_author_email.yml@main
-

.github/workflows/close_issue_for_scylla_associate.yml

@@ -1,62 +0,0 @@
-name: Close issues created by Scylla associates
-
-on:
-  issues:
-    types: [opened, reopened]
-
-permissions:
-  issues: write
-
-jobs:
-  comment-and-close:
-    runs-on: ubuntu-latest
-
-    steps:
-      - name: Comment and close if author email is scylladb.com
-        uses: actions/github-script@v7
-        with:
-          github-token: ${{ secrets.GITHUB_TOKEN }}
-          script: |
-            const issue = context.payload.issue;
-            const actor = context.actor;
-
-            // Get user data (only public email is available)
-            const { data: user } = await github.rest.users.getByUsername({
-              username: actor,
-            });
-
-            const email = user.email || "";
-            console.log(`Actor: ${actor}, public email: ${email || "<none>"}`);
-
-            // Only continue if email exists and ends with @scylladb.com
-            if (!email || !email.toLowerCase().endsWith("@scylladb.com")) {
-              console.log("User is not a scylladb.com email (or email not public); skipping.");
-              return;
-            }
-
-            const owner = context.repo.owner;
-            const repo = context.repo.repo;
-            const issue_number = issue.number;
-
-            const body = "Issues in this repository are closed automatically. Scylla associates should use Jira to manage issues.\nPlease move this issue to Jira https://scylladb.atlassian.net/jira/software/c/projects/SCYLLADB/list";
-
-            // Add the comment
-            await github.rest.issues.createComment({
-              owner,
-              repo,
-              issue_number,
-              body,
-            });
-
-            console.log(`Comment added to #${issue_number}`);
-
-            // Close the issue
-            await github.rest.issues.update({
-              owner,
-              repo,
-              issue_number,
-              state: "closed",
-              state_reason: "not_planned"
-            });
-
-            console.log(`Issue #${issue_number} closed.`);

.github/workflows/codespell.yaml

@@ -13,5 +13,5 @@ jobs:
       - uses: codespell-project/actions-codespell@master
         with:
           only_warn: 1
-          ignore_words_list: "ans,datas,fo,ser,ue,crate,nd,reenable,strat,stap,te,raison,iif,tread"
+          ignore_words_list: "ans,datas,fo,ser,ue,crate,nd,reenable,strat,stap,te,raison"
           skip: "./.git,./build,./tools,*.js,*.lock,./test,./licenses,./redis/lolwut.cc,*.svg"

.github/workflows/docs-pages.yaml

@@ -18,8 +18,6 @@ on:
 
 jobs:
   release:
-    permissions:
-      contents: write
     runs-on: ubuntu-latest
     steps:
       - name: Checkout

.github/workflows/docs-pr.yaml

@@ -2,9 +2,6 @@ name: "Docs / Build PR"
 # For more information,
 # see https://sphinx-theme.scylladb.com/stable/deployment/production.html#available-workflows
 
-permissions:
-  contents: read
-
 env:
   FLAG: ${{ github.repository == 'scylladb/scylla-enterprise' && 'enterprise' || 'opensource' }}
 

.github/workflows/docs-validate-metrics.yml

@@ -1,8 +1,5 @@
 name: Docs / Validate metrics
 
-permissions:
-  contents: read
-
 on:
   pull_request:
     branches:
@@ -10,7 +7,7 @@ on:
       - enterprise
     paths:
       - '**/*.cc'
-      - 'scripts/metrics-config.yml'
+      - 'scripts/metrics-config.yml' 
       - 'scripts/get_description.py'
       - 'docs/_ext/scylladb_metrics.py'
 
@@ -18,20 +15,20 @@ jobs:
   validate-metrics:
     runs-on: ubuntu-latest
     name: Check metrics documentation coverage
-
+    
     steps:
     - name: Checkout code
       uses: actions/checkout@v4
       with:
         submodules: true
-
+      
     - name: Set up Python
       uses: actions/setup-python@v6
       with:
         python-version: '3.10'
-
+        
     - name: Install dependencies
       run: pip install PyYAML
-
+        
     - name: Validate metrics
       run: python3 scripts/get_description.py --validate -c scripts/metrics-config.yml

.github/workflows/iwyu.yaml

@@ -14,8 +14,7 @@ env:
   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
 
-permissions:
-  contents: read
+permissions: {}
 
 # cancel the in-progress run upon a repush
 concurrency:
@@ -35,6 +34,8 @@ jobs:
       - uses: actions/checkout@v4
         with:
           submodules: true
+      - run: |
+          sudo dnf -y install clang-tools-extra
       - name: Generate compilation database
         run: |
           cmake                                         \

.github/workflows/read-toolchain.yaml

@@ -10,8 +10,6 @@ on:
 jobs:
   read-toolchain:
     runs-on: ubuntu-latest
-    permissions:
-      contents: read
     outputs:
       image: ${{ steps.read.outputs.image }}
     steps:

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

@@ -3,58 +3,19 @@ name: Trigger Scylla CI Route
 on:
   issue_comment:
     types: [created]
-  pull_request_target:
-    types:
-      - unlabeled
 
 jobs:
   trigger-jenkins:
-    if: (github.event_name == 'issue_comment' && github.event.comment.user.login != 'scylladbbot') || github.event.label.name == 'conflicts'
+    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: Verify Org Membership
-        id: verify_author
-        shell: bash
-        run: |
-          if [[ "${{ github.event_name }}" == "pull_request_target" ]]; then
-            AUTHOR="${{ github.event.pull_request.user.login }}"
-            ASSOCIATION="${{ github.event.pull_request.author_association }}"
-          else
-            AUTHOR="${{ github.event.comment.user.login }}"
-            ASSOCIATION="${{ github.event.comment.author_association }}"
-          fi
-          if [[ "$ASSOCIATION" == "MEMBER" || "$ASSOCIATION" == "OWNER" ]]; then
-            echo "member=true" >> $GITHUB_OUTPUT
-          else
-            echo "::warning::${AUTHOR} is not a member of scylladb (association: ${ASSOCIATION}); skipping CI trigger."
-            echo "member=false" >> $GITHUB_OUTPUT
-          fi
-
-      - name: Validate Comment Trigger
-        if: github.event_name == 'issue_comment'
-        id: verify_comment
-        shell: bash
-        run: |
-          BODY=$(cat << 'EOF'
-          ${{ github.event.comment.body }}
-          EOF
-          )
-          CLEAN_BODY=$(echo "$BODY" | grep -v '^[[:space:]]*>')
-
-          if echo "$CLEAN_BODY" | grep -qi '@scylladbbot' && echo "$CLEAN_BODY" | grep -qi 'trigger-ci'; then
-            echo "trigger=true" >> $GITHUB_OUTPUT
-          else
-            echo "trigger=false" >> $GITHUB_OUTPUT
-          fi
-
       - name: Trigger Scylla-CI-Route Jenkins Job
-        if: steps.verify_author.outputs.member == 'true' && (github.event_name == 'pull_request_target' || steps.verify_comment.outputs.trigger == 'true')
         env:
           JENKINS_USER: ${{ secrets.JENKINS_USERNAME }}
           JENKINS_API_TOKEN: ${{ secrets.JENKINS_TOKEN }}
           JENKINS_URL: "https://jenkins.scylladb.com"
-          PR_NUMBER: "${{ github.event.issue.number || github.event.pull_request.number }}"
-          PR_REPO_NAME: "${{ github.event.repository.full_name }}"
         run: |
+          PR_NUMBER=${{ 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
+          --user "$JENKINS_USER:$JENKINS_API_TOKEN" --fail -i -v

CMakeLists.txt

@@ -300,6 +300,7 @@ add_subdirectory(locator)
 add_subdirectory(message)
 add_subdirectory(mutation)
 add_subdirectory(mutation_writer)
+add_subdirectory(node_ops)
 add_subdirectory(readers)
 add_subdirectory(replica)
 add_subdirectory(raft)

README.md

@@ -43,7 +43,7 @@ For further information, please see:
 
 [developer documentation]: HACKING.md
 [build documentation]: docs/dev/building.md
-[docker image build documentation]: dist/docker/redhat/README.md
+[docker image build documentation]: dist/docker/debian/README.md
 
 ## Running Scylla
 

SCYLLA-VERSION-GEN

@@ -78,7 +78,7 @@ fi
 
 # Default scylla product/version tags
 PRODUCT=scylla
-VERSION=2026.2.0-dev
+VERSION=2026.1.0-dev
 
 if test -f version
 then

alternator/CMakeLists.txt

@@ -18,7 +18,6 @@ target_sources(alternator
     consumed_capacity.cc
     ttl.cc
     parsed_expression_cache.cc
-    http_compression.cc
     ${cql_grammar_srcs})
 target_include_directories(alternator
   PUBLIC

alternator/conditions.cc

@@ -42,7 +42,7 @@ comparison_operator_type get_comparison_operator(const rjson::value& comparison_
     if (!comparison_operator.IsString()) {
         throw api_error::validation(fmt::format("Invalid comparison operator definition {}", rjson::print(comparison_operator)));
     }
-    std::string op = rjson::to_string(comparison_operator);
+    std::string op = comparison_operator.GetString();
     auto it = ops.find(op);
     if (it == ops.end()) {
         throw api_error::validation(fmt::format("Unsupported comparison operator {}", op));
@@ -377,8 +377,8 @@ bool check_compare(const rjson::value* v1, const rjson::value& v2, const Compara
         return cmp(unwrap_number(*v1, cmp.diagnostic), unwrap_number(v2, cmp.diagnostic));
     }
     if (kv1.name == "S") {
-        return cmp(rjson::to_string_view(kv1.value),
-                   rjson::to_string_view(kv2.value));
+        return cmp(std::string_view(kv1.value.GetString(), kv1.value.GetStringLength()),
+                   std::string_view(kv2.value.GetString(), kv2.value.GetStringLength()));
     }
     if (kv1.name == "B") {
         auto d_kv1 = unwrap_bytes(kv1.value, v1_from_query);
@@ -470,9 +470,9 @@ static bool check_BETWEEN(const rjson::value* v, const rjson::value& lb, const r
         return check_BETWEEN(unwrap_number(*v, diag), unwrap_number(lb, diag), unwrap_number(ub, diag), bounds_from_query);
     }
     if (kv_v.name == "S") {
-        return check_BETWEEN(rjson::to_string_view(kv_v.value),
-                             rjson::to_string_view(kv_lb.value),
-                             rjson::to_string_view(kv_ub.value),
+        return check_BETWEEN(std::string_view(kv_v.value.GetString(), kv_v.value.GetStringLength()),
+                             std::string_view(kv_lb.value.GetString(), kv_lb.value.GetStringLength()),
+                             std::string_view(kv_ub.value.GetString(), kv_ub.value.GetStringLength()),
                              bounds_from_query);
     }
     if (kv_v.name == "B") {
@@ -618,7 +618,7 @@ conditional_operator_type get_conditional_operator(const rjson::value& req) {
 // Check if the existing values of the item (previous_item) match the
 // conditions given by the Expected and ConditionalOperator parameters
 // (if they exist) in the request (an UpdateItem, PutItem or DeleteItem).
-// This function can throw a ValidationException API error if there
+// This function can throw an ValidationException API error if there
 // are errors in the format of the condition itself.
 bool verify_expected(const rjson::value& req, const rjson::value* previous_item) {
     const rjson::value* expected = rjson::find(req, "Expected");

alternator/consumed_capacity.cc

@@ -8,8 +8,6 @@
 
 #include "consumed_capacity.hh"
 #include "error.hh"
-#include "utils/rjson.hh"
-#include <fmt/format.h>
 
 namespace alternator {
 
@@ -34,18 +32,18 @@ bool consumed_capacity_counter::should_add_capacity(const rjson::value& request)
     if (!return_consumed->IsString()) {
         throw api_error::validation("Non-string ReturnConsumedCapacity field in request");
     }
-    std::string_view consumed = rjson::to_string_view(*return_consumed);
+    std::string consumed = return_consumed->GetString();
     if (consumed == "INDEXES") {
         throw api_error::validation("INDEXES consumed capacity is not supported");
     }
     if (consumed != "TOTAL") {
-        throw api_error::validation(fmt::format("Unknown consumed capacity {}", consumed));
+        throw api_error::validation("Unknown consumed capacity "+ consumed);
     }
     return true;
 }
 
 void consumed_capacity_counter::add_consumed_capacity_to_response_if_needed(rjson::value& response) const noexcept {
-    if (_should_add_to_response) {
+    if (_should_add_to_reponse) {
         auto consumption = rjson::empty_object();
         rjson::add(consumption, "CapacityUnits", get_consumed_capacity_units());
         rjson::add(response, "ConsumedCapacity", std::move(consumption));

alternator/consumed_capacity.hh

@@ -28,9 +28,9 @@ namespace alternator {
 class consumed_capacity_counter {
 public:
     consumed_capacity_counter() = default;
-    consumed_capacity_counter(bool should_add_to_response) : _should_add_to_response(should_add_to_response){}
+    consumed_capacity_counter(bool should_add_to_reponse) : _should_add_to_reponse(should_add_to_reponse){}
     bool operator()() const noexcept {
-        return _should_add_to_response;
+        return _should_add_to_reponse;
     }
 
     consumed_capacity_counter& operator +=(uint64_t bytes);
@@ -44,7 +44,7 @@ public:
     uint64_t _total_bytes = 0;
     static bool should_add_capacity(const rjson::value& request);
 protected:
-    bool _should_add_to_response = false;
+    bool _should_add_to_reponse = false;
 };
 
 class rcu_consumed_capacity_counter : public consumed_capacity_counter {

alternator/controller.cc

@@ -28,7 +28,6 @@ static logging::logger logger("alternator_controller");
 controller::controller(
         sharded<gms::gossiper>& gossiper,
         sharded<service::storage_proxy>& proxy,
-        sharded<service::storage_service>& ss,
         sharded<service::migration_manager>& mm,
         sharded<db::system_distributed_keyspace>& sys_dist_ks,
         sharded<cdc::generation_service>& cdc_gen_svc,
@@ -40,7 +39,6 @@ controller::controller(
     : protocol_server(sg)
     , _gossiper(gossiper)
     , _proxy(proxy)
-    , _ss(ss)
     , _mm(mm)
     , _sys_dist_ks(sys_dist_ks)
     , _cdc_gen_svc(cdc_gen_svc)
@@ -91,7 +89,7 @@ future<> controller::start_server() {
         auto get_timeout_in_ms = [] (const db::config& cfg) -> utils::updateable_value<uint32_t> {
             return cfg.alternator_timeout_in_ms;
         };
-        _executor.start(std::ref(_gossiper), std::ref(_proxy), std::ref(_ss), std::ref(_mm), std::ref(_sys_dist_ks),
+        _executor.start(std::ref(_gossiper), std::ref(_proxy), std::ref(_mm), std::ref(_sys_dist_ks),
                         sharded_parameter(get_cdc_metadata, std::ref(_cdc_gen_svc)), _ssg.value(),
                         sharded_parameter(get_timeout_in_ms, std::ref(_config))).get();
         _server.start(std::ref(_executor), std::ref(_proxy), std::ref(_gossiper), std::ref(_auth_service), std::ref(_sl_controller)).get();
@@ -105,23 +103,11 @@ future<> controller::start_server() {
             alternator_port = _config.alternator_port();
             _listen_addresses.push_back({addr, *alternator_port});
         }
-        std::optional<uint16_t> alternator_port_proxy_protocol;
-        if (_config.alternator_port_proxy_protocol()) {
-            alternator_port_proxy_protocol = _config.alternator_port_proxy_protocol();
-            _listen_addresses.push_back({addr, *alternator_port_proxy_protocol});
-        }
         std::optional<uint16_t> alternator_https_port;
-        std::optional<uint16_t> alternator_https_port_proxy_protocol;
         std::optional<tls::credentials_builder> creds;
-        if (_config.alternator_https_port() || _config.alternator_https_port_proxy_protocol()) {
-            if (_config.alternator_https_port()) {
-                alternator_https_port = _config.alternator_https_port();
-                _listen_addresses.push_back({addr, *alternator_https_port});
-            }
-            if (_config.alternator_https_port_proxy_protocol()) {
-                alternator_https_port_proxy_protocol = _config.alternator_https_port_proxy_protocol();
-                _listen_addresses.push_back({addr, *alternator_https_port_proxy_protocol});
-            }
+        if (_config.alternator_https_port()) {
+            alternator_https_port = _config.alternator_https_port();
+            _listen_addresses.push_back({addr, *alternator_https_port});
             creds.emplace();
             auto opts = _config.alternator_encryption_options();
             if (opts.empty()) {
@@ -147,29 +133,20 @@ future<> controller::start_server() {
             }
         }
         _server.invoke_on_all(
-                [this, addr, alternator_port, alternator_https_port, alternator_port_proxy_protocol, alternator_https_port_proxy_protocol, creds = std::move(creds)] (server& server) mutable {
-            return server.init(addr, alternator_port, alternator_https_port, alternator_port_proxy_protocol, alternator_https_port_proxy_protocol, creds,
+                [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_warn_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, alternator_port_proxy_protocol, alternator_https_port_proxy_protocol] (std::exception_ptr ep) {
-            logger.error("Failed to set up Alternator HTTP server on {} port {}, TLS port {}, proxy-protocol port {}, TLS proxy-protocol port {}: {}",
-                    addr,
-                    alternator_port ? std::to_string(*alternator_port) : "OFF",
-                    alternator_https_port ? std::to_string(*alternator_https_port) : "OFF",
-                    alternator_port_proxy_protocol ? std::to_string(*alternator_port_proxy_protocol) : "OFF",
-                    alternator_https_port_proxy_protocol ? std::to_string(*alternator_https_port_proxy_protocol) : "OFF",
-                    ep);
+        }).handle_exception([this, addr, alternator_port, alternator_https_port] (std::exception_ptr ep) {
+            logger.error("Failed to set up Alternator HTTP server on {} port {}, TLS port {}: {}",
+                    addr, alternator_port ? std::to_string(*alternator_port) : "OFF", alternator_https_port ? std::to_string(*alternator_https_port) : "OFF", ep);
             return stop_server().then([ep = std::move(ep)] { return make_exception_future<>(ep); });
-        }).then([addr, alternator_port, alternator_https_port, alternator_port_proxy_protocol, alternator_https_port_proxy_protocol] {
-            logger.info("Alternator server listening on {}, HTTP port {}, HTTPS port {}, proxy-protocol port {}, TLS proxy-protocol port {}",
-                    addr,
-                    alternator_port ? std::to_string(*alternator_port) : "OFF",
-                    alternator_https_port ? std::to_string(*alternator_https_port) : "OFF",
-                    alternator_port_proxy_protocol ? std::to_string(*alternator_port_proxy_protocol) : "OFF",
-                    alternator_https_port_proxy_protocol ? std::to_string(*alternator_https_port_proxy_protocol) : "OFF");
+        }).then([addr, alternator_port, alternator_https_port] {
+            logger.info("Alternator server listening on {}, HTTP port {}, HTTPS port {}",
+                    addr, alternator_port ? std::to_string(*alternator_port) : "OFF", alternator_https_port ? std::to_string(*alternator_https_port) : "OFF");
         }).get();
     });
 }
@@ -192,7 +169,7 @@ future<> controller::request_stop_server() {
     });
 }
 
-future<utils::chunked_vector<foreign_ptr<std::unique_ptr<client_data>>>> controller::get_client_data() {
+future<utils::chunked_vector<client_data>> controller::get_client_data() {
     return _server.local().get_client_data();
 }
 

alternator/controller.hh

@@ -15,7 +15,6 @@
 
 namespace service {
 class storage_proxy;
-class storage_service;
 class migration_manager;
 class memory_limiter;
 }
@@ -58,7 +57,6 @@ class server;
 class controller : public protocol_server {
     sharded<gms::gossiper>& _gossiper;
     sharded<service::storage_proxy>& _proxy;
-    sharded<service::storage_service>& _ss;
     sharded<service::migration_manager>& _mm;
     sharded<db::system_distributed_keyspace>& _sys_dist_ks;
     sharded<cdc::generation_service>& _cdc_gen_svc;
@@ -76,7 +74,6 @@ public:
     controller(
         sharded<gms::gossiper>& gossiper,
         sharded<service::storage_proxy>& proxy,
-        sharded<service::storage_service>& ss,
         sharded<service::migration_manager>& mm,
         sharded<db::system_distributed_keyspace>& sys_dist_ks,
         sharded<cdc::generation_service>& cdc_gen_svc,
@@ -96,7 +93,7 @@ public:
     // 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<foreign_ptr<std::unique_ptr<client_data>>>> get_client_data() override;
+    virtual future<utils::chunked_vector<client_data>> get_client_data() override;
 };
 
 }

alternator/executor.cc

@@ -17,7 +17,6 @@
 #include "auth/service.hh"
 #include "db/config.hh"
 #include "db/view/view_build_status.hh"
-#include "locator/tablets.hh"
 #include "mutation/tombstone.hh"
 #include "locator/abstract_replication_strategy.hh"
 #include "utils/log.hh"
@@ -63,20 +62,11 @@
 #include "types/types.hh"
 #include "db/system_keyspace.hh"
 #include "cql3/statements/ks_prop_defs.hh"
-#include "alternator/ttl_tag.hh"
 
 using namespace std::chrono_literals;
 
 logging::logger elogger("alternator-executor");
 
-namespace std {
-    template <> struct hash<std::pair<sstring, sstring>> {
-        size_t operator () (const std::pair<sstring, sstring>& p) const {
-            return std::hash<sstring>()(p.first) * 1009 + std::hash<sstring>()(p.second) * 3;
-        }
-    };
-}
-
 namespace alternator {
 
 // Alternator-specific table properties stored as hidden table tags:
@@ -165,7 +155,7 @@ static map_type attrs_type() {
 
 static const column_definition& attrs_column(const schema& schema) {
     const column_definition* cdef = schema.get_column_definition(bytes(executor::ATTRS_COLUMN_NAME));
-    throwing_assert(cdef);
+    SCYLLA_ASSERT(cdef);
     return *cdef;
 }
 
@@ -238,7 +228,7 @@ static void validate_is_object(const rjson::value& value, const char* caller) {
 }
 
 // This function assumes the given value is an object and returns requested member value.
-// If it is not possible, an api_error::validation is thrown.
+// If it is not possible an api_error::validation is thrown.
 static const rjson::value& get_member(const rjson::value& obj, const char* member_name, const char* caller) {
     validate_is_object(obj, caller);
     const rjson::value* ret = rjson::find(obj, member_name);
@@ -250,7 +240,7 @@ static const rjson::value& get_member(const rjson::value& obj, const char* membe
 
 
 // This function assumes the given value is an object with a single member, and returns this member.
-// In case the requirements are not met, an api_error::validation is thrown.
+// In case the requirements are not met an api_error::validation is thrown.
 static const rjson::value::Member& get_single_member(const rjson::value& v, const char* caller) {
     if (!v.IsObject() || v.MemberCount() != 1) {
         throw api_error::validation(format("{}: expected an object with a single member.", caller));
@@ -258,66 +248,14 @@ static const rjson::value::Member& get_single_member(const rjson::value& v, cons
     return *(v.MemberBegin());
 }
 
-class executor::describe_table_info_manager : public service::migration_listener::empty_listener {
-    executor &_executor;
-
-    struct table_info {
-        utils::simple_value_with_expiry<std::uint64_t> size_in_bytes;
-    };
-    std::unordered_map<std::pair<sstring, sstring>, table_info> info_for_tables;
-    bool active = false;
-
-public:
-    describe_table_info_manager(executor& executor) : _executor(executor) {
-        _executor._proxy.data_dictionary().real_database_ptr()->get_notifier().register_listener(this);
-        active = true;
-    }
-    describe_table_info_manager(const describe_table_info_manager &) = delete;
-    describe_table_info_manager(describe_table_info_manager&&) = delete;
-    ~describe_table_info_manager() {
-        if (active) {
-            on_fatal_internal_error(elogger, "describe_table_info_manager was not stopped before destruction");
-        }
-    }
-
-    describe_table_info_manager &operator = (const describe_table_info_manager &) = delete;
-    describe_table_info_manager &operator = (describe_table_info_manager&&) = delete;
-
-    static std::chrono::high_resolution_clock::time_point now() {
-        return std::chrono::high_resolution_clock::now();
-    }
-
-    std::optional<std::uint64_t> get_cached_size_in_bytes(const sstring &ks_name, const sstring &cf_name) const {
-        auto it = info_for_tables.find({ks_name, cf_name});
-        if (it != info_for_tables.end()) {
-            return it->second.size_in_bytes.get();
-        }
-        return std::nullopt;
-    }
-    void cache_size_in_bytes(sstring ks_name, sstring cf_name, std::uint64_t size_in_bytes, std::chrono::high_resolution_clock::time_point expiry) {
-        info_for_tables[{std::move(ks_name), std::move(cf_name)}].size_in_bytes.set_if_longer_expiry(size_in_bytes, expiry);
-    }
-    future<> stop() {
-        co_await _executor._proxy.data_dictionary().real_database_ptr()->get_notifier().unregister_listener(this);
-        active = false;
-        co_return;
-    }
-    void on_drop_column_family(const sstring& ks_name, const sstring& cf_name) override {
-        if (!ks_name.starts_with(executor::KEYSPACE_NAME_PREFIX)) return;
-        info_for_tables.erase({ks_name, cf_name});
-    }
-};
-
 executor::executor(gms::gossiper& gossiper,
          service::storage_proxy& proxy,
-         service::storage_service& ss,
          service::migration_manager& mm,
          db::system_distributed_keyspace& sdks,
          cdc::metadata& cdc_metadata,
          smp_service_group ssg,
          utils::updateable_value<uint32_t> default_timeout_in_ms)
     : _gossiper(gossiper),
-      _ss(ss),
       _proxy(proxy),
       _mm(mm),
       _sdks(sdks),
@@ -330,7 +268,6 @@ executor::executor(gms::gossiper& gossiper,
         _stats))
 {
     s_default_timeout_in_ms = std::move(default_timeout_in_ms);
-    _describe_table_info_manager = std::make_unique<describe_table_info_manager>(*this);
     register_metrics(_metrics, _stats);
 }
 
@@ -482,7 +419,7 @@ static std::optional<std::string> find_table_name(const rjson::value& request) {
     if (!table_name_value->IsString()) {
         throw api_error::validation("Non-string TableName field in request");
     }
-    std::string table_name = rjson::to_string(*table_name_value);
+    std::string table_name = table_name_value->GetString();
     return table_name;
 }
 
@@ -609,7 +546,7 @@ get_table_or_view(service::storage_proxy& proxy, const rjson::value& request) {
             // does exist but the index does not (ValidationException).
             if (proxy.data_dictionary().has_schema(keyspace_name, orig_table_name)) {
                 throw api_error::validation(
-                    fmt::format("Requested resource not found: Index '{}' for table '{}'", rjson::to_string_view(*index_name), orig_table_name));
+                    fmt::format("Requested resource not found: Index '{}' for table '{}'", index_name->GetString(), orig_table_name));
             } else {
                 throw api_error::resource_not_found(
                     fmt::format("Requested resource not found: Table: {} not found", orig_table_name));
@@ -650,7 +587,7 @@ static std::string get_string_attribute(const rjson::value& value, std::string_v
         throw api_error::validation(fmt::format("Expected string value for attribute {}, got: {}",
                 attribute_name, value));
     }
-    return rjson::to_string(*attribute_value);
+    return std::string(attribute_value->GetString(), attribute_value->GetStringLength());
 }
 
 // Convenience function for getting the value of a boolean attribute, or a
@@ -683,7 +620,7 @@ static std::optional<int> get_int_attribute(const rjson::value& value, std::stri
 }
 
 // Sets a KeySchema object inside the given JSON parent describing the key
-// attributes of the given schema as being either HASH or RANGE keys.
+// attributes of the the given schema as being either HASH or RANGE keys.
 // Additionally, adds to a given map mappings between the key attribute
 // names and their type (as a DynamoDB type string).
 void executor::describe_key_schema(rjson::value& parent, const schema& schema, std::unordered_map<std::string,std::string>* attribute_types, const std::map<sstring, sstring> *tags) {
@@ -815,44 +752,12 @@ static future<bool> is_view_built(
 
 }
 
-future<> executor::cache_newly_calculated_size_on_all_shards(schema_ptr schema, std::uint64_t size_in_bytes, std::chrono::nanoseconds ttl) {
-    auto expiry = describe_table_info_manager::now() + ttl;
-    return container().invoke_on_all(
-        [schema, size_in_bytes, expiry] (executor& exec) {
-            exec._describe_table_info_manager->cache_size_in_bytes(schema->ks_name(), schema->cf_name(), size_in_bytes, expiry);
-        });
-}
-
-future<> executor::fill_table_size(rjson::value &table_description, schema_ptr schema, bool deleting) {
-    auto cached_size = _describe_table_info_manager->get_cached_size_in_bytes(schema->ks_name(), schema->cf_name());
-    std::uint64_t total_size = 0;
-    if (cached_size) {
-        total_size = *cached_size;
-    } else {
-        // there's no point in trying to estimate value of table that is being deleted, as other nodes more often than not might
-        // move forward with deletion faster than we calculate the size
-        if (!deleting) {
-            total_size = co_await _ss.estimate_total_sstable_volume(schema->id(), service::storage_service::ignore_errors::yes);
-            const auto expiry = std::chrono::seconds{ _proxy.data_dictionary().get_config().alternator_describe_table_info_cache_validity_in_seconds() };
-            // Note: we don't care when the notification of other shards will finish, as long as it will be done
-            // it's possible to get into race condition (next DescribeTable comes to other shard, that new shard doesn't have
-            // the size yet, so it will calculate it again) - this is not a problem, because it will call cache_newly_calculated_size_on_all_shards
-            // with expiry, which is extremely unlikely to be exactly the same as the previous one, all shards will keep the size coming with expiry that is further into the future.
-            // In case of the same expiry, some shards will have different size, which means DescribeTable will return different values depending on the shard
-            // which is also fine, as the specification doesn't give precision guarantees of any kind.
-            co_await cache_newly_calculated_size_on_all_shards(schema, total_size, expiry);
-        }
-    }
-    rjson::add(table_description, "TableSizeBytes", total_size);
-}
-
-future<rjson::value> executor::fill_table_description(schema_ptr schema, table_status tbl_status, service::client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit)
+static future<rjson::value> fill_table_description(schema_ptr schema, table_status tbl_status, service::storage_proxy& proxy, service::client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit)
 {
     rjson::value table_description = rjson::empty_object();
     auto tags_ptr = db::get_tags_of_table(schema);
 
     rjson::add(table_description, "TableName", rjson::from_string(schema->cf_name()));
-    co_await fill_table_size(table_description, schema, tbl_status == table_status::deleting);
 
     auto creation_timestamp = get_table_creation_time(*schema);
 
@@ -896,7 +801,9 @@ future<rjson::value> executor::fill_table_description(schema_ptr schema, table_s
     rjson::add(table_description["ProvisionedThroughput"], "WriteCapacityUnits", wcu);
     rjson::add(table_description["ProvisionedThroughput"], "NumberOfDecreasesToday", 0);
 
-    data_dictionary::table t = _proxy.data_dictionary().find_column_family(schema);
+
+
+    data_dictionary::table t = proxy.data_dictionary().find_column_family(schema);
 
     if (tbl_status != table_status::deleting) {
         rjson::add(table_description, "CreationDateTime", rjson::value(creation_timestamp));
@@ -917,7 +824,7 @@ future<rjson::value> executor::fill_table_description(schema_ptr schema, table_s
                 sstring index_name = cf_name.substr(delim_it + 1);
                 rjson::add(view_entry, "IndexName", rjson::from_string(index_name));
                 rjson::add(view_entry, "IndexArn", generate_arn_for_index(*schema, index_name));
-                // Add index's KeySchema and collect types for AttributeDefinitions:
+                // Add indexes's KeySchema and collect types for AttributeDefinitions:
                 executor::describe_key_schema(view_entry, *vptr, key_attribute_types, db::get_tags_of_table(vptr));
                 // Add projection type
                 rjson::value projection = rjson::empty_object();
@@ -933,7 +840,7 @@ future<rjson::value> executor::fill_table_description(schema_ptr schema, table_s
                 // (for a built view) or CREATING+Backfilling (if view building
                 // is in progress).
                 if (!is_lsi) {
-                    if (co_await is_view_built(vptr, _proxy, client_state, trace_state, permit)) {
+                    if (co_await is_view_built(vptr, proxy, client_state, trace_state, permit)) {
                         rjson::add(view_entry, "IndexStatus", "ACTIVE");
                     } else {
                         rjson::add(view_entry, "IndexStatus", "CREATING");
@@ -961,8 +868,9 @@ future<rjson::value> executor::fill_table_description(schema_ptr schema, table_s
         }
         rjson::add(table_description, "AttributeDefinitions", std::move(attribute_definitions));
     }
-    executor::supplement_table_stream_info(table_description, *schema, _proxy);
+    executor::supplement_table_stream_info(table_description, *schema, proxy);
 
+    // FIXME: still missing some response fields (issue #5026)
     co_return table_description;
 }
 
@@ -982,7 +890,7 @@ future<executor::request_return_type> executor::describe_table(client_state& cli
     get_stats_from_schema(_proxy, *schema)->api_operations.describe_table++;
     tracing::add_alternator_table_name(trace_state, schema->cf_name());
 
-    rjson::value table_description = co_await fill_table_description(schema, table_status::active, client_state, trace_state, permit);
+    rjson::value table_description = co_await fill_table_description(schema, table_status::active, _proxy, client_state, trace_state, permit);
     rjson::value response = rjson::empty_object();
     rjson::add(response, "Table", std::move(table_description));
     elogger.trace("returning {}", response);
@@ -1085,7 +993,7 @@ future<executor::request_return_type> executor::delete_table(client_state& clien
     auto& p = _proxy.container();
 
     schema_ptr schema = get_table(_proxy, request);
-    rjson::value table_description = co_await fill_table_description(schema, table_status::deleting, client_state, trace_state, permit);
+    rjson::value table_description = co_await fill_table_description(schema, table_status::deleting, _proxy, client_state, trace_state, permit);
     co_await verify_permission(_enforce_authorization, _warn_authorization, client_state, schema, auth::permission::DROP, _stats);
     co_await _mm.container().invoke_on(0, [&, cs = client_state.move_to_other_shard()] (service::migration_manager& mm) -> future<> {
         size_t retries = mm.get_concurrent_ddl_retries();
@@ -1172,8 +1080,8 @@ static void add_column(schema_builder& builder, const std::string& name, const r
     }
     for (auto it = attribute_definitions.Begin(); it != attribute_definitions.End(); ++it) {
         const rjson::value& attribute_info = *it;
-        if (rjson::to_string_view(attribute_info["AttributeName"]) == name) {
-            std::string_view type = rjson::to_string_view(attribute_info["AttributeType"]);
+        if (attribute_info["AttributeName"].GetString() == name) {
+            auto type = attribute_info["AttributeType"].GetString();
             data_type dt = parse_key_type(type);
             if (computed_column) {
                 // Computed column for GSI (doesn't choose a real column as-is
@@ -1208,7 +1116,7 @@ static std::pair<std::string, std::string> parse_key_schema(const rjson::value&
         throw api_error::validation("First element of KeySchema must be an object");
     }
     const rjson::value *v = rjson::find((*key_schema)[0], "KeyType");
-    if (!v || !v->IsString() || rjson::to_string_view(*v) != "HASH") {
+    if (!v || !v->IsString() || v->GetString() != std::string("HASH")) {
         throw api_error::validation("First key in KeySchema must be a HASH key");
     }
     v = rjson::find((*key_schema)[0], "AttributeName");
@@ -1216,14 +1124,14 @@ static std::pair<std::string, std::string> parse_key_schema(const rjson::value&
         throw api_error::validation("First key in KeySchema must have string AttributeName");
     }
     validate_attr_name_length(supplementary_context, v->GetStringLength(), true, "HASH key in KeySchema - ");
-    std::string hash_key = rjson::to_string(*v);
+    std::string hash_key = v->GetString();
     std::string range_key;
     if (key_schema->Size() == 2) {
         if (!(*key_schema)[1].IsObject()) {
             throw api_error::validation("Second element of KeySchema must be an object");
         }
         v = rjson::find((*key_schema)[1], "KeyType");
-        if (!v || !v->IsString() || rjson::to_string_view(*v) != "RANGE") {
+        if (!v || !v->IsString() || v->GetString() != std::string("RANGE")) {
             throw api_error::validation("Second key in KeySchema must be a RANGE key");
         }
         v = rjson::find((*key_schema)[1], "AttributeName");
@@ -1649,8 +1557,9 @@ static future<> mark_view_schemas_as_built(utils::chunked_vector<mutation>& out,
     }
 }
 
-future<executor::request_return_type> executor::create_table_on_shard0(service::client_state&& client_state, tracing::trace_state_ptr trace_state, rjson::value request, bool enforce_authorization, bool warn_authorization, const db::tablets_mode_t::mode tablets_mode) {
-    throwing_assert(this_shard_id() == 0);
+static future<executor::request_return_type> create_table_on_shard0(service::client_state&& client_state, tracing::trace_state_ptr trace_state, rjson::value request,
+            service::storage_proxy& sp, service::migration_manager& mm, gms::gossiper& gossiper, bool enforce_authorization, bool warn_authorization, stats& stats, const db::tablets_mode_t::mode tablets_mode) {
+    SCYLLA_ASSERT(this_shard_id() == 0);
 
     // We begin by parsing and validating the content of the CreateTable
     // command. We can't inspect the current database schema at this point
@@ -1836,7 +1745,7 @@ future<executor::request_return_type> executor::create_table_on_shard0(service::
 
     rjson::value* stream_specification = rjson::find(request, "StreamSpecification");
     if (stream_specification && stream_specification->IsObject()) {
-        if (executor::add_stream_options(*stream_specification, builder, _proxy)) {
+        if (executor::add_stream_options(*stream_specification, builder, sp)) {
             validate_cdc_log_name_length(builder.cf_name());
         }
     }
@@ -1855,7 +1764,7 @@ future<executor::request_return_type> executor::create_table_on_shard0(service::
     set_table_creation_time(tags_map, db_clock::now());
     builder.add_extension(db::tags_extension::NAME, ::make_shared<db::tags_extension>(tags_map));
 
-    co_await verify_create_permission(enforce_authorization, warn_authorization, client_state, _stats);
+    co_await verify_create_permission(enforce_authorization, warn_authorization, client_state, stats);
 
     schema_ptr schema = builder.build();
     for (auto& view_builder : view_builders) {
@@ -1871,49 +1780,33 @@ future<executor::request_return_type> executor::create_table_on_shard0(service::
         view_builder.with_view_info(schema, include_all_columns, ""/*where clause*/);
     }
 
-    size_t retries = _mm.get_concurrent_ddl_retries();
+    size_t retries = mm.get_concurrent_ddl_retries();
     for (;;) {
-        auto group0_guard = co_await _mm.start_group0_operation();
+        auto group0_guard = co_await mm.start_group0_operation();
         auto ts = group0_guard.write_timestamp();
         utils::chunked_vector<mutation> schema_mutations;
-        auto ksm = create_keyspace_metadata(keyspace_name, _proxy, _gossiper, ts, tags_map, _proxy.features(), tablets_mode);
-        locator::replication_strategy_params params(ksm->strategy_options(), ksm->initial_tablets(), ksm->consistency_option());
-        const auto& topo = _proxy.local_db().get_token_metadata().get_topology();
-        auto rs = locator::abstract_replication_strategy::create_replication_strategy(ksm->strategy_name(), params, topo);
+        auto ksm = create_keyspace_metadata(keyspace_name, sp, gossiper, ts, tags_map, sp.features(), tablets_mode);
         // Alternator Streams doesn't yet work when the table uses tablets (#23838)
         if (stream_specification && stream_specification->IsObject()) {
             auto stream_enabled = rjson::find(*stream_specification, "StreamEnabled");
             if (stream_enabled && stream_enabled->IsBool() && stream_enabled->GetBool()) {
+                locator::replication_strategy_params params(ksm->strategy_options(), ksm->initial_tablets(), ksm->consistency_option());
+                const auto& topo = sp.local_db().get_token_metadata().get_topology();
+                auto rs = locator::abstract_replication_strategy::create_replication_strategy(ksm->strategy_name(), params, topo);
                 if (rs->uses_tablets()) {
                     co_return api_error::validation("Streams not yet supported on a table using tablets (issue #23838). "
                     "If you want to use streams, create a table with vnodes by setting the tag 'system:initial_tablets' set to 'none'.");
                 }
             }
         }
-        // Creating an index in tablets mode requires the keyspace to be RF-rack-valid.
-        // GSI and LSI indexes are based on materialized views which require RF-rack-validity to avoid consistency issues.
-        if (!view_builders.empty() || _proxy.data_dictionary().get_config().rf_rack_valid_keyspaces()) {
-            try {
-                locator::assert_rf_rack_valid_keyspace(keyspace_name, _proxy.local_db().get_token_metadata_ptr(), *rs);
-            } catch (const std::invalid_argument& ex) {
-                if (!view_builders.empty()) {
-                    co_return api_error::validation(fmt::format("GlobalSecondaryIndexes and LocalSecondaryIndexes on a table "
-                        "using tablets require the number of racks in the cluster to be either 1 or 3"));
-                } else {
-                    co_return api_error::validation(fmt::format("Cannot create table '{}' with tablets: the configuration "
-                        "option 'rf_rack_valid_keyspaces' is enabled, which enforces that tables using tablets can only be created in clusters "
-                        "that have either 1 or 3 racks", table_name));
-                }
-            }
-        }
         try {
-            schema_mutations = service::prepare_new_keyspace_announcement(_proxy.local_db(), ksm, ts);
+            schema_mutations = service::prepare_new_keyspace_announcement(sp.local_db(), ksm, ts);
         } catch (exceptions::already_exists_exception&) {
-            if (_proxy.data_dictionary().has_schema(keyspace_name, table_name)) {
+            if (sp.data_dictionary().has_schema(keyspace_name, table_name)) {
                 co_return api_error::resource_in_use(fmt::format("Table {} already exists", table_name));
             }
         }
-        if (_proxy.data_dictionary().try_find_table(schema->id())) {
+        if (sp.data_dictionary().try_find_table(schema->id())) {
             // This should never happen, the ID is supposed to be unique
             co_return api_error::internal(format("Table with ID {} already exists", schema->id()));
         }
@@ -1922,9 +1815,9 @@ future<executor::request_return_type> executor::create_table_on_shard0(service::
         for (schema_builder& view_builder : view_builders) {
             schemas.push_back(view_builder.build());
         }
-        co_await service::prepare_new_column_families_announcement(schema_mutations, _proxy, *ksm, schemas, ts);
+        co_await service::prepare_new_column_families_announcement(schema_mutations, sp, *ksm, schemas, ts);
         if (ksm->uses_tablets()) {
-            co_await mark_view_schemas_as_built(schema_mutations, schemas, ts, _proxy);
+            co_await mark_view_schemas_as_built(schema_mutations, schemas, ts, sp);
         }
 
         // If a role is allowed to create a table, we must give it permissions to
@@ -1949,7 +1842,7 @@ future<executor::request_return_type> executor::create_table_on_shard0(service::
         }
         std::tie(schema_mutations, group0_guard) = co_await std::move(mc).extract();
         try {
-            co_await _mm.announce(std::move(schema_mutations), std::move(group0_guard), fmt::format("alternator-executor: create {} table", table_name));
+            co_await mm.announce(std::move(schema_mutations), std::move(group0_guard), fmt::format("alternator-executor: create {} table", table_name));
             break;
         }  catch (const service::group0_concurrent_modification& ex) {
             elogger.info("Failed to execute CreateTable {} due to concurrent schema modifications. {}.",
@@ -1961,9 +1854,9 @@ future<executor::request_return_type> executor::create_table_on_shard0(service::
         }
     }
 
-    co_await _mm.wait_for_schema_agreement(_proxy.local_db(), db::timeout_clock::now() + 10s, nullptr);
+    co_await mm.wait_for_schema_agreement(sp.local_db(), db::timeout_clock::now() + 10s, nullptr);
     rjson::value status = rjson::empty_object();
-    executor::supplement_table_info(request, *schema, _proxy);
+    executor::supplement_table_info(request, *schema, sp);
     rjson::add(status, "TableDescription", std::move(request));
     co_return rjson::print(std::move(status));
 }
@@ -1972,11 +1865,10 @@ future<executor::request_return_type> executor::create_table(client_state& clien
     _stats.api_operations.create_table++;
     elogger.trace("Creating table {}", request);
 
-    co_return co_await _mm.container().invoke_on(0, [&, tr = tracing::global_trace_state_ptr(trace_state), request = std::move(request), &e = this->container(), client_state_other_shard = client_state.move_to_other_shard(), enforce_authorization = bool(_enforce_authorization), warn_authorization = bool(_warn_authorization)]
+    co_return co_await _mm.container().invoke_on(0, [&, tr = tracing::global_trace_state_ptr(trace_state), request = std::move(request), &sp = _proxy.container(), &g = _gossiper.container(), &e = this->container(), client_state_other_shard = client_state.move_to_other_shard(), enforce_authorization = bool(_enforce_authorization), warn_authorization = bool(_warn_authorization)]
                                         (service::migration_manager& mm) mutable -> future<executor::request_return_type> {
         const db::tablets_mode_t::mode tablets_mode = _proxy.data_dictionary().get_config().tablets_mode_for_new_keyspaces(); // type cast
-        // `invoke_on` hopped us to shard 0, but `this` points to `executor` is from 'old' shard, we need to hop it too.
-        co_return co_await e.local().create_table_on_shard0(client_state_other_shard.get(), tr, std::move(request), enforce_authorization, warn_authorization, std::move(tablets_mode));
+        co_return co_await create_table_on_shard0(client_state_other_shard.get(), tr, std::move(request), sp.local(), mm, g.local(), enforce_authorization, warn_authorization, e.local()._stats, std::move(tablets_mode));
     });
 }
 
@@ -1995,8 +1887,8 @@ future<executor::request_return_type> executor::create_table(client_state& clien
         std::string def_type = type_to_string(def.type);
         for (auto it = attribute_definitions.Begin(); it != attribute_definitions.End(); ++it) {
             const rjson::value& attribute_info = *it;
-            if (rjson::to_string_view(attribute_info["AttributeName"]) == def.name_as_text()) {
-                std::string_view type = rjson::to_string_view(attribute_info["AttributeType"]);
+            if (attribute_info["AttributeName"].GetString() == def.name_as_text()) {
+                auto type = attribute_info["AttributeType"].GetString();
                 if (type != def_type) {
                     throw api_error::validation(fmt::format("AttributeDefinitions redefined {} to {} already a key attribute of type {} in this table", def.name_as_text(), type, def_type));
                 }
@@ -2127,13 +2019,6 @@ future<executor::request_return_type> executor::update_table(client_state& clien
                             co_return api_error::validation(fmt::format(
                                 "LSI {} already exists in table {}, can't use same name for GSI", index_name, table_name));
                         }
-                        try {
-                            locator::assert_rf_rack_valid_keyspace(keyspace_name, p.local().local_db().get_token_metadata_ptr(),
-                                    p.local().local_db().find_keyspace(keyspace_name).get_replication_strategy());
-                        } catch (const std::invalid_argument& ex) {
-                            co_return api_error::validation(fmt::format("GlobalSecondaryIndexes on a table "
-                                "using tablets require the number of racks in the cluster to be either 1 or 3"));
-                        }
 
                         elogger.trace("Adding GSI {}", index_name);
                         // FIXME: read and handle "Projection" parameter. This will
@@ -2436,7 +2321,7 @@ std::unordered_map<bytes, std::string> si_key_attributes(data_dictionary::table
 //   case, this function simply won't be called for this attribute.)
 //
 // This function checks if the given attribute update is an update to some
-// GSI's key, and if the value is unsuitable, an api_error::validation is
+// GSI's key, and if the value is unsuitable, a api_error::validation is
 // thrown. The checking here is similar to the checking done in
 // get_key_from_typed_value() for the base table's key columns.
 //
@@ -2477,7 +2362,7 @@ put_or_delete_item::put_or_delete_item(const rjson::value& item, schema_ptr sche
     _cells = std::vector<cell>();
     _cells->reserve(item.MemberCount());
     for (auto it = item.MemberBegin(); it != item.MemberEnd(); ++it) {
-        bytes column_name = to_bytes(rjson::to_string_view(it->name));
+        bytes column_name = to_bytes(it->name.GetString());
         validate_value(it->value, "PutItem");
         const column_definition* cdef = find_attribute(*schema, column_name);
         validate_attr_name_length("", column_name.size(), cdef && cdef->is_primary_key());
@@ -2838,12 +2723,14 @@ future<executor::request_return_type> rmw_operation::execute(service::storage_pr
         }
     } else if (_write_isolation != write_isolation::LWT_ALWAYS) {
         std::optional<mutation> m = apply(nullptr, api::new_timestamp(), cdc_opts);
-        throwing_assert(m); // !needs_read_before_write, so apply() did not check a condition
+        SCYLLA_ASSERT(m); // !needs_read_before_write, so apply() did not check a condition
         return proxy.mutate(utils::chunked_vector<mutation>{std::move(*m)}, db::consistency_level::LOCAL_QUORUM, executor::default_timeout(), trace_state, std::move(permit), db::allow_per_partition_rate_limit::yes, false, std::move(cdc_opts)).then([this, &wcu_total] () mutable {
             return rmw_operation_return(std::move(_return_attributes), _consumed_capacity, wcu_total);
         });
     }
-    throwing_assert(cas_shard);
+    if (!cas_shard) {
+        on_internal_error(elogger, "cas_shard is not set");
+    }
     // If we're still here, we need to do this write using LWT:
     global_stats.write_using_lwt++;
     per_table_stats.write_using_lwt++;
@@ -2896,10 +2783,10 @@ static void verify_all_are_used(const rjson::value* field,
         return;
     }
     for (auto it = field->MemberBegin(); it != field->MemberEnd(); ++it) {
-        if (!used.contains(rjson::to_string(it->name))) {
+        if (!used.contains(it->name.GetString())) {
             throw api_error::validation(
                 format("{} has spurious '{}', not used in {}",
-                    field_name, rjson::to_string_view(it->name), operation));
+                    field_name, it->name.GetString(), operation));
         }
     }
 }
@@ -3113,7 +3000,7 @@ future<executor::request_return_type> executor::delete_item(client_state& client
 }
 
 static schema_ptr get_table_from_batch_request(const service::storage_proxy& proxy, const rjson::value::ConstMemberIterator& batch_request) {
-    sstring table_name = rjson::to_sstring(batch_request->name); // JSON keys are always strings
+    sstring table_name = batch_request->name.GetString(); // JSON keys are always strings
     try {
         return proxy.data_dictionary().find_schema(sstring(executor::KEYSPACE_NAME_PREFIX) + table_name, table_name);
     } catch(data_dictionary::no_such_column_family&) {
@@ -3499,7 +3386,7 @@ static bool hierarchy_filter(rjson::value& val, const attribute_path_map_node<T>
         }
         rjson::value newv = rjson::empty_object();
         for (auto it = v.MemberBegin(); it != v.MemberEnd(); ++it) {
-            std::string attr = rjson::to_string(it->name);
+            std::string attr = it->name.GetString();
             auto x = members.find(attr);
             if (x != members.end()) {
                 if (x->second) {
@@ -3547,7 +3434,7 @@ static bool hierarchy_filter(rjson::value& val, const attribute_path_map_node<T>
     return true;
 }
 
-// Add a path to an attribute_path_map. Throws a validation error if the path
+// Add a path to a attribute_path_map. Throws a validation error if the path
 // "overlaps" with one already in the filter (one is a sub-path of the other)
 // or "conflicts" with it (both a member and index is requested).
 template<typename T>
@@ -3719,7 +3606,7 @@ static std::optional<attrs_to_get> calculate_attrs_to_get(const rjson::value& re
         const rjson::value& attributes_to_get = req["AttributesToGet"];
         attrs_to_get ret;
         for (auto it = attributes_to_get.Begin(); it != attributes_to_get.End(); ++it) {
-            attribute_path_map_add("AttributesToGet", ret, rjson::to_string(*it));
+            attribute_path_map_add("AttributesToGet", ret, it->GetString());
             validate_attr_name_length("AttributesToGet", it->GetStringLength(), false);
         }
         if (ret.empty()) {
@@ -4385,12 +4272,12 @@ inline void update_item_operation::apply_attribute_updates(const std::unique_ptr
         attribute_collector& modified_attrs, bool& any_updates, bool& any_deletes) const {
     for (auto it = _attribute_updates->MemberBegin(); it != _attribute_updates->MemberEnd(); ++it) {
         // Note that it.key() is the name of the column, *it is the operation
-        bytes column_name = to_bytes(rjson::to_string_view(it->name));
+        bytes column_name = to_bytes(it->name.GetString());
         const column_definition* cdef = _schema->get_column_definition(column_name);
         if (cdef && cdef->is_primary_key()) {
-            throw api_error::validation(format("UpdateItem cannot update key column {}", rjson::to_string_view(it->name)));
+            throw api_error::validation(format("UpdateItem cannot update key column {}", it->name.GetString()));
         }
-        std::string action = rjson::to_string((it->value)["Action"]);
+        std::string action = (it->value)["Action"].GetString();
         if (action == "DELETE") {
             // The DELETE operation can do two unrelated tasks. Without a
             // "Value" option, it is used to delete an attribute. With a
@@ -5412,7 +5299,7 @@ static future<executor::request_return_type> do_query(service::storage_proxy& pr
 }
 
 static dht::token token_for_segment(int segment, int total_segments) {
-    throwing_assert(total_segments > 1 && segment >= 0 && segment < total_segments);
+    SCYLLA_ASSERT(total_segments > 1 && segment >= 0 && segment < total_segments);
     uint64_t delta = std::numeric_limits<uint64_t>::max() / total_segments;
     return dht::token::from_int64(std::numeric_limits<int64_t>::min() + delta * segment);
 }
@@ -5587,7 +5474,7 @@ calculate_bounds_conditions(schema_ptr schema, const rjson::value& conditions) {
     std::vector<query::clustering_range> ck_bounds;
 
     for (auto it = conditions.MemberBegin(); it != conditions.MemberEnd(); ++it) {
-        sstring key = rjson::to_sstring(it->name);
+        std::string key = it->name.GetString();
         const rjson::value& condition = it->value;
 
         const rjson::value& comp_definition = rjson::get(condition, "ComparisonOperator");
@@ -5595,13 +5482,13 @@ calculate_bounds_conditions(schema_ptr schema, const rjson::value& conditions) {
 
         const column_definition& pk_cdef = schema->partition_key_columns().front();
         const column_definition* ck_cdef = schema->clustering_key_size() > 0 ? &schema->clustering_key_columns().front() : nullptr;
-        if (key == pk_cdef.name_as_text()) {
+        if (sstring(key) == pk_cdef.name_as_text()) {
             if (!partition_ranges.empty()) {
                 throw api_error::validation("Currently only a single restriction per key is allowed");
             }
             partition_ranges.push_back(calculate_pk_bound(schema, pk_cdef, comp_definition, attr_list));
         }
-        if (ck_cdef && key == ck_cdef->name_as_text()) {
+        if (ck_cdef && sstring(key) == ck_cdef->name_as_text()) {
             if (!ck_bounds.empty()) {
                 throw api_error::validation("Currently only a single restriction per key is allowed");
             }
@@ -6000,14 +5887,9 @@ future<executor::request_return_type> executor::list_tables(client_state& client
     _stats.api_operations.list_tables++;
     elogger.trace("Listing tables {}", request);
 
-    co_await utils::get_local_injector().inject("alternator_list_tables", [] (auto& handler) -> future<> {
-        handler.set("waiting", true);
-        co_await handler.wait_for_message(std::chrono::steady_clock::now() + std::chrono::minutes{5});
-    });
-
     rjson::value* exclusive_start_json = rjson::find(request, "ExclusiveStartTableName");
     rjson::value* limit_json = rjson::find(request, "Limit");
-    std::string exclusive_start = exclusive_start_json ? rjson::to_string(*exclusive_start_json) : "";
+    std::string exclusive_start = exclusive_start_json ? exclusive_start_json->GetString() : "";
     int limit = limit_json ? limit_json->GetInt() : 100;
     if (limit < 1 || limit > 100) {
         co_return api_error::validation("Limit must be greater than 0 and no greater than 100");
@@ -6196,10 +6078,9 @@ future<> executor::start() {
 }
 
 future<> executor::stop() {
-    co_await _describe_table_info_manager->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()};
-    co_await _parsed_expression_cache->stop();
+    return _parsed_expression_cache->stop();
 }
 
 } // namespace alternator

alternator/executor.hh

@@ -17,13 +17,11 @@
 #include "service/client_state.hh"
 #include "service_permit.hh"
 #include "db/timeout_clock.hh"
-#include "db/config.hh"
 
 #include "alternator/error.hh"
 #include "stats.hh"
 #include "utils/rjson.hh"
 #include "utils/updateable_value.hh"
-#include "utils/simple_value_with_expiry.hh"
 
 #include "tracing/trace_state.hh"
 
@@ -43,7 +41,6 @@ namespace cql3::selection {
 namespace service {
     class storage_proxy;
     class cas_shard;
-    class storage_service;
 }
 
 namespace cdc {
@@ -60,7 +57,6 @@ class schema_builder;
 
 namespace alternator {
 
-enum class table_status;
 class rmw_operation;
 class put_or_delete_item;
 
@@ -140,7 +136,6 @@ class expression_cache;
 
 class executor : public peering_sharded_service<executor> {
     gms::gossiper& _gossiper;
-    service::storage_service& _ss;
     service::storage_proxy& _proxy;
     service::migration_manager& _mm;
     db::system_distributed_keyspace& _sdks;
@@ -153,11 +148,6 @@ class executor : public peering_sharded_service<executor> {
 
     std::unique_ptr<parsed::expression_cache> _parsed_expression_cache;
 
-    struct describe_table_info_manager;
-    std::unique_ptr<describe_table_info_manager> _describe_table_info_manager;
-
-    future<> cache_newly_calculated_size_on_all_shards(schema_ptr schema, std::uint64_t size_in_bytes, std::chrono::nanoseconds ttl);
-    future<> fill_table_size(rjson::value &table_description, schema_ptr schema, bool deleting);
 public:
     using client_state = service::client_state;
     // request_return_type is the return type of the executor methods, which
@@ -183,7 +173,6 @@ public:
 
     executor(gms::gossiper& gossiper,
              service::storage_proxy& proxy,
-             service::storage_service& ss,
              service::migration_manager& mm,
              db::system_distributed_keyspace& sdks,
              cdc::metadata& cdc_metadata,
@@ -231,8 +220,6 @@ 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);
-    future<rjson::value> fill_table_description(schema_ptr schema, table_status tbl_status, service::client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit);
-    future<executor::request_return_type> create_table_on_shard0(service::client_state&& client_state, tracing::trace_state_ptr trace_state, rjson::value request, bool enforce_authorization, bool warn_authorization, const db::tablets_mode_t::mode tablets_mode);
 
     future<> do_batch_write(
         std::vector<std::pair<schema_ptr, put_or_delete_item>> mutation_builders,

alternator/expressions_types.hh

@@ -50,7 +50,7 @@ public:
         _operators.emplace_back(i);
         check_depth_limit();
     }
-    void add_dot(std::string name) {
+    void add_dot(std::string(name)) {
         _operators.emplace_back(std::move(name));
         check_depth_limit();
     }
@@ -85,7 +85,7 @@ struct constant {
     }
 };
 
-// "value" is a value used in the right hand side of an assignment
+// "value" is is a value used in the right hand side of an assignment
 // expression, "SET a = ...". It can be a constant (a reference to a value
 // included in the request, e.g., ":val"), a path to an attribute from the
 // existing item (e.g., "a.b[3].c"), or a function of other such values.
@@ -205,7 +205,7 @@ public:
 // The supported primitive conditions are:
 // 1. Binary operators - v1 OP v2, where OP is =, <>, <, <=, >, or >= and
 //    v1 and v2 are values - from the item (an attribute path), the query
-//    (a ":val" reference), or a function of the above (only the size()
+//    (a ":val" reference), or a function of the the above (only the size()
 //    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, ... )

alternator/http_compression.cc

@@ -1,301 +0,0 @@
-/*
- * Copyright 2025-present ScyllaDB
- */
-
-/*
- * SPDX-License-Identifier: LicenseRef-ScyllaDB-Source-Available-1.0
- */
-
-#include "alternator/http_compression.hh"
-#include "alternator/server.hh"
-#include <seastar/coroutine/maybe_yield.hh>
-#include <zlib.h>
-
-static logging::logger slogger("alternator-http-compression");
-
-namespace alternator {
-
-
-static constexpr size_t compressed_buffer_size = 1024;
-class zlib_compressor {
-    z_stream _zs;
-    temporary_buffer<char> _output_buf;
-    noncopyable_function<future<>(temporary_buffer<char>&&)> _write_func;
-public:
-    zlib_compressor(bool gzip, int compression_level, noncopyable_function<future<>(temporary_buffer<char>&&)> write_func)
-     : _write_func(std::move(write_func)) {
-        memset(&_zs, 0, sizeof(_zs));
-        if (deflateInit2(&_zs, std::clamp(compression_level, Z_NO_COMPRESSION, Z_BEST_COMPRESSION), Z_DEFLATED,
-                (gzip ? 16 : 0) + MAX_WBITS, 8, Z_DEFAULT_STRATEGY) != Z_OK) {
-            // Should only happen if memory allocation fails
-            throw std::bad_alloc();
-        }
-    }
-    ~zlib_compressor() {
-        deflateEnd(&_zs);
-    }
-    future<> close() {
-        return compress(nullptr, 0, true);
-    }
-
-    future<> compress(const char* buf, size_t len, bool is_last_chunk = false) {
-        _zs.next_in = reinterpret_cast<unsigned char*>(const_cast<char*>(buf));
-        _zs.avail_in = (uInt) len;
-        int mode = is_last_chunk ? Z_FINISH : Z_NO_FLUSH;
-        while(_zs.avail_in > 0 || is_last_chunk) {
-            co_await coroutine::maybe_yield();
-            if (_output_buf.empty()) {
-                if (is_last_chunk) {
-                    uint32_t max_buffer_size = 0;
-                    deflatePending(&_zs, &max_buffer_size, nullptr);
-                    max_buffer_size += deflateBound(&_zs, _zs.avail_in) + 1;
-                    _output_buf = temporary_buffer<char>(std::min(compressed_buffer_size, (size_t) max_buffer_size));
-                } else {
-                    _output_buf = temporary_buffer<char>(compressed_buffer_size);
-                }
-                _zs.next_out = reinterpret_cast<unsigned char*>(_output_buf.get_write());
-                _zs.avail_out = compressed_buffer_size;
-            }
-            int e = deflate(&_zs, mode);
-            if (e < Z_OK) {
-                throw api_error::internal("Error during compression of response body");
-            }
-            if (e == Z_STREAM_END || _zs.avail_out < compressed_buffer_size / 4) {
-                _output_buf.trim(compressed_buffer_size - _zs.avail_out);
-                co_await _write_func(std::move(_output_buf));
-                if (e == Z_STREAM_END) {
-                    break;
-                }
-            }
-        }
-    }
-};
-
-// Helper string_view functions for parsing Accept-Encoding header
-struct case_insensitive_cmp_sv {
-    bool operator()(std::string_view s1, std::string_view s2) const {
-        return std::equal(s1.begin(), s1.end(), s2.begin(), s2.end(),
-            [](char a, char b) { return ::tolower(a) == ::tolower(b); });
-    }
-};
-static inline std::string_view trim_left(std::string_view sv) {
-    while (!sv.empty() && std::isspace(static_cast<unsigned char>(sv.front())))
-        sv.remove_prefix(1);
-    return sv;
-}
-static inline std::string_view trim_right(std::string_view sv) {
-    while (!sv.empty() && std::isspace(static_cast<unsigned char>(sv.back())))
-        sv.remove_suffix(1);
-    return sv;
-}
-static inline std::string_view trim(std::string_view sv) {
-    return trim_left(trim_right(sv));
-}
-
-inline std::vector<std::string_view> split(std::string_view text, char separator) {
-    std::vector<std::string_view> tokens;
-    if (text == "") {
-        return tokens;
-    }
-
-    while (true) {
-        auto pos = text.find_first_of(separator);
-        if (pos != std::string_view::npos) {
-            tokens.emplace_back(text.data(), pos);
-            text.remove_prefix(pos + 1);
-        } else {
-            tokens.emplace_back(text);
-            break;
-        }
-    }
-    return tokens;
-}
-
-constexpr response_compressor::compression_type response_compressor::get_compression_type(std::string_view encoding) {
-    for (size_t i = 0; i < static_cast<size_t>(compression_type::count); ++i) {
-        if (case_insensitive_cmp_sv{}(encoding, compression_names[i])) {
-            return static_cast<compression_type>(i);
-        }
-    }
-    return compression_type::unknown;
-}
-
-response_compressor::compression_type response_compressor::find_compression(std::string_view accept_encoding, size_t response_size) {
-    std::optional<float> ct_q[static_cast<size_t>(compression_type::count)];
-    ct_q[static_cast<size_t>(compression_type::none)] = std::numeric_limits<float>::min(); // enabled, but lowest priority
-    compression_type selected_ct = compression_type::none;
-
-    std::vector<std::string_view> entries = split(accept_encoding, ',');
-    for (auto& e : entries) {
-        std::vector<std::string_view> params = split(e, ';');
-        if (params.size() == 0) {
-            continue;
-        }
-        compression_type ct = get_compression_type(trim(params[0]));
-        if (ct == compression_type::unknown) {
-            continue; // ignore unknown encoding types
-        }
-        if (ct_q[static_cast<size_t>(ct)].has_value() && ct_q[static_cast<size_t>(ct)] != 0.0f) {
-            continue; // already processed this encoding
-        }
-        if (response_size < _threshold[static_cast<size_t>(ct)]) {
-            continue; // below threshold treat as unknown
-        }
-        for (size_t i = 1; i < params.size(); ++i) { // find "q=" parameter
-            auto pos = params[i].find("q=");
-            if (pos == std::string_view::npos) {
-                continue;
-            }
-            std::string_view param = params[i].substr(pos + 2);
-            param = trim(param);
-            // parse quality value
-            float q_value = 1.0f;
-            auto [ptr, ec] = std::from_chars(param.data(), param.data() + param.size(), q_value);
-            if (ec != std::errc() || ptr != param.data() + param.size()) {
-                continue;
-            }
-            if (q_value < 0.0) {
-                q_value = 0.0;
-            } else if (q_value > 1.0) {
-                q_value = 1.0;
-            }
-            ct_q[static_cast<size_t>(ct)] = q_value;
-            break; // we parsed quality value
-        }
-        if (!ct_q[static_cast<size_t>(ct)].has_value()) {
-            ct_q[static_cast<size_t>(ct)] = 1.0f; // default quality value
-        }
-        // keep the highest encoding (in the order, unless 'any')
-        if (selected_ct == compression_type::any) {
-            if (ct_q[static_cast<size_t>(ct)] >= ct_q[static_cast<size_t>(selected_ct)]) {
-                selected_ct = ct;
-            }
-        } else {
-            if (ct_q[static_cast<size_t>(ct)] > ct_q[static_cast<size_t>(selected_ct)]) {
-                selected_ct = ct;
-            }
-        }
-    }
-    if (selected_ct == compression_type::any) {
-        // select any not mentioned or highest quality
-        selected_ct = compression_type::none;
-        for (size_t i = 0; i < static_cast<size_t>(compression_type::compressions_count); ++i) {
-            if (!ct_q[i].has_value()) {
-                return static_cast<compression_type>(i);
-            }
-            if (ct_q[i] > ct_q[static_cast<size_t>(selected_ct)]) {
-                selected_ct = static_cast<compression_type>(i);
-            }
-        }
-    }
-    return selected_ct;
-}
-
-static future<chunked_content> compress(response_compressor::compression_type ct, const db::config& cfg, std::string str) {
-    chunked_content compressed;
-    auto write = [&compressed](temporary_buffer<char>&& buf) -> future<> {
-        compressed.push_back(std::move(buf));
-        return make_ready_future<>();
-    };
-    zlib_compressor compressor(ct != response_compressor::compression_type::deflate,
-        cfg.alternator_response_gzip_compression_level(), std::move(write));
-    co_await compressor.compress(str.data(), str.size(), true);
-    co_return compressed;
-}
-
-static sstring flatten(chunked_content&& cc) {
-    size_t total_size = 0;
-    for (const auto& chunk : cc) {
-        total_size += chunk.size();
-    }
-    sstring result = sstring{ sstring::initialized_later{}, total_size };
-    size_t offset = 0;
-    for (const auto& chunk : cc) {
-        std::copy(chunk.begin(), chunk.end(), result.begin() + offset);
-        offset += chunk.size();
-    }
-    return result;
-}
-
-future<std::unique_ptr<http::reply>> response_compressor::generate_reply(std::unique_ptr<http::reply> rep, sstring accept_encoding, const char* content_type, std::string&& response_body) {
-    response_compressor::compression_type ct = find_compression(accept_encoding, response_body.size());
-    if (ct != response_compressor::compression_type::none) {
-        rep->add_header("Content-Encoding", get_encoding_name(ct));
-        rep->set_content_type(content_type);
-        return compress(ct, cfg, std::move(response_body)).then([rep = std::move(rep)] (chunked_content compressed) mutable {
-            rep->_content = flatten(std::move(compressed));
-            return make_ready_future<std::unique_ptr<http::reply>>(std::move(rep));
-        });
-    } else {
-        // Note that despite the move, there is a copy here -
-        // as str is std::string and rep->_content is sstring.
-        rep->_content = std::move(response_body);
-        rep->set_content_type(content_type);
-    }
-    return make_ready_future<std::unique_ptr<http::reply>>(std::move(rep));
-}
-
-template<typename Compressor>
-class compressed_data_sink_impl : public data_sink_impl {
-    output_stream<char> _out;
-    Compressor _compressor;
-public:
-    template<typename... Args>
-    compressed_data_sink_impl(output_stream<char>&& out, Args&&... args)
-     : _out(std::move(out)), _compressor(std::forward<Args>(args)..., [this](temporary_buffer<char>&& buf) {
-        return _out.write(std::move(buf));
-    }) { }
-
-    future<> put(std::span<temporary_buffer<char>> data) override {
-        return data_sink_impl::fallback_put(data, [this] (temporary_buffer<char>&& buf) {
-            return do_put(std::move(buf));
-        });
-    }
-
-private:
-    future<> do_put(temporary_buffer<char> buf) {
-        co_return co_await _compressor.compress(buf.get(), buf.size());
-
-    }
-    future<> close() override {
-        return _compressor.close().then([this] {
-            return _out.close();
-        });
-    }
-};
-
-executor::body_writer compress(response_compressor::compression_type ct, const db::config& cfg, executor::body_writer&& bw) {
-    return [bw = std::move(bw), ct, level = cfg.alternator_response_gzip_compression_level()](output_stream<char>&& out) mutable -> future<> {
-        output_stream_options opts;
-        opts.trim_to_size = true;
-        std::unique_ptr<data_sink_impl> data_sink_impl;
-        switch (ct) {
-            case response_compressor::compression_type::gzip:
-                data_sink_impl = std::make_unique<compressed_data_sink_impl<zlib_compressor>>(std::move(out), true, level);
-                break;
-            case response_compressor::compression_type::deflate:
-                data_sink_impl = std::make_unique<compressed_data_sink_impl<zlib_compressor>>(std::move(out), false, level);
-                break;
-            case response_compressor::compression_type::none:
-            case response_compressor::compression_type::any:
-            case response_compressor::compression_type::unknown:
-                on_internal_error(slogger,"Compression not selected");
-            default:
-                on_internal_error(slogger, "Unsupported compression type for data sink");
-        }
-        return bw(output_stream<char>(data_sink(std::move(data_sink_impl)), compressed_buffer_size, opts));
-    };
-}
-
-future<std::unique_ptr<http::reply>> response_compressor::generate_reply(std::unique_ptr<http::reply> rep, sstring accept_encoding, const char* content_type, executor::body_writer&& body_writer) {
-    response_compressor::compression_type ct = find_compression(accept_encoding, std::numeric_limits<size_t>::max());
-    if (ct != response_compressor::compression_type::none) {
-        rep->add_header("Content-Encoding", get_encoding_name(ct));
-        rep->write_body(content_type, compress(ct, cfg, std::move(body_writer)));
-    } else {
-        rep->write_body(content_type, std::move(body_writer));
-    }
-    return make_ready_future<std::unique_ptr<http::reply>>(std::move(rep));
-}
-
-} // namespace alternator

alternator/http_compression.hh

@@ -1,91 +0,0 @@
-/*
- * Copyright 2025-present ScyllaDB
- */
-
-/*
- * SPDX-License-Identifier: LicenseRef-ScyllaDB-Source-Available-1.0
- */
-
-#pragma once
-
-#include "alternator/executor.hh"
-#include <seastar/http/httpd.hh>
-#include "db/config.hh"
-
-namespace alternator {
-
-class response_compressor {
-public:
-    enum class compression_type {
-        gzip,
-        deflate,
-        compressions_count,
-        any = compressions_count,
-        none,
-        count,
-        unknown = count
-    };
-    static constexpr std::string_view compression_names[] = {
-        "gzip",
-        "deflate",
-        "*",
-        "identity"
-    };
-
-    static sstring get_encoding_name(compression_type ct) {
-        return sstring(compression_names[static_cast<size_t>(ct)]);
-    }
-    static constexpr compression_type get_compression_type(std::string_view encoding);
-
-    sstring get_accepted_encoding(const http::request& req) {
-        if (get_threshold() == 0) {
-            return "";
-        }
-        return req.get_header("Accept-Encoding");
-    }
-    compression_type find_compression(std::string_view accept_encoding, size_t response_size);
-
-    response_compressor(const db::config& cfg)
-        : cfg(cfg)
-        ,_gzip_level_observer(
-            cfg.alternator_response_gzip_compression_level.observe([this](int v) {
-                    update_threshold();
-                }))
-        ,_gzip_threshold_observer(
-            cfg.alternator_response_compression_threshold_in_bytes.observe([this](uint32_t v) {
-                    update_threshold();
-                }))
-    {
-        update_threshold();
-    }
-    response_compressor(const response_compressor& rhs) : response_compressor(rhs.cfg) {}
-
-private:
-    const db::config& cfg;
-    utils::observable<int>::observer _gzip_level_observer;
-    utils::observable<uint32_t>::observer _gzip_threshold_observer;
-    uint32_t _threshold[static_cast<size_t>(compression_type::count)];
-
-    size_t get_threshold() { return _threshold[static_cast<size_t>(compression_type::any)]; }
-    void update_threshold() {
-        _threshold[static_cast<size_t>(compression_type::none)] = std::numeric_limits<uint32_t>::max();
-        _threshold[static_cast<size_t>(compression_type::any)] = std::numeric_limits<uint32_t>::max();
-        uint32_t gzip = cfg.alternator_response_gzip_compression_level() <= 0 ? std::numeric_limits<uint32_t>::max()
-            : cfg.alternator_response_compression_threshold_in_bytes();
-        _threshold[static_cast<size_t>(compression_type::gzip)] = gzip;
-        _threshold[static_cast<size_t>(compression_type::deflate)] = gzip;
-        for (size_t i = 0; i < static_cast<size_t>(compression_type::compressions_count); ++i) {
-            if (_threshold[i] < _threshold[static_cast<size_t>(compression_type::any)]) {
-                _threshold[static_cast<size_t>(compression_type::any)] = _threshold[i];
-            }
-        }
-    }
-
-public:
-    future<std::unique_ptr<http::reply>> generate_reply(std::unique_ptr<http::reply> rep,
-         sstring accept_encoding, const char* content_type, std::string&& response_body);
-    future<std::unique_ptr<http::reply>> generate_reply(std::unique_ptr<http::reply> rep,
-         sstring accept_encoding, const char* content_type, executor::body_writer&& body_writer);
-};
-
-}

alternator/serialization.cc

@@ -496,7 +496,7 @@ const std::pair<std::string, const rjson::value*> unwrap_set(const rjson::value&
         return {"", nullptr};
     }
     auto it = v.MemberBegin();
-    const std::string it_key = rjson::to_string(it->name);
+    const std::string it_key = it->name.GetString();
     if (it_key != "SS" && it_key != "BS" && it_key != "NS") {
         return {std::move(it_key), nullptr};
     }

alternator/serialization.hh

@@ -55,7 +55,7 @@ partition_key pk_from_json(const rjson::value& item, schema_ptr schema);
 clustering_key ck_from_json(const rjson::value& item, schema_ptr schema);
 position_in_partition pos_from_json(const rjson::value& item, schema_ptr schema);
 
-// If v encodes a number (i.e., it is a {"N": [...]}), returns an object representing it.  Otherwise,
+// If v encodes a number (i.e., it is a {"N": [...]}, returns an object representing it.  Otherwise,
 // raises ValidationException with diagnostic.
 big_decimal unwrap_number(const rjson::value& v, std::string_view diagnostic);
 

alternator/server.cc

@@ -34,7 +34,6 @@
 #include "client_data.hh"
 #include "utils/updateable_value.hh"
 #include <zlib.h>
-#include "alternator/http_compression.hh"
 
 static logging::logger slogger("alternator-server");
 
@@ -112,12 +111,9 @@ class api_handler : public handler_base {
     // 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,
-                const db::config& config) : _response_compressor(config), _f_handle(
+    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) {
-         sstring accept_encoding = _response_compressor.get_accepted_encoding(*req);
-         return seastar::futurize_invoke(_handle, std::move(req)).then_wrapped(
-            [this, rep = std::move(rep), accept_encoding=std::move(accept_encoding)](future<executor::request_return_type> resf) mutable {
+         return seastar::futurize_invoke(_handle, std::move(req)).then_wrapped([this, rep = std::move(rep)](future<executor::request_return_type> resf) mutable {
              if (resf.failed()) {
                  // Exceptions of type api_error are wrapped as JSON and
                  // returned to the client as expected. Other types of
@@ -137,20 +133,22 @@ public:
                  return make_ready_future<std::unique_ptr<reply>>(std::move(rep));
              }
              auto res = resf.get();
-             return std::visit(overloaded_functor {
+             std::visit(overloaded_functor {
                 [&] (std::string&& str) {
-                    return _response_compressor.generate_reply(std::move(rep), std::move(accept_encoding),
-                                                               REPLY_CONTENT_TYPE, std::move(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) {
-                    return _response_compressor.generate_reply(std::move(rep), std::move(accept_encoding),
-                                                               REPLY_CONTENT_TYPE, std::move(body_writer));
+                    rep->write_body(REPLY_CONTENT_TYPE, std::move(body_writer));
                 },
                 [&] (const api_error& err) {
                     generate_error_reply(*rep, err);
-                    return make_ready_future<std::unique_ptr<reply>>(std::move(rep));
                 }
              }, std::move(res));
+
+             return make_ready_future<std::unique_ptr<reply>>(std::move(rep));
          });
     }) { }
 
@@ -179,7 +177,6 @@ protected:
         slogger.trace("api_handler error case: {}", rep._content);
     }
 
-    response_compressor _response_compressor;
     future_handler_function _f_handle;
 };
 
@@ -374,40 +371,13 @@ future<std::string> server::verify_signature(const request& req, const chunked_c
     for (const auto& header : signed_headers) {
         signed_headers_map.emplace(header, std::string_view());
     }
-    std::vector<std::string> modified_values;
     for (auto& header : req._headers) {
         std::string header_str;
         header_str.resize(header.first.size());
         std::transform(header.first.begin(), header.first.end(), header_str.begin(), ::tolower);
         auto it = signed_headers_map.find(header_str);
         if (it != signed_headers_map.end()) {
-            // replace multiple spaces in the header value header.second with
-            // a single space, as required by AWS SigV4 header canonization.
-            // If we modify the value, we need to save it in modified_values
-            // to keep it alive.
-            std::string value;
-            value.reserve(header.second.size());
-            bool prev_space = false;
-            bool modified = false;
-            for (char ch : header.second) {
-                if (ch == ' ') {
-                    if (!prev_space) {
-                        value += ch;
-                        prev_space = true;
-                    } else {
-                        modified = true; // skip a space
-                    }
-                } else {
-                    value += ch;
-                    prev_space = false;
-                }
-            }
-            if (modified) {
-                modified_values.emplace_back(std::move(value));
-                it->second = std::string_view(modified_values.back());
-            } else {
-                it->second = std::string_view(header.second);
-            }
+            it->second = std::string_view(header.second);
         }
     }
 
@@ -420,7 +390,6 @@ future<std::string> server::verify_signature(const request& req, const chunked_c
                                                     datestamp = std::move(datestamp),
                                                     signed_headers_str = std::move(signed_headers_str),
                                                     signed_headers_map = std::move(signed_headers_map),
-                                                    modified_values = std::move(modified_values),
                                                     region = std::move(region),
                                                     service = std::move(service),
                                                     user_signature = std::move(user_signature)] (future<key_cache::value_ptr> key_ptr_fut) {
@@ -591,11 +560,11 @@ read_entire_stream(input_stream<char>& inp, size_t length_limit) {
 class safe_gzip_zstream {
     z_stream _zs;
 public:
-    // If gzip is true, decode a gzip header (for "Content-Encoding: gzip").
-    // Otherwise, a zlib header (for "Content-Encoding: deflate").
-    safe_gzip_zstream(bool gzip = true) {
+    safe_gzip_zstream() {
         memset(&_zs, 0, sizeof(_zs));
-        if (inflateInit2(&_zs, gzip ? 16 + MAX_WBITS : MAX_WBITS) != Z_OK) {
+        // The strange 16 + WMAX_BITS tells zlib to expect and decode
+        // a gzip header, not a zlib header.
+        if (inflateInit2(&_zs, 16 + MAX_WBITS) != Z_OK) {
             // Should only happen if memory allocation fails
             throw std::bad_alloc();
         }
@@ -614,21 +583,19 @@ public:
     }
 };
 
-// ungzip() takes a chunked_content of a compressed request body, and returns
-// the uncompressed content as a chunked_content. If gzip is true, we expect
-// gzip header (for "Content-Encoding: gzip"), if gzip is false, we expect a
-// zlib header (for "Content-Encoding: deflate").
+// ungzip() takes a chunked_content with a gzip-compressed request body,
+// uncompresses it, and returns the uncompressed content as a chunked_content.
 // If the uncompressed content exceeds length_limit, an error is thrown.
 static future<chunked_content>
-ungzip(chunked_content&& compressed_body, size_t length_limit, bool gzip = true) {
+ungzip(chunked_content&& compressed_body, size_t length_limit) {
     chunked_content ret;
     // output_buf can be any size - when uncompressing input_buf, it doesn't
     // need to fit in a single output_buf, we'll use multiple output_buf for
     // a single input_buf if needed.
     constexpr size_t OUTPUT_BUF_SIZE = 4096;
     temporary_buffer<char> output_buf;
-    safe_gzip_zstream strm(gzip);
-    bool complete_stream = false; // empty input is not a valid gzip/deflate
+    safe_gzip_zstream strm;
+    bool complete_stream = false; // empty input is not a valid gzip
     size_t total_out_bytes = 0;
     for (const temporary_buffer<char>& input_buf : compressed_body) {
         if (input_buf.empty()) {
@@ -710,7 +677,7 @@ future<executor::request_return_type> server::handle_api_request(std::unique_ptr
         ++_executor._stats.requests_blocked_memory;
     }
     auto units = co_await std::move(units_fut);
-    throwing_assert(req->content_stream);
+    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
@@ -731,8 +698,6 @@ future<executor::request_return_type> server::handle_api_request(std::unique_ptr
     sstring content_encoding = req->get_header("Content-Encoding");
     if (content_encoding == "gzip") {
         content = co_await ungzip(std::move(content), request_content_length_limit);
-    } else if (content_encoding == "deflate") {
-        content = co_await ungzip(std::move(content), request_content_length_limit, false);
     } else if (!content_encoding.empty()) {
         // DynamoDB returns a 500 error for unsupported Content-Encoding.
         // I'm not sure if this is the best error code, but let's do it too.
@@ -743,12 +708,8 @@ future<executor::request_return_type> server::handle_api_request(std::unique_ptr
     // 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 user_agent_header = co_await _connection_options_keys_and_values.get_or_load(req->get_header("User-Agent"), [] (const client_options_cache_key_type&) {
-        return make_ready_future<options_cache_value_type>(options_cache_value_type{});
-    });
-
     auto system_clients_entry = _ongoing_requests.emplace(
-        req->get_client_address(), std::move(user_agent_header),
+        req->get_client_address(), req->get_header("User-Agent"),
         username, current_scheduling_group(),
         req->get_protocol_name() == "https");
 
@@ -793,7 +754,7 @@ future<executor::request_return_type> server::handle_api_request(std::unique_ptr
 void server::set_routes(routes& r) {
     api_handler* req_handler = new api_handler([this] (std::unique_ptr<request> req) mutable {
         return handle_api_request(std::move(req));
-    }, _proxy.data_dictionary().get_config());
+    });
 
     r.put(operation_type::POST, "/", req_handler);
     r.put(operation_type::GET, "/", new health_handler(_pending_requests));
@@ -904,9 +865,7 @@ 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<uint16_t> port_proxy_protocol, std::optional<uint16_t> https_port_proxy_protocol,
-        std::optional<tls::credentials_builder> creds,
+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<bool> warn_authorization, utils::updateable_value<uint64_t> max_users_query_size_in_trace_output,
         semaphore* memory_limiter, utils::updateable_value<uint32_t> max_concurrent_requests) {
     _memory_limiter = memory_limiter;
@@ -914,28 +873,20 @@ future<> server::init(net::inet_address addr, std::optional<uint16_t> port, std:
     _warn_authorization = std::move(warn_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 && !port_proxy_protocol && !https_port_proxy_protocol) {
+    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"));
     }
-    return seastar::async([this, addr, port, https_port, port_proxy_protocol, https_port_proxy_protocol, creds] {
+    return seastar::async([this, addr, port, https_port, creds] {
         _executor.start().get();
 
-        if (port || port_proxy_protocol) {
+        if (port) {
             set_routes(_http_server._routes);
             _http_server.set_content_streaming(true);
-            if (port) {
-                _http_server.listen(socket_address{addr, *port}).get();
-            }
-            if (port_proxy_protocol) {
-                listen_options lo;
-                lo.reuse_address = true;
-                lo.proxy_protocol = true;
-                _http_server.listen(socket_address{addr, *port_proxy_protocol}, lo).get();
-            }
+            _http_server.listen(socket_address{addr, *port}).get();
             _enabled_servers.push_back(std::ref(_http_server));
         }
-        if (https_port || https_port_proxy_protocol) {
+        if (https_port) {
             set_routes(_https_server._routes);
             _https_server.set_content_streaming(true);
 
@@ -955,15 +906,7 @@ future<> server::init(net::inet_address addr, std::optional<uint16_t> port, std:
             } else {
                 _credentials = creds->build_server_credentials();
             }
-            if (https_port) {
-                _https_server.listen(socket_address{addr, *https_port}, _credentials).get();
-            }
-            if (https_port_proxy_protocol) {
-                listen_options lo;
-                lo.reuse_address = true;
-                lo.proxy_protocol = true;
-                _https_server.listen(socket_address{addr, *https_port_proxy_protocol}, lo, _credentials).get();
-            }
+            _https_server.listen(socket_address{addr, *https_port}, _credentials).get();
             _enabled_servers.push_back(std::ref(_https_server));
         }
     });
@@ -1036,15 +979,16 @@ client_data server::ongoing_request::make_client_data() const {
     // 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 for Alternator.
-    // Note: CQL sets ssl_protocol and ssl_cipher_suite via generic_server::connection base class.
+    // 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<foreign_ptr<std::unique_ptr<client_data>>>> server::get_client_data() {
-    utils::chunked_vector<foreign_ptr<std::unique_ptr<client_data>>> ret;
+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(make_foreign(std::make_unique<client_data>(r.make_client_data())));
+        ret.emplace_back(r.make_client_data());
     });
     co_return ret;
 }

alternator/server.hh

@@ -55,7 +55,6 @@ class server : public peering_sharded_service<server> {
     // though it isn't really relevant for Alternator which defines its own
     // timeouts separately. We can create this object only once.
     updateable_timeout_config _timeout_config;
-    client_options_cache_type _connection_options_keys_and_values;
 
     alternator_callbacks_map _callbacks;
 
@@ -89,7 +88,7 @@ class server : public peering_sharded_service<server> {
     // is called when reading the "system.clients" virtual table.
     struct ongoing_request {
         socket_address _client_address;
-        client_options_cache_entry_type _user_agent;
+        sstring _user_agent;
         sstring _username;
         scheduling_group _scheduling_group;
         bool _is_https;
@@ -100,9 +99,7 @@ class server : public peering_sharded_service<server> {
 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<uint16_t> port_proxy_protocol, std::optional<uint16_t> https_port_proxy_protocol,
-            std::optional<tls::credentials_builder> creds,
+    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<bool> warn_authorization, utils::updateable_value<uint64_t> max_users_query_size_in_trace_output,
             semaphore* memory_limiter, utils::updateable_value<uint32_t> max_concurrent_requests);
     future<> stop();
@@ -110,7 +107,7 @@ public:
     // 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<foreign_ptr<std::unique_ptr<client_data>>>> 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/streams.cc

@@ -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));
-        co_return rjson::print(std::move(ret));
+        return make_ready_future<executor::request_return_type>(rjson::print(std::move(ret)));
     }
 
     // TODO: label
@@ -502,121 +502,123 @@ future<executor::request_return_type> executor::describe_stream(client_state& cl
     // filter out cdc generations older than the table or now() - cdc::ttl (typically dynamodb_streams_max_window - 24h)
     auto low_ts = std::max(as_timepoint(schema->id()), db_clock::now() - ttl);
 
-    std::map<db_clock::time_point, cdc::streams_version> topologies = co_await _sdks.cdc_get_versioned_streams(low_ts, { normal_token_owners });
-    auto e = topologies.end();
-    auto prev = e;
-    auto shards = rjson::empty_array();
+    return _sdks.cdc_get_versioned_streams(low_ts, { normal_token_owners }).then([db, shard_start, limit, ret = std::move(ret), stream_desc = std::move(stream_desc)] (std::map<db_clock::time_point, cdc::streams_version> topologies) mutable {
 
-    std::optional<shard_id> last;
+        auto e = topologies.end();
+        auto prev = e;
+        auto shards = rjson::empty_array();
 
-    auto i = topologies.begin();
-    // if we're a paged query, skip to the generation where we left of.
-    if (shard_start) {
-        i = topologies.find(shard_start->time);
-    }
+        std::optional<shard_id> last;
 
-    // for parent-child stuff we need id:s to be sorted by token
-    // (see explanation above) since we want to find closest
-    // token boundary when determining parent.
-    // #7346 - we processed and searched children/parents in
-    // stored order, which is not necessarily token order,
-    // so the finding of "closest" token boundary (using upper bound)
-    // could give somewhat weird results.
-    static auto token_cmp = [](const cdc::stream_id& id1, const cdc::stream_id& id2) {
-        return id1.token() < id2.token();
-    };
+        auto i = topologies.begin();
+        // if we're a paged query, skip to the generation where we left of.
+        if (shard_start) {
+            i = topologies.find(shard_start->time);
+        }
 
-    // #7409 - shards must be returned in lexicographical order,
-    // normal bytes compare is string_traits<int8_t>::compare.
-    // thus bytes 0x8000 is less than 0x0000. By doing unsigned
-    // compare instead we inadvertently will sort in string lexical.
-    static auto id_cmp = [](const cdc::stream_id& id1, const cdc::stream_id& id2) {
-        return compare_unsigned(id1.to_bytes(), id2.to_bytes()) < 0;
-    };
-
-    // need a prev even if we are skipping stuff
-    if (i != topologies.begin()) {
-        prev = std::prev(i);
-    }
-
-    for (; limit > 0 && i != e; prev = i, ++i) {
-        auto& [ts, sv] = *i;
-
-        last = std::nullopt;
-
-        auto lo = sv.streams.begin();
-        auto end = sv.streams.end();
+        // for parent-child stuff we need id:s to be sorted by token
+        // (see explanation above) since we want to find closest
+        // token boundary when determining parent.
+        // #7346 - we processed and searched children/parents in
+        // stored order, which is not necessarily token order,
+        // so the finding of "closest" token boundary (using upper bound)
+        // could give somewhat weird results.
+        static auto token_cmp = [](const cdc::stream_id& id1, const cdc::stream_id& id2) {
+            return id1.token() < id2.token();
+        };
 
         // #7409 - shards must be returned in lexicographical order,
-        std::sort(lo, end, id_cmp);
+        // normal bytes compare is string_traits<int8_t>::compare.
+        // thus bytes 0x8000 is less than 0x0000. By doing unsigned
+        // compare instead we inadvertently will sort in string lexical.
+        static auto id_cmp = [](const cdc::stream_id& id1, const cdc::stream_id& id2) {
+            return compare_unsigned(id1.to_bytes(), id2.to_bytes()) < 0;
+        };
 
-        if (shard_start) {
-            // find next shard position
-            lo = std::upper_bound(lo, end, shard_start->id, id_cmp);
-            shard_start = std::nullopt;
+        // need a prev even if we are skipping stuff
+        if (i != topologies.begin()) {
+            prev = std::prev(i);
         }
 
-        if (lo != end && prev != e) {
-            // We want older stuff sorted in token order so we can find matching
-            // token range when determining parent shard.
-            std::stable_sort(prev->second.streams.begin(), prev->second.streams.end(), token_cmp);
-        }
-
-        auto expired = [&]() -> std::optional<db_clock::time_point> {
-            auto j = std::next(i);
-            if (j == e) {
-                return std::nullopt;
-            }
-            // add this so we sort of match potential 
-            // sequence numbers in get_records result.
-            return j->first + confidence_interval(db);
-        }();
-
-        while (lo != end) {
-            auto& id = *lo++;
-
-            auto shard = rjson::empty_object();
-
-            if (prev != e) {
-                auto& pids = prev->second.streams;
-                auto pid = std::upper_bound(pids.begin(), pids.end(), id.token(), [](const dht::token& t, const cdc::stream_id& id) {
-                    return t < id.token();
-                });
-                if (pid != pids.begin()) {
-                    pid = std::prev(pid);
-                }
-                if (pid != pids.end()) {
-                    rjson::add(shard, "ParentShardId", shard_id(prev->first, *pid));
-                }
-            }
-
-            last.emplace(ts, id);
-            rjson::add(shard, "ShardId", *last);
-            auto range = rjson::empty_object();
-            rjson::add(range, "StartingSequenceNumber", sequence_number(utils::UUID_gen::min_time_UUID(ts.time_since_epoch())));
-            if (expired) {
-                rjson::add(range, "EndingSequenceNumber", sequence_number(utils::UUID_gen::min_time_UUID(expired->time_since_epoch())));
-            }
-
-            rjson::add(shard, "SequenceNumberRange", std::move(range));
-            rjson::push_back(shards, std::move(shard));
-            
-            if (--limit == 0) {
-                break;
-            }
+        for (; limit > 0 && i != e; prev = i, ++i) {
+            auto& [ts, sv] = *i;
 
             last = std::nullopt;
+
+            auto lo = sv.streams.begin();
+            auto end = sv.streams.end();
+
+            // #7409 - shards must be returned in lexicographical order,
+            std::sort(lo, end, id_cmp);
+
+            if (shard_start) {
+                // find next shard position
+                lo = std::upper_bound(lo, end, shard_start->id, id_cmp);
+                shard_start = std::nullopt;
+            }
+
+            if (lo != end && prev != e) {
+                // We want older stuff sorted in token order so we can find matching
+                // token range when determining parent shard.
+                std::stable_sort(prev->second.streams.begin(), prev->second.streams.end(), token_cmp);
+            }
+
+            auto expired = [&]() -> std::optional<db_clock::time_point> {
+                auto j = std::next(i);
+                if (j == e) {
+                    return std::nullopt;
+                }
+                // add this so we sort of match potential 
+                // sequence numbers in get_records result.
+                return j->first + confidence_interval(db);
+            }();
+
+            while (lo != end) {
+                auto& id = *lo++;
+
+                auto shard = rjson::empty_object();
+
+                if (prev != e) {
+                    auto& pids = prev->second.streams;
+                    auto pid = std::upper_bound(pids.begin(), pids.end(), id.token(), [](const dht::token& t, const cdc::stream_id& id) {
+                        return t < id.token();
+                    });
+                    if (pid != pids.begin()) {
+                        pid = std::prev(pid);
+                    }
+                    if (pid != pids.end()) {
+                        rjson::add(shard, "ParentShardId", shard_id(prev->first, *pid));
+                    }
+                }
+
+                last.emplace(ts, id);
+                rjson::add(shard, "ShardId", *last);
+                auto range = rjson::empty_object();
+                rjson::add(range, "StartingSequenceNumber", sequence_number(utils::UUID_gen::min_time_UUID(ts.time_since_epoch())));
+                if (expired) {
+                    rjson::add(range, "EndingSequenceNumber", sequence_number(utils::UUID_gen::min_time_UUID(expired->time_since_epoch())));
+                }
+
+                rjson::add(shard, "SequenceNumberRange", std::move(range));
+                rjson::push_back(shards, std::move(shard));
+                
+                if (--limit == 0) {
+                    break;
+                }
+
+                last = std::nullopt;
+            }
         }
-    }
 
-    if (last) {
-        rjson::add(stream_desc, "LastEvaluatedShardId", *last);
-    }
+        if (last) {
+            rjson::add(stream_desc, "LastEvaluatedShardId", *last);
+        }
 
-    rjson::add(stream_desc, "Shards", std::move(shards));
-    rjson::add(ret, "StreamDescription", std::move(stream_desc));
-        
-    co_return rjson::print(std::move(ret));
+        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)));
+    });
 }
 
 enum class shard_iterator_type {
@@ -896,169 +898,172 @@ future<executor::request_return_type> executor::get_records(client_state& client
     auto command = ::make_lw_shared<query::read_command>(schema->id(), schema->version(), partition_slice, _proxy.get_max_result_size(partition_slice),
             query::tombstone_limit(_proxy.get_tombstone_limit()), query::row_limit(limit * mul));
 
-    service::storage_proxy::coordinator_query_result qr = co_await _proxy.query(schema, std::move(command), std::move(partition_ranges), cl, service::storage_proxy::coordinator_query_options(default_timeout(), std::move(permit), client_state));
-    cql3::selection::result_set_builder builder(*selection, gc_clock::now());
-    query::result_view::consume(*qr.query_result, partition_slice, cql3::selection::result_set_builder::visitor(builder, *schema, *selection));
+    co_return co_await _proxy.query(schema, std::move(command), std::move(partition_ranges), cl, service::storage_proxy::coordinator_query_options(default_timeout(), std::move(permit), client_state)).then(
+            [this, schema, partition_slice = std::move(partition_slice), selection = std::move(selection), start_time = std::move(start_time), limit, key_names = std::move(key_names), attr_names = std::move(attr_names), type, iter, high_ts] (service::storage_proxy::coordinator_query_result qr) mutable {       
+        cql3::selection::result_set_builder builder(*selection, gc_clock::now());
+        query::result_view::consume(*qr.query_result, partition_slice, cql3::selection::result_set_builder::visitor(builder, *schema, *selection));
 
-    auto result_set = builder.build();
-    auto records = rjson::empty_array();
+        auto result_set = builder.build();
+        auto records = rjson::empty_array();
 
-    auto& metadata = result_set->get_metadata();
+        auto& metadata = result_set->get_metadata();
 
-    auto op_index = std::distance(metadata.get_names().begin(), 
-        std::find_if(metadata.get_names().begin(), metadata.get_names().end(), [](const lw_shared_ptr<cql3::column_specification>& cdef) {
-            return cdef->name->name() == op_column_name;
-        })
-    );
-    auto ts_index = std::distance(metadata.get_names().begin(), 
-        std::find_if(metadata.get_names().begin(), metadata.get_names().end(), [](const lw_shared_ptr<cql3::column_specification>& cdef) {
-            return cdef->name->name() == timestamp_column_name;
-        })
-    );
-    auto eor_index = std::distance(metadata.get_names().begin(), 
-        std::find_if(metadata.get_names().begin(), metadata.get_names().end(), [](const lw_shared_ptr<cql3::column_specification>& cdef) {
-            return cdef->name->name() == eor_column_name;
-        })
-    );
+        auto op_index = std::distance(metadata.get_names().begin(), 
+            std::find_if(metadata.get_names().begin(), metadata.get_names().end(), [](const lw_shared_ptr<cql3::column_specification>& cdef) {
+                return cdef->name->name() == op_column_name;
+            })
+        );
+        auto ts_index = std::distance(metadata.get_names().begin(), 
+            std::find_if(metadata.get_names().begin(), metadata.get_names().end(), [](const lw_shared_ptr<cql3::column_specification>& cdef) {
+                return cdef->name->name() == timestamp_column_name;
+            })
+        );
+        auto eor_index = std::distance(metadata.get_names().begin(), 
+            std::find_if(metadata.get_names().begin(), metadata.get_names().end(), [](const lw_shared_ptr<cql3::column_specification>& cdef) {
+                return cdef->name->name() == eor_column_name;
+            })
+        );
 
-    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();
+        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>;
+        using op_utype = std::underlying_type_t<cdc::operation>;
 
-    auto maybe_add_record = [&] {
-        if (!dynamodb.ObjectEmpty()) {
-            rjson::add(record, "dynamodb", std::move(dynamodb));
-            dynamodb = rjson::empty_object();
-        }
-        if (!record.ObjectEmpty()) {
-            rjson::add(record, "awsRegion", rjson::from_string(dc_name));
-            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;
-        }
-    };
+        auto maybe_add_record = [&] {
+            if (!dynamodb.ObjectEmpty()) {
+                rjson::add(record, "dynamodb", std::move(dynamodb));
+                dynamodb = rjson::empty_object();
+            }
+            if (!record.ObjectEmpty()) {
+                rjson::add(record, "awsRegion", rjson::from_string(dc_name));
+                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;
+            }
+        };
 
-    for (auto& row : result_set->rows()) {
-        auto op = static_cast<cdc::operation>(value_cast<op_utype>(data_type_for<op_utype>()->deserialize(*row[op_index])));
-        auto ts = value_cast<utils::UUID>(data_type_for<utils::UUID>()->deserialize(*row[ts_index]));
-        auto eor = row[eor_index].has_value() ? value_cast<bool>(boolean_type->deserialize(*row[eor_index])) : false;
+        for (auto& row : result_set->rows()) {
+            auto op = static_cast<cdc::operation>(value_cast<op_utype>(data_type_for<op_utype>()->deserialize(*row[op_index])));
+            auto ts = value_cast<utils::UUID>(data_type_for<utils::UUID>()->deserialize(*row[ts_index]));
+            auto eor = row[eor_index].has_value() ? value_cast<bool>(boolean_type->deserialize(*row[eor_index])) : false;
 
-        if (!dynamodb.HasMember("Keys")) {
-            auto keys = rjson::empty_object();
-            describe_single_item(*selection, row, key_names, keys);
-            rjson::add(dynamodb, "Keys", std::move(keys));
-            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
-        }
+            if (!dynamodb.HasMember("Keys")) {
+                auto keys = rjson::empty_object();
+                describe_single_item(*selection, row, key_names, keys);
+                rjson::add(dynamodb, "Keys", std::move(keys));
+                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
+            }
 
-        /**
-         * We merge rows with same timestamp into a single event.
-         * This is pretty much needed, because a CDC row typically
-         * encodes ~half the info of an alternator write. 
-         * 
-         * A big, big downside to how alternator records are written
-         * (i.e. CQL), is that the distinction between INSERT and UPDATE
-         * is somewhat lost/unmappable to actual eventName. 
-         * A write (currently) always looks like an insert+modify
-         * regardless whether we wrote existing record or not. 
-         * 
-         * Maybe RMW ops could be done slightly differently so 
-         * we can distinguish them here...
-         * 
-         * For now, all writes will become MODIFY.
-         * 
-         * Note: we do not check the current pre/post
-         * flags on CDC log, instead we use data to 
-         * drive what is returned. This is (afaict)
-         * consistent with dynamo streams
-         */
-        switch (op) {
-        case cdc::operation::pre_image:
-        case cdc::operation::post_image:
-        {
-            auto item = rjson::empty_object();
-            describe_single_item(*selection, row, attr_names, item, nullptr, true);
-            describe_single_item(*selection, row, key_names, item);
-            rjson::add(dynamodb, op == cdc::operation::pre_image ? "OldImage" : "NewImage", std::move(item));
-            break;
-        }
-        case cdc::operation::update:
-            rjson::add(record, "eventName", "MODIFY");
-            break;
-        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;
-        }
-        if (eor) {
-            maybe_add_record();
-            timestamp = ts;
-            if (limit == 0) {
+            /**
+             * We merge rows with same timestamp into a single event.
+             * This is pretty much needed, because a CDC row typically
+             * encodes ~half the info of an alternator write. 
+             * 
+             * A big, big downside to how alternator records are written
+             * (i.e. CQL), is that the distinction between INSERT and UPDATE
+             * is somewhat lost/unmappable to actual eventName. 
+             * A write (currently) always looks like an insert+modify
+             * regardless whether we wrote existing record or not. 
+             * 
+             * Maybe RMW ops could be done slightly differently so 
+             * we can distinguish them here...
+             * 
+             * For now, all writes will become MODIFY.
+             * 
+             * Note: we do not check the current pre/post
+             * flags on CDC log, instead we use data to 
+             * drive what is returned. This is (afaict)
+             * consistent with dynamo streams
+             */
+            switch (op) {
+            case cdc::operation::pre_image:
+            case cdc::operation::post_image:
+            {
+                auto item = rjson::empty_object();
+                describe_single_item(*selection, row, attr_names, item, nullptr, true);
+                describe_single_item(*selection, row, key_names, item);
+                rjson::add(dynamodb, op == cdc::operation::pre_image ? "OldImage" : "NewImage", std::move(item));
                 break;
             }
+            case cdc::operation::update:
+                rjson::add(record, "eventName", "MODIFY");
+                break;
+            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;
+            }
+            if (eor) {
+                maybe_add_record();
+                timestamp = ts;
+                if (limit == 0) {
+                    break;
+                }
+            }
         }
-    }
 
-    auto ret = rjson::empty_object();
-    auto nrecords = records.Size();
-    rjson::add(ret, "Records", std::move(records));
+        auto ret = rjson::empty_object();
+        auto nrecords = records.Size();
+        rjson::add(ret, "Records", std::move(records));
 
-    if (nrecords != 0) {
-        // #9642. Set next iterators threshold to > last
-        shard_iterator next_iter(iter.table, iter.shard, *timestamp, false);
-        // Note that here we unconditionally return NextShardIterator,
-        // without checking if maybe we reached the end-of-shard. If the
-        // shard did end, then the next read will have nrecords == 0 and
-        // 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);
-        co_return rjson::print(std::move(ret));
-    }
+        if (nrecords != 0) {
+            // #9642. Set next iterators threshold to > last
+            shard_iterator next_iter(iter.table, iter.shard, *timestamp, false);
+            // Note that here we unconditionally return NextShardIterator,
+            // without checking if maybe we reached the end-of-shard. If the
+            // shard did end, then the next read will have nrecords == 0 and
+            // 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)));
+        }
 
-    // ugh. figure out if we are and end-of-shard
-    auto normal_token_owners = _proxy.get_token_metadata_ptr()->count_normal_token_owners();
+        // ugh. figure out if we are and end-of-shard
+        auto normal_token_owners = _proxy.get_token_metadata_ptr()->count_normal_token_owners();
 
-    db_clock::time_point ts = co_await _sdks.cdc_current_generation_timestamp({ normal_token_owners });
-    auto& shard = iter.shard;
+        return _sdks.cdc_current_generation_timestamp({ normal_token_owners }).then([this, iter, high_ts, start_time, ret = std::move(ret)](db_clock::time_point ts) mutable {
+            auto& shard = iter.shard;            
 
-    if (shard.time < ts && ts < high_ts) {
-        // The DynamoDB documentation states that when a shard is
-        // closed, reading it until the end has NextShardIterator
-        // "set to null". Our test test_streams_closed_read
-        // confirms that by "null" they meant not set at all.
-    } else {
-        // We could have return the same iterator again, but we did
-        // a search from it until high_ts and found nothing, so we
-        // can also start the next search from high_ts.
-        // TODO: but why? It's simpler just to leave the iterator be.
-        shard_iterator next_iter(iter.table, iter.shard, utils::UUID_gen::min_time_UUID(high_ts.time_since_epoch()), true);
-        rjson::add(ret, "NextShardIterator", iter);
-    }
-    _stats.api_operations.get_records_latency.mark(std::chrono::steady_clock::now() - start_time);
-    if (is_big(ret)) {
-        co_return make_streamed(std::move(ret));
-    }
-    co_return rjson::print(std::move(ret));
+            if (shard.time < ts && ts < high_ts) {
+                // The DynamoDB documentation states that when a shard is
+                // closed, reading it until the end has NextShardIterator
+                // "set to null". Our test test_streams_closed_read
+                // confirms that by "null" they meant not set at all.
+            } else {
+                // We could have return the same iterator again, but we did
+                // a search from it until high_ts and found nothing, so we
+                // can also start the next search from high_ts.
+                // TODO: but why? It's simpler just to leave the iterator be.
+                shard_iterator next_iter(iter.table, iter.shard, utils::UUID_gen::min_time_UUID(high_ts.time_since_epoch()), true);
+                rjson::add(ret, "NextShardIterator", iter);
+            }
+            _stats.api_operations.get_records_latency.mark(std::chrono::steady_clock::now() - start_time);
+            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)));
+        });
+    });
 }
 
 bool executor::add_stream_options(const rjson::value& stream_specification, schema_builder& builder, service::storage_proxy& sp) {

alternator/ttl.cc

@@ -46,7 +46,6 @@
 #include "alternator/executor.hh"
 #include "alternator/controller.hh"
 #include "alternator/serialization.hh"
-#include "alternator/ttl_tag.hh"
 #include "dht/sharder.hh"
 #include "db/config.hh"
 #include "db/tags/utils.hh"
@@ -58,10 +57,19 @@ static logging::logger tlogger("alternator_ttl");
 
 namespace alternator {
 
+// We write the expiration-time attribute enabled on a table in 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;
+
 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++;
     if (!_proxy.features().alternator_ttl) {
-        co_return api_error::unknown_operation("UpdateTimeToLive not yet supported. Upgrade all nodes to a version that supports it.");
+        co_return api_error::unknown_operation("UpdateTimeToLive not yet supported. Experimental support is available if the 'alternator-ttl' experimental feature is enabled on all nodes.");
     }
 
     schema_ptr schema = get_table(_proxy, request);
@@ -85,7 +93,7 @@ future<executor::request_return_type> executor::update_time_to_live(client_state
     if (v->GetStringLength() < 1 || v->GetStringLength() > 255) {
         co_return api_error::validation("The length of AttributeName must be between 1 and 255");
     }
-    sstring attribute_name = rjson::to_sstring(*v);
+    sstring attribute_name(v->GetString(), v->GetStringLength());
 
     co_await verify_permission(_enforce_authorization, _warn_authorization, client_state, schema, auth::permission::ALTER, _stats);
     co_await db::modify_tags(_mm, schema->ks_name(), schema->cf_name(), [&](std::map<sstring, sstring>& tags_map) {
@@ -133,7 +141,7 @@ future<executor::request_return_type> executor::describe_time_to_live(client_sta
 
 // expiration_service is a sharded service responsible for cleaning up expired
 // items in all tables with per-item expiration enabled. Currently, this means
-// Alternator tables with TTL configured via an UpdateTimeToLive request.
+// Alternator tables with TTL configured via a UpdateTimeToLive request.
 //
 // Here is a brief overview of how the expiration service works:
 //
@@ -316,7 +324,9 @@ static future<std::vector<std::pair<dht::token_range, locator::host_id>>> get_se
     const auto& tm = *erm->get_token_metadata_ptr();
     const auto& sorted_tokens = tm.sorted_tokens();
     std::vector<std::pair<dht::token_range, locator::host_id>> ret;
-    throwing_assert(!sorted_tokens.empty());
+    if (sorted_tokens.empty()) {
+        on_internal_error(tlogger, "Token metadata is empty");
+    }
     auto prev_tok = sorted_tokens.back();
     for (const auto& tok : sorted_tokens) {
         co_await coroutine::maybe_yield();
@@ -553,7 +563,7 @@ static future<> scan_table_ranges(
         expiration_service::stats& expiration_stats)
 {
     const schema_ptr& s = scan_ctx.s;
-    throwing_assert(partition_ranges.size() == 1); // otherwise issue #9167 will cause incorrect results.
+    SCYLLA_ASSERT (partition_ranges.size() == 1); // otherwise issue #9167 will cause incorrect results.
     auto p = service::pager::query_pagers::pager(proxy, s, scan_ctx.selection, *scan_ctx.query_state_ptr,
             *scan_ctx.query_options, scan_ctx.command, std::move(partition_ranges), nullptr);
     while (!p->is_exhausted()) {
@@ -583,7 +593,7 @@ static future<> scan_table_ranges(
             if (retries >= 10) {
                 // Don't get stuck forever asking the same page, maybe there's
                 // a bug or a real problem in several replicas. Give up on
-                // this scan and retry the scan from a random position later,
+                // this scan an retry the scan from a random position later,
                 // in the next scan period.
                 throw runtime_exception("scanner thread failed after too many timeouts for the same page");
             }
@@ -630,38 +640,13 @@ static future<> scan_table_ranges(
                 }
             } else {
                 // For a real column to contain an expiration time, it
-                // must be a numeric type. We currently support decimal
-                // (used by Alternator TTL) as well as bigint, int and
-                // timestamp (used by CQL per-row TTL).
-                switch (meta[*expiration_column]->type->get_kind()) {
-                    case abstract_type::kind::decimal:
-                        // Used by Alternator TTL for key columns not stored
-                        // in the map. The value is in seconds, fractional
-                        // part is ignored.
-                        expired = is_expired(value_cast<big_decimal>(v), now);
-                        break;
-                    case abstract_type::kind::long_kind:
-                        // Used by CQL per-row TTL. The value is in seconds.
-                        expired = is_expired(gc_clock::time_point(std::chrono::seconds(value_cast<int64_t>(v))), now);
-                        break;
-                    case abstract_type::kind::int32:
-                        // Used by CQL per-row TTL. The value is in seconds.
-                        // Using int type is not recommended because it will
-                        // overflow in 2038, but we support it to allow users
-                        // to use existing int columns for expiration.
-                        expired = is_expired(gc_clock::time_point(std::chrono::seconds(value_cast<int32_t>(v))), now);
-                        break;
-                    case abstract_type::kind::timestamp:
-                        // Used by CQL per-row TTL. The value is in milliseconds
-                        // but we truncate it to gc_clock's precision (whole seconds).
-                        expired = is_expired(gc_clock::time_point(std::chrono::duration_cast<gc_clock::duration>(value_cast<db_clock::time_point>(v).time_since_epoch())), now);
-                        break;
-                    default:
-                        // Should never happen - we verified the column's type
-                        // before starting the scan.
-                        [[unlikely]]
-                        on_internal_error(tlogger, format("expiration scanner value of unsupported type {} in column {}", meta[*expiration_column]->type->cql3_type_name(), scan_ctx.column_name) );
-                }
+                // must be a numeric type.
+                // FIXME: Currently we only support decimal_type (which is
+                // what Alternator uses), but other numeric types can be
+                // supported as well to make this feature more useful in CQL.
+                // Note that kind::decimal is also checked above.
+                big_decimal n = value_cast<big_decimal>(v);
+                expired = is_expired(n, now);
             }
             if (expired) {
                 expiration_stats.items_deleted++;
@@ -723,12 +708,16 @@ static future<bool> scan_table(
         co_return false;
     }
     // attribute_name may be one of the schema's columns (in Alternator, this
-    // means a key column, in CQL it's a regular column), or an element in
-    // Alternator's attrs map encoded in Alternator's JSON encoding (which we
-    // decode). If attribute_name is a real column, in Alternator it will have
-    // the type decimal, counting seconds since the UNIX epoch, while in CQL
-    // it will one of the types bigint or int (counting seconds) or timestamp
-    // (counting milliseconds).
+    // means it's a key column), or an element in Alternator's attrs map
+    // encoded in Alternator's JSON encoding.
+    // FIXME: To make this less Alternators-specific, we should encode in the
+    // single key's value three things:
+    // 1. The name of a column
+    // 2. Optionally if column is a map, a member in the map
+    // 3. The deserializer for the value: CQL or Alternator (JSON).
+    // The deserializer can be guessed: If the given column or map item is
+    // numeric, it can be used directly. If it is a "bytes" type, it needs to
+    // be deserialized using Alternator's deserializer.
     bytes column_name = to_bytes(*attribute_name);
     const column_definition *cd = s->get_column_definition(column_name);
     std::optional<std::string> member;
@@ -747,14 +736,11 @@ static future<bool> scan_table(
     data_type column_type = cd->type;
     // Verify that the column has the right type: If "member" exists
     // the column must be a map, and if it doesn't, the column must
-    // be decimal_type (Alternator), bigint, int or timestamp (CQL).
-    // If the column has the wrong type nothing can get expired in
-    // this table, and it's pointless to scan it.
+    // (currently) be a decimal_type. If the column has the wrong type
+    // nothing can get expired in this table, and it's pointless to
+    // scan it.
     if ((member && column_type->get_kind() != abstract_type::kind::map) ||
-        (!member && column_type->get_kind() != abstract_type::kind::decimal &&
-         column_type->get_kind() != abstract_type::kind::long_kind &&
-         column_type->get_kind() != abstract_type::kind::int32 &&
-         column_type->get_kind() != abstract_type::kind::timestamp)) {
+        (!member && column_type->get_kind() != abstract_type::kind::decimal)) {
         tlogger.info("table {} TTL column has unsupported type, not scanning", s->cf_name());
         co_return false;
     }
@@ -781,7 +767,7 @@ static future<bool> scan_table(
                 // 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, erm->get_topology()); // throws if no secondary replica
+                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);
@@ -892,10 +878,12 @@ future<> expiration_service::run() {
 future<> expiration_service::start() {
     // Called by main() on each shard to start the expiration-service
     // thread. Just runs run() in the background and allows stop().
-    if (!shutting_down()) {
-        _end = run().handle_exception([] (std::exception_ptr ep) {
-            tlogger.error("expiration_service failed: {}", ep);
-        });
+    if (_db.features().alternator_ttl) {
+        if (!shutting_down()) {
+            _end = run().handle_exception([] (std::exception_ptr ep) {
+                tlogger.error("expiration_service failed: {}", ep);
+            });
+        }
     }
     return make_ready_future<>();
 }

alternator/ttl.hh

@@ -30,7 +30,7 @@ namespace alternator {
 
 // expiration_service is a sharded service responsible for cleaning up expired
 // items in all tables with per-item expiration enabled. Currently, this means
-// Alternator tables with TTL configured via an UpdateTimeToLive request.
+// Alternator tables with TTL configured via a UpdateTimeToLeave request.
 class expiration_service final : public seastar::peering_sharded_service<expiration_service> {
 public:
     // Object holding per-shard statistics related to the expiration service.
@@ -52,7 +52,7 @@ private:
     data_dictionary::database _db;
     service::storage_proxy& _proxy;
     gms::gossiper& _gossiper;
-    // _end is set by start(), and resolves when the background service
+    // _end is set by start(), and resolves when the the background service
     // started by it ends. To ask the background service to end, _abort_source
     // should be triggered. stop() below uses both _abort_source and _end.
     std::optional<future<>> _end;

alternator/ttl_tag.hh

@@ -1,26 +0,0 @@
-/*
- * Copyright 2026-present ScyllaDB
- */
-
-/*
- * SPDX-License-Identifier: LicenseRef-ScyllaDB-Source-Available-1.0
- */
-
-#pragma once
-
-#include "seastarx.hh"
-#include <seastar/core/sstring.hh>
-
-namespace alternator {
-// We use the table tag TTL_TAG_KEY ("system:ttl_attribute") to remember
-// which attribute was chosen as the expiration-time attribute for
-// Alternator's TTL and CQL's per-row TTL features.
-// Currently, the *value* of this tag is simply the name of the attribute:
-// It can refer to a real column or if that doesn't exist, to a member of
-// the ":attrs" map column (which Alternator uses).
-extern const sstring TTL_TAG_KEY;
-} // namespace alternator
-
-// let users use TTL_TAG_KEY without the "alternator::" prefix,
-// to make it easier to move it to a different namespace later.
-using alternator::TTL_TAG_KEY;

api/CMakeLists.txt

@@ -31,7 +31,6 @@ set(swagger_files
   api-doc/column_family.json
   api-doc/commitlog.json
   api-doc/compaction_manager.json
-  api-doc/client_routes.json
   api-doc/config.json
   api-doc/cql_server_test.json
   api-doc/endpoint_snitch_info.json
@@ -69,7 +68,6 @@ target_sources(api
   PRIVATE
     api.cc
     cache_service.cc
-    client_routes.cc
     collectd.cc
     column_family.cc
     commitlog.cc

api/api-doc/authorization_cache.json

@@ -12,7 +12,7 @@
       "operations":[
         {
           "method":"POST",
-          "summary":"Resets authorized prepared statements cache",
+          "summary":"Reset cache",
           "type":"void",
           "nickname":"authorization_cache_reset",
           "produces":[

api/api-doc/client_routes.def.json

@@ -1,23 +0,0 @@
-    , "client_routes_entry": {
-        "id": "client_routes_entry",
-        "summary": "An entry storing client routes",
-        "properties": {
-            "connection_id": {"type": "string"},
-            "host_id": {"type": "string", "format": "uuid"},
-            "address": {"type": "string"},
-            "port": {"type": "integer"},
-            "tls_port": {"type": "integer"},
-            "alternator_port": {"type": "integer"},
-            "alternator_https_port": {"type": "integer"}
-        },
-        "required": ["connection_id", "host_id", "address"]
-    }
-    , "client_routes_key": {
-        "id": "client_routes_key",
-        "summary": "A key of client_routes_entry",
-        "properties": {
-            "connection_id": {"type": "string"},
-            "host_id": {"type": "string", "format": "uuid"}
-        }
-    }
-

api/api-doc/client_routes.json

@@ -1,74 +0,0 @@
-    , "/v2/client-routes":{
-        "get": {
-            "description":"List all client route entries",
-            "operationId":"get_client_routes",
-            "tags":["client_routes"],
-            "produces":[
-                "application/json"
-            ],
-            "parameters":[],
-            "responses":{
-                "200":{
-                    "schema":{
-                        "type":"array",
-                        "items":{ "$ref":"#/definitions/client_routes_entry" }
-                    }
-                },
-                "default":{
-                    "description":"unexpected error",
-                    "schema":{"$ref":"#/definitions/ErrorModel"}
-                }
-            }
-        },
-        "post": {
-            "description":"Upsert one or more client route entries",
-            "operationId":"set_client_routes",
-            "tags":["client_routes"],
-            "parameters":[
-                {
-                    "name":"body",
-                    "in":"body",
-                    "required":true,
-                    "schema":{
-                        "type":"array",
-                        "items":{ "$ref":"#/definitions/client_routes_entry" }
-                    }
-                }
-            ],
-            "responses":{
-                "200":{ "description": "OK" },
-                "default":{
-                    "description":"unexpected error",
-                    "schema":{ "$ref":"#/definitions/ErrorModel" }
-                }
-            }
-        },
-        "delete": {
-            "description":"Delete one or more client route entries",
-            "operationId":"delete_client_routes",
-            "tags":["client_routes"],
-            "parameters":[
-                {
-                    "name":"body",
-                    "in":"body",
-                    "required":true,
-                    "schema":{
-                        "type":"array",
-                        "items":{ "$ref":"#/definitions/client_routes_key" }
-                    }
-                }
-            ],
-            "responses":{
-                "200":{
-                    "description": "OK"
-                },
-                "default":{
-                    "description":"unexpected error",
-                    "schema":{
-                        "$ref":"#/definitions/ErrorModel"
-                    }
-                }
-            }
-        }
-    }
-

api/api-doc/storage_service.json

@@ -729,6 +729,14 @@
                      "allowMultiple":false,
                      "type":"boolean",
                      "paramType":"query"
+                  },
+                  {
+                     "name":"use_sstable_identifier",
+                     "description":"Use the sstable identifier UUID, if available, rather than the sstable generation.",
+                     "required":false,
+                     "allowMultiple":false,
+                     "type":"boolean",
+                     "paramType":"query"
                   }
                ]
             },
@@ -3085,48 +3093,6 @@
             }
          ]
       },
-
-      {
-         "path":"/storage_service/tablets/snapshots",
-         "operations":[
-            {
-               "method":"POST",
-               "summary":"Takes the snapshot for the given keyspaces/tables. A snapshot name must be specified.",
-               "type":"void",
-               "nickname":"take_cluster_snapshot",
-               "produces":[
-                  "application/json"
-               ],
-               "parameters":[
-                  {
-                     "name":"tag",
-                     "description":"the tag given to the snapshot",
-                     "required":true,
-                     "allowMultiple":false,
-                     "type":"string",
-                     "paramType":"query"
-                  },
-                  {
-                     "name":"keyspace",
-                     "description":"Keyspace(s) to snapshot. Multiple keyspaces can be provided using a comma-separated list. If omitted, snapshot all keyspaces.",
-                     "required":false,
-                     "allowMultiple":false,
-                     "type":"string",
-                     "paramType":"query"
-                  },
-                  {
-                     "name":"table",
-                     "description":"Table(s) to snapshot. Multiple tables (in a single keyspace) can be provided using a comma-separated list. If omitted, snapshot all tables in the given keyspace(s).",
-                     "required":false,
-                     "allowMultiple":false,
-                     "type":"string",
-                     "paramType":"query"
-                  }
-               ]
-            }
-         ]
-      },
-
       {
          "path":"/storage_service/quiesce_topology",
          "operations":[

api/api.cc

@@ -37,7 +37,6 @@
 #include "raft.hh"
 #include "gms/gossip_address_map.hh"
 #include "service_levels.hh"
-#include "client_routes.hh"
 
 logging::logger apilog("api");
 
@@ -68,11 +67,9 @@ future<> set_server_init(http_context& ctx) {
         rb02->set_api_doc(r);
         rb02->register_api_file(r, "swagger20_header");
         rb02->register_api_file(r, "metrics");
-        rb02->register_api_file(r, "client_routes");
         rb->register_function(r, "system",
                 "The system related API");
         rb02->add_definitions_file(r, "metrics");
-        rb02->add_definitions_file(r, "client_routes");
         set_system(ctx, r);
         rb->register_function(r, "error_injection",
             "The error injection API");
@@ -132,16 +129,6 @@ future<> unset_server_storage_service(http_context& ctx) {
     return ctx.http_server.set_routes([&ctx] (routes& r) { unset_storage_service(ctx, r); });
 }
 
-future<> set_server_client_routes(http_context& ctx, sharded<service::client_routes_service>& cr) {
-    return ctx.http_server.set_routes([&ctx, &cr] (routes& r) {
-        set_client_routes(ctx, r, cr);
-    });
-}
-
-future<> unset_server_client_routes(http_context& ctx) {
-    return ctx.http_server.set_routes([&ctx] (routes& r) { unset_client_routes(ctx, r); });
-}
-
 future<> set_load_meter(http_context& ctx, service::load_meter& lm) {
     return ctx.http_server.set_routes([&ctx, &lm] (routes& r) { set_load_meter(ctx, r, lm); });
 }

api/api.hh

@@ -23,6 +23,31 @@
 
 namespace api {
 
+template<class T>
+std::vector<T> map_to_key_value(const std::map<sstring, sstring>& map) {
+    std::vector<T> res;
+    res.reserve(map.size());
+
+    for (const auto& [key, value] : map) {
+        res.push_back(T());
+        res.back().key = key;
+        res.back().value = value;
+    }
+    return res;
+}
+
+template<class T, class MAP>
+std::vector<T>& map_to_key_value(const MAP& map, std::vector<T>& res) {
+    res.reserve(res.size() + std::size(map));
+
+    for (const auto& [key, value] : map) {
+        T val;
+        val.key = fmt::to_string(key);
+        val.value = fmt::to_string(value);
+        res.push_back(val);
+    }
+    return res;
+}
 template <typename T, typename S = T>
 T map_sum(T&& dest, const S& src) {
     for (const auto& i : src) {

api/api_init.hh

@@ -29,7 +29,6 @@ class storage_proxy;
 class storage_service;
 class raft_group0_client;
 class raft_group_registry;
-class client_routes_service;
 
 } // namespace service
 
@@ -100,8 +99,6 @@ future<> set_server_snitch(http_context& ctx, sharded<locator::snitch_ptr>& snit
 future<> unset_server_snitch(http_context& ctx);
 future<> set_server_storage_service(http_context& ctx, sharded<service::storage_service>& ss, service::raft_group0_client&);
 future<> unset_server_storage_service(http_context& ctx);
-future<> set_server_client_routes(http_context& ctx, sharded<service::client_routes_service>& cr);
-future<> unset_server_client_routes(http_context& ctx);
 future<> set_server_sstables_loader(http_context& ctx, sharded<sstables_loader>& sst_loader);
 future<> unset_server_sstables_loader(http_context& ctx);
 future<> set_server_view_builder(http_context& ctx, sharded<db::view::view_builder>& vb, sharded<gms::gossiper>& g);

api/client_routes.cc

@@ -1,176 +0,0 @@
-/*
- * Copyright (C) 2025-present ScyllaDB
- *
- */
-
-/*
- * SPDX-License-Identifier: LicenseRef-ScyllaDB-Source-Available-1.0
- */
-
- #include <seastar/http/short_streams.hh>
-
-#include "client_routes.hh"
-#include "api/api.hh"
-#include "service/storage_service.hh"
-#include "service/client_routes.hh"
-#include "utils/rjson.hh"
-
-
-#include "api/api-doc/client_routes.json.hh"
-
-using namespace seastar::httpd;
-using namespace std::chrono_literals;
-using namespace json;
-
-extern logging::logger apilog;
-
-namespace api {
-
-static void validate_client_routes_endpoint(sharded<service::client_routes_service>& cr, sstring endpoint_name) {
-    if (!cr.local().get_feature_service().client_routes) {
-        apilog.warn("{}: called before the cluster feature was enabled", endpoint_name);
-        throw std::runtime_error(fmt::format("{} requires all nodes to support the CLIENT_ROUTES cluster feature", endpoint_name));
-    }
-}
-
-static sstring parse_string(const char* name, rapidjson::Value const& v) {
-    const auto it = v.FindMember(name);
-    if (it == v.MemberEnd()) {
-        throw bad_param_exception(fmt::format("Missing '{}'", name));
-    }
-    if (!it->value.IsString()) {
-        throw bad_param_exception(fmt::format("'{}' must be a string", name));
-    }
-    return {it->value.GetString(), it->value.GetStringLength()};
-}
-
-static std::optional<uint32_t> parse_port(const char* name, rapidjson::Value const& v) {
-    const auto it = v.FindMember(name);
-    if (it == v.MemberEnd()) {
-        return std::nullopt;
-    }
-    if (!it->value.IsInt()) {
-        throw bad_param_exception(fmt::format("'{}' must be an integer", name));
-    }
-    auto port = it->value.GetInt();
-    if (port < 1 || port > 65535) {
-        throw bad_param_exception(fmt::format("'{}' value={} is outside the allowed port range", name, port));
-    }
-    return port;
-}
-
-static std::vector<service::client_routes_service::client_route_entry> parse_set_client_array(const rapidjson::Document& root) {
-    if (!root.IsArray()) {
-        throw bad_param_exception("Body must be a JSON array");
-    }
-
-    std::vector<service::client_routes_service::client_route_entry> v;
-    v.reserve(root.GetArray().Size());
-    for (const auto& element : root.GetArray()) {
-        if (!element.IsObject()) { throw bad_param_exception("Each element must be object"); }
-
-        const auto port = parse_port("port", element);
-        const auto tls_port = parse_port("tls_port", element);
-        const auto alternator_port = parse_port("alternator_port", element);
-        const auto alternator_https_port = parse_port("alternator_https_port", element);
-
-        if (!port.has_value() && !tls_port.has_value() && !alternator_port.has_value() && !alternator_https_port.has_value()) {
-            throw bad_param_exception("At least one port field ('port', 'tls_port', 'alternator_port', 'alternator_https_port') must be specified");
-        }
-
-        v.emplace_back(
-            parse_string("connection_id", element),
-            utils::UUID{parse_string("host_id", element)},
-            parse_string("address", element),
-            port,
-            tls_port,
-            alternator_port,
-            alternator_https_port
-        );
-    }
-
-    return v;
-}
-
-static
-future<json::json_return_type>
-rest_set_client_routes(http_context& ctx, sharded<service::client_routes_service>& cr, std::unique_ptr<http::request> req) {
-    validate_client_routes_endpoint(cr, "rest_set_client_routes");
-
-    rapidjson::Document root;
-    auto content = co_await util::read_entire_stream_contiguous(*req->content_stream);
-    root.Parse(content.c_str());
-
-    co_await cr.local().set_client_routes(parse_set_client_array(root));
-    co_return seastar::json::json_void();
-}
-
-static std::vector<service::client_routes_service::client_route_key> parse_delete_client_array(const rapidjson::Document& root) {
-    if (!root.IsArray()) {
-        throw bad_param_exception("Body must be a JSON array");
-    }
-
-    std::vector<service::client_routes_service::client_route_key> v;
-    v.reserve(root.GetArray().Size());
-    for (const auto& element : root.GetArray()) {
-        v.emplace_back(
-            parse_string("connection_id", element),
-            utils::UUID{parse_string("host_id", element)}
-        );
-    }
-
-    return v;
-}
-
-static
-future<json::json_return_type>
-rest_delete_client_routes(http_context& ctx, sharded<service::client_routes_service>& cr, std::unique_ptr<http::request> req) {
-    validate_client_routes_endpoint(cr, "delete_client_routes");
-
-    rapidjson::Document root;
-    auto content = co_await util::read_entire_stream_contiguous(*req->content_stream);
-    root.Parse(content.c_str());
-
-    co_await cr.local().delete_client_routes(parse_delete_client_array(root));
-    co_return seastar::json::json_void();
-}
-
-static
-future<json::json_return_type>
-rest_get_client_routes(http_context& ctx, sharded<service::client_routes_service>& cr, std::unique_ptr<http::request> req) {
-    validate_client_routes_endpoint(cr, "get_client_routes");
-
-    co_return co_await cr.invoke_on(0, [] (service::client_routes_service& cr) -> future<json::json_return_type> {
-        co_return json::json_return_type(stream_range_as_array(co_await cr.get_client_routes(), [](const service::client_routes_service::client_route_entry & entry) {
-            seastar::httpd::client_routes_json::client_routes_entry obj;
-            obj.connection_id = entry.connection_id;
-            obj.host_id = fmt::to_string(entry.host_id);
-            obj.address = entry.address;
-            if (entry.port.has_value()) { obj.port = entry.port.value(); }
-            if (entry.tls_port.has_value()) { obj.tls_port = entry.tls_port.value(); }
-            if (entry.alternator_port.has_value()) { obj.alternator_port = entry.alternator_port.value(); }
-            if (entry.alternator_https_port.has_value()) { obj.alternator_https_port = entry.alternator_https_port.value(); }
-            return obj;
-        }));
-    });
-}
-
-void set_client_routes(http_context& ctx, routes& r, sharded<service::client_routes_service>& cr) {
-    seastar::httpd::client_routes_json::set_client_routes.set(r, [&ctx, &cr] (std::unique_ptr<seastar::http::request> req) {
-        return rest_set_client_routes(ctx, cr, std::move(req));
-    });
-    seastar::httpd::client_routes_json::delete_client_routes.set(r, [&ctx, &cr] (std::unique_ptr<seastar::http::request> req) {
-        return rest_delete_client_routes(ctx, cr, std::move(req));
-    });
-    seastar::httpd::client_routes_json::get_client_routes.set(r, [&ctx, &cr] (std::unique_ptr<seastar::http::request> req) {
-        return rest_get_client_routes(ctx, cr, std::move(req));
-    });
-}
-
-void unset_client_routes(http_context& ctx, routes& r) {
-    seastar::httpd::client_routes_json::set_client_routes.unset(r);
-    seastar::httpd::client_routes_json::delete_client_routes.unset(r);
-    seastar::httpd::client_routes_json::get_client_routes.unset(r);
-}
-
-}

api/client_routes.hh

@@ -1,20 +0,0 @@
-/*
- * Copyright (C) 2025-present ScyllaDB
- *
- */
-
-/*
- * SPDX-License-Identifier: LicenseRef-ScyllaDB-Source-Available-1.0
- */
-#pragma once
-
-#include <seastar/core/sharded.hh>
-#include <seastar/json/json_elements.hh>
-#include "api/api_init.hh"
-
-namespace api {
-
-void set_client_routes(http_context& ctx, httpd::routes& r, sharded<service::client_routes_service>& cr);
-void unset_client_routes(http_context& ctx, httpd::routes& r);
-
-}

api/storage_service.cc

@@ -515,15 +515,6 @@ void set_sstables_loader(http_context& ctx, routes& r, sharded<sstables_loader>&
         auto sstables = parsed.GetArray() |
             std::views::transform([] (const auto& s) { return sstring(rjson::to_string_view(s)); }) |
             std::ranges::to<std::vector>();
-        apilog.info("Restore invoked with following parameters: keyspace={}, table={}, endpoint={}, bucket={}, prefix={}, sstables_count={}, scope={}, primary_replica_only={}",
-                    keyspace,
-                    table,
-                    endpoint,
-                    bucket,
-                    prefix,
-                    sstables.size(),
-                    scope,
-                    primary_replica_only);
         auto task_id = co_await sst_loader.local().download_new_sstables(keyspace, table, prefix, std::move(sstables), endpoint, bucket, scope, primary_replica_only);
         co_return json::json_return_type(fmt::to_string(task_id));
     });
@@ -536,15 +527,13 @@ void unset_sstables_loader(http_context& ctx, routes& r) {
 }
 
 void set_view_builder(http_context& ctx, routes& r, sharded<db::view::view_builder>& vb, sharded<gms::gossiper>& g) {
-    ss::view_build_statuses.set(r, [&ctx, &vb, &g] (std::unique_ptr<http::request> req) -> future<json::json_return_type> {
+    ss::view_build_statuses.set(r, [&ctx, &vb, &g] (std::unique_ptr<http::request> req) {
         auto keyspace = validate_keyspace(ctx, req);
         auto view = req->get_path_param("view");
-        co_return json::json_return_type(stream_range_as_array(co_await vb.local().view_build_statuses(std::move(keyspace), std::move(view), g.local()), [] (const auto& i) {
-            storage_service_json::mapper res;
-            res.key = i.first;
-            res.value = i.second;
-            return res;
-        }));
+        return vb.local().view_build_statuses(std::move(keyspace), std::move(view), g.local()).then([] (std::unordered_map<sstring, sstring> status) {
+            std::vector<storage_service_json::mapper> res;
+            return make_ready_future<json::json_return_type>(map_to_key_value(std::move(status), res));
+        });
     });
 
     cf::get_built_indexes.set(r, [&vb](std::unique_ptr<http::request> req) -> future<json::json_return_type> {
@@ -558,13 +547,17 @@ void set_view_builder(http_context& ctx, routes& r, sharded<db::view::view_build
                 vp.insert(b.second);
             }
         }
+        std::vector<sstring> res;
         replica::database& db = vb.local().get_db();
         auto uuid = validate_table(db, ks, cf_name);
         replica::column_family& cf = db.find_column_family(uuid);
-        co_return cf.get_index_manager().list_indexes()
-                | std::views::transform([] (const auto& i) { return i.metadata().name(); })
-                | std::views::filter([&vp] (const auto& n) { return vp.contains(secondary_index::index_table_name(n)); })
-                | std::ranges::to<std::vector>();
+        res.reserve(cf.get_index_manager().list_indexes().size());
+        for (auto&& i : cf.get_index_manager().list_indexes()) {
+            if (vp.contains(secondary_index::index_table_name(i.metadata().name()))) {
+                res.emplace_back(i.metadata().name());
+            }
+        }
+        co_return res;
     });
 
 }
@@ -582,16 +575,6 @@ static future<json::json_return_type> describe_ring_as_json_for_table(const shar
     co_return json::json_return_type(stream_range_as_array(co_await ss.local().describe_ring_for_table(keyspace, table), token_range_endpoints_to_json));
 }
 
-namespace {
-template <typename Key, typename Value>
-storage_service_json::mapper map_to_json(const std::pair<Key, Value>& i) {
-    storage_service_json::mapper val;
-    val.key = fmt::to_string(i.first);
-    val.value = fmt::to_string(i.second);
-    return val;
-}
-}
-
 static
 future<json::json_return_type>
 rest_get_token_endpoint(http_context& ctx, sharded<service::storage_service>& ss, std::unique_ptr<http::request> req) {
@@ -609,7 +592,12 @@ rest_get_token_endpoint(http_context& ctx, sharded<service::storage_service>& ss
             throw bad_param_exception("Either provide both keyspace and table (for tablet table) or neither (for vnodes)");
         }
 
-        co_return json::json_return_type(stream_range_as_array(token_endpoints, &map_to_json<dht::token, gms::inet_address>));
+        co_return json::json_return_type(stream_range_as_array(token_endpoints, [](const auto& i) {
+            storage_service_json::mapper val;
+            val.key = fmt::to_string(i.first);
+            val.value = fmt::to_string(i.second);
+            return val;
+        }));
 }
 
 static
@@ -693,6 +681,7 @@ rest_get_range_to_endpoint_map(http_context& ctx, sharded<service::storage_servi
             table_id = validate_table(ctx.db.local(), keyspace, table);
         }
 
+        std::vector<ss::maplist_mapper> res;
         co_return stream_range_as_array(co_await ss.local().get_range_to_address_map(keyspace, table_id),
                 [](const std::pair<dht::token_range, inet_address_vector_replica_set>& entry){
             ss::maplist_mapper m;
@@ -783,13 +772,17 @@ rest_cleanup_all(http_context& ctx, sharded<service::storage_service>& ss, std::
 
         apilog.info("cleanup_all global={}", global);
 
-        if (global) {
-            co_await ss.invoke_on(0, [] (service::storage_service& ss) -> future<> {
-                co_return co_await ss.do_clusterwide_vnodes_cleanup();
-            });
+        auto done = !global ? false : co_await ss.invoke_on(0, [] (service::storage_service& ss) -> future<bool> {
+            if (!ss.is_topology_coordinator_enabled()) {
+                co_return false;
+            }
+            co_await ss.do_clusterwide_vnodes_cleanup();
+            co_return true;
+        });
+        if (done) {
             co_return json::json_return_type(0);
         }
-        // fall back to the local cleanup if local cleanup is requested
+        // fall back to the local cleanup if topology coordinator is not enabled or local cleanup is requested
         auto& db = ctx.db;
         auto& compaction_module = db.local().get_compaction_manager().get_task_manager_module();
         auto task = co_await compaction_module.make_and_start_task<compaction::global_cleanup_compaction_task_impl>({}, db);
@@ -797,7 +790,9 @@ rest_cleanup_all(http_context& ctx, sharded<service::storage_service>& ss, std::
 
         // Mark this node as clean
         co_await ss.invoke_on(0, [] (service::storage_service& ss) -> future<> {
-            co_await ss.reset_cleanup_needed();
+            if (ss.is_topology_coordinator_enabled()) {
+                co_await ss.reset_cleanup_needed();
+            }
         });
 
         co_return json::json_return_type(0);
@@ -808,6 +803,9 @@ future<json::json_return_type>
 rest_reset_cleanup_needed(http_context& ctx, sharded<service::storage_service>& ss, std::unique_ptr<http::request> req) {
         apilog.info("reset_cleanup_needed");
         co_await ss.invoke_on(0, [] (service::storage_service& ss) {
+            if (!ss.is_topology_coordinator_enabled()) {
+                throw std::runtime_error("mark_node_as_clean is only supported when topology over raft is enabled");
+            }
             return ss.reset_cleanup_needed();
         });
         co_return json_void();
@@ -1314,7 +1312,10 @@ rest_get_ownership(http_context& ctx, sharded<service::storage_service>& ss, std
             throw httpd::bad_param_exception("storage_service/ownership cannot be used when a keyspace uses tablets");
         }
 
-        co_return json::json_return_type(stream_range_as_array(co_await ss.local().get_ownership(), &map_to_json<gms::inet_address, float>));
+        return ss.local().get_ownership().then([] (auto&& ownership) {
+            std::vector<storage_service_json::mapper> res;
+            return make_ready_future<json::json_return_type>(map_to_key_value(ownership, res));
+        });
 }
 
 static
@@ -1331,7 +1332,10 @@ rest_get_effective_ownership(http_context& ctx, sharded<service::storage_service
             }
         }
 
-        co_return json::json_return_type(stream_range_as_array(co_await ss.local().effective_ownership(keyspace_name, table_name), &map_to_json<gms::inet_address, float>));
+        return ss.local().effective_ownership(keyspace_name, table_name).then([] (auto&& ownership) {
+            std::vector<storage_service_json::mapper> res;
+            return make_ready_future<json::json_return_type>(map_to_key_value(ownership, res));
+        });
 }
 
 static
@@ -1341,7 +1345,7 @@ rest_estimate_compression_ratios(http_context& ctx, sharded<service::storage_ser
         apilog.warn("estimate_compression_ratios: called before the cluster feature was enabled");
         throw std::runtime_error("estimate_compression_ratios requires all nodes to support the SSTABLE_COMPRESSION_DICTS cluster feature");
     }
-    auto ticket = co_await get_units(ss.local().get_do_sample_sstables_concurrency_limiter(), 1);
+    auto ticket = get_units(ss.local().get_do_sample_sstables_concurrency_limiter(), 1);
     auto ks = api::req_param<sstring>(*req, "keyspace", {}).value;
     auto cf = api::req_param<sstring>(*req, "cf", {}).value;
     apilog.debug("estimate_compression_ratios: called with ks={} cf={}", ks, cf);
@@ -1407,7 +1411,7 @@ rest_retrain_dict(http_context& ctx, sharded<service::storage_service>& ss, serv
         apilog.warn("retrain_dict: called before the cluster feature was enabled");
         throw std::runtime_error("retrain_dict requires all nodes to support the SSTABLE_COMPRESSION_DICTS cluster feature");
     }
-    auto ticket = co_await get_units(ss.local().get_do_sample_sstables_concurrency_limiter(), 1);
+    auto ticket = get_units(ss.local().get_do_sample_sstables_concurrency_limiter(), 1);
     auto ks = api::req_param<sstring>(*req, "keyspace", {}).value;
     auto cf = api::req_param<sstring>(*req, "cf", {}).value;
     apilog.debug("retrain_dict: called with ks={} cf={}", ks, cf);
@@ -1565,7 +1569,16 @@ rest_reload_raft_topology_state(sharded<service::storage_service>& ss, service::
 static
 future<json::json_return_type>
 rest_upgrade_to_raft_topology(sharded<service::storage_service>& ss, std::unique_ptr<http::request> req) {
-        apilog.info("Requested to schedule upgrade to raft topology, but this version does not need it since it uses raft topology by default.");
+        apilog.info("Requested to schedule upgrade to raft topology");
+        try {
+            co_await ss.invoke_on(0, [] (auto& ss) {
+                return ss.start_upgrade_to_raft_topology();
+            });
+        } catch (...) {
+            auto ex = std::current_exception();
+            apilog.error("Failed to schedule upgrade to raft topology: {}", ex);
+            std::rethrow_exception(std::move(ex));
+        }
         co_return json_void();
 }
 
@@ -2007,10 +2020,10 @@ void set_snapshot(http_context& ctx, routes& r, sharded<db::snapshot_ctl>& snap_
         auto tag = req->get_query_param("tag");
         auto column_families = split(req->get_query_param("cf"), ",");
         auto sfopt = req->get_query_param("sf");
-        auto tcopt = req->get_query_param("tc");
-
+        auto usiopt = req->get_query_param("use_sstable_identifier");
         db::snapshot_options opts = {
             .skip_flush = strcasecmp(sfopt.c_str(), "true") == 0,
+            .use_sstable_identifier = strcasecmp(usiopt.c_str(), "true") == 0
         };
 
         std::vector<sstring> keynames = split(req->get_query_param("kn"), ",");
@@ -2033,27 +2046,6 @@ void set_snapshot(http_context& ctx, routes& r, sharded<db::snapshot_ctl>& snap_
         }
     });
 
-    ss::take_cluster_snapshot.set(r, [&snap_ctl](std::unique_ptr<http::request> req) -> future<json::json_return_type> {
-        apilog.info("take_cluster_snapshot: {}", req->get_query_params());
-        auto tag = req->get_query_param("tag");
-        auto column_families = split(req->get_query_param("table"), ",");
-        // Note: not published/active. Retain as internal option, but...
-        auto sfopt = req->get_query_param("skip_flush");
-
-        db::snapshot_options opts = {
-            .skip_flush = strcasecmp(sfopt.c_str(), "true") == 0,
-        };
-
-        std::vector<sstring> keynames = split(req->get_query_param("keyspace"), ",");
-        try {
-            co_await snap_ctl.local().take_cluster_column_family_snapshot(keynames, column_families, tag, opts);
-            co_return json_void();
-        } catch (...) {
-            apilog.error("take_cluster_snapshot failed: {}", std::current_exception());
-            throw;
-        }
-    });
-
     ss::del_snapshot.set(r, [&snap_ctl](std::unique_ptr<http::request> req) -> future<json::json_return_type> {
         apilog.info("del_snapshot: {}", req->get_query_params());
         auto tag = req->get_query_param("tag");
@@ -2080,7 +2072,7 @@ void set_snapshot(http_context& ctx, routes& r, sharded<db::snapshot_ctl>& snap_
         auto info = parse_scrub_options(ctx, std::move(req));
 
         if (!info.snapshot_tag.empty()) {
-            db::snapshot_options opts = {.skip_flush = false};
+            db::snapshot_options opts = {.skip_flush = false, .use_sstable_identifier = false};
             co_await snap_ctl.local().take_column_family_snapshot(info.keyspace, info.column_families, info.snapshot_tag, opts);
         }
 

api/task_manager.cc

@@ -9,7 +9,6 @@
 #include <seastar/core/chunked_fifo.hh>
 #include <seastar/core/coroutine.hh>
 #include <seastar/coroutine/exception.hh>
-#include <seastar/coroutine/maybe_yield.hh>
 #include <seastar/http/exception.hh>
 
 #include "task_manager.hh"
@@ -265,7 +264,7 @@ void set_task_manager(http_context& ctx, routes& r, sharded<tasks::task_manager>
                 if (id) {
                     module->unregister_task(id);
                 }
-                co_await coroutine::maybe_yield();
+                co_await maybe_yield();
             }
         });
         co_return json_void();

api/tasks.cc

@@ -146,7 +146,7 @@ void set_tasks_compaction_module(http_context& ctx, routes& r, sharded<service::
         auto info = parse_scrub_options(ctx, std::move(req));
 
         if (!info.snapshot_tag.empty()) {
-            db::snapshot_options opts = {.skip_flush = false};
+            db::snapshot_options opts = {.skip_flush = false, .use_sstable_identifier = false};
             co_await snap_ctl.local().take_column_family_snapshot(info.keyspace, info.column_families, info.snapshot_tag, opts);
         }
 

audit/audit.cc

@@ -209,11 +209,15 @@ future<> audit::stop_audit() {
     });
 }
 
-audit_info_ptr audit::create_audit_info(statement_category cat, const sstring& keyspace, const sstring& table, bool batch) {
+audit_info_ptr audit::create_audit_info(statement_category cat, const sstring& keyspace, const sstring& table) {
     if (!audit_instance().local_is_initialized()) {
         return nullptr;
     }
-    return std::make_unique<audit_info>(cat, keyspace, table, batch);
+    return std::make_unique<audit_info>(cat, keyspace, table);
+}
+
+audit_info_ptr audit::create_no_audit_info() {
+    return audit_info_ptr();
 }
 
 future<> audit::start(const db::config& cfg) {
@@ -263,21 +267,18 @@ future<> audit::log_login(const sstring& username, socket_address client_ip, boo
 }
 
 future<> inspect(shared_ptr<cql3::cql_statement> statement, service::query_state& query_state, const cql3::query_options& options, bool error) {
-    auto audit_info = statement->get_audit_info();
-    if (!audit_info) {
-        return make_ready_future<>();
-    }
-    if (audit_info->batch()) {
-        cql3::statements::batch_statement* batch = static_cast<cql3::statements::batch_statement*>(statement.get());
+    cql3::statements::batch_statement* batch = dynamic_cast<cql3::statements::batch_statement*>(statement.get());
+    if (batch != nullptr) {
         return do_for_each(batch->statements().begin(), batch->statements().end(), [&query_state, &options, error] (auto&& m) {
             return inspect(m.statement, query_state, options, error);
         });
     } else {
-        if (audit::local_audit_instance().should_log(audit_info)) {
+        auto audit_info = statement->get_audit_info();
+        if (bool(audit_info) && audit::local_audit_instance().should_log(audit_info)) {
             return audit::local_audit_instance().log(audit_info, query_state, options, error);
         }
-        return make_ready_future<>();
     }
+    return make_ready_future<>();
 }
 
 future<> inspect_login(const sstring& username, socket_address client_ip, bool error) {

audit/audit.hh

@@ -75,13 +75,11 @@ class audit_info final {
     sstring _keyspace;
     sstring _table;
     sstring _query;
-    bool _batch;
 public:
-    audit_info(statement_category cat, sstring keyspace, sstring table, bool batch)
+    audit_info(statement_category cat, sstring keyspace, sstring table)
         : _category(cat)
         , _keyspace(std::move(keyspace))
         , _table(std::move(table))
-        , _batch(batch)
     { }
     void set_query_string(const std::string_view& query_string) {
         _query = sstring(query_string);
@@ -91,7 +89,6 @@ public:
     const sstring& query() const { return _query; }
     sstring category_string() const;
     statement_category category() const { return _category; }
-    bool batch() const { return _batch; }
 };
 
 using audit_info_ptr = std::unique_ptr<audit_info>;
@@ -129,7 +126,8 @@ public:
     }
     static future<> start_audit(const db::config& cfg, sharded<locator::shared_token_metadata>& stm, sharded<cql3::query_processor>& qp, sharded<service::migration_manager>& mm);
     static future<> stop_audit();
-    static audit_info_ptr create_audit_info(statement_category cat, const sstring& keyspace, const sstring& table, bool batch = false);
+    static audit_info_ptr create_audit_info(statement_category cat, const sstring& keyspace, const sstring& table);
+    static audit_info_ptr create_no_audit_info();
     audit(locator::shared_token_metadata& stm,
           cql3::query_processor& qp,
           service::migration_manager& mm,

audit/audit_syslog_storage_helper.cc

@@ -53,10 +53,10 @@ static std::string json_escape(std::string_view str) {
 
 }
 
-future<> audit_syslog_storage_helper::syslog_send_helper(temporary_buffer<char> msg) {
+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, std::span(&msg, 1));
+        co_await _sender.send(_syslog_address, net::packet{msg.data(), msg.size()});
     }
     catch (const std::exception& e) {
         auto error_msg = seastar::format(
@@ -90,7 +90,7 @@ future<> audit_syslog_storage_helper::start(const db::config& cfg) {
         co_return;
     }
 
-    co_await syslog_send_helper(temporary_buffer<char>::copy_of("Initializing syslog audit backend."));
+    co_await syslog_send_helper("Initializing syslog audit backend.");
 }
 
 future<> audit_syslog_storage_helper::stop() {
@@ -120,7 +120,7 @@ future<> audit_syslog_storage_helper::write(const audit_info* audit_info,
                                     audit_info->table(),
                                     username);
 
-    co_await syslog_send_helper(std::move(msg).release());
+    co_await syslog_send_helper(msg);
 }
 
 future<> audit_syslog_storage_helper::write_login(const sstring& username,
@@ -139,7 +139,7 @@ future<> audit_syslog_storage_helper::write_login(const sstring& username,
                                     client_ip,
                                     username);
 
-    co_await syslog_send_helper(std::move(msg).release());
+    co_await syslog_send_helper(msg.c_str());
 }
 
 }

audit/audit_syslog_storage_helper.hh

@@ -26,7 +26,7 @@ class audit_syslog_storage_helper : public storage_helper {
     net::datagram_channel _sender;
     seastar::semaphore _semaphore;
 
-    future<> syslog_send_helper(seastar::temporary_buffer<char> msg);
+    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/CMakeLists.txt

@@ -17,6 +17,7 @@ target_sources(scylla_auth
     password_authenticator.cc
     passwords.cc
     permission.cc
+    permissions_cache.cc
     resource.cc
     role_or_anonymous.cc
     roles-metadata.cc

auth/allow_all_authenticator.cc

@@ -9,6 +9,7 @@
 #include "auth/allow_all_authenticator.hh"
 
 #include "service/migration_manager.hh"
+#include "utils/alien_worker.hh"
 #include "utils/class_registrator.hh"
 
 namespace auth {
@@ -22,6 +23,7 @@ static const class_registrator<
         cql3::query_processor&,
         ::service::raft_group0_client&,
         ::service::migration_manager&,
-        cache&> registration("org.apache.cassandra.auth.AllowAllAuthenticator");
+        cache&,
+        utils::alien_worker&> registration("org.apache.cassandra.auth.AllowAllAuthenticator");
 
 }

auth/allow_all_authenticator.hh

@@ -14,6 +14,7 @@
 #include "auth/authenticator.hh"
 #include "auth/cache.hh"
 #include "auth/common.hh"
+#include "utils/alien_worker.hh"
 
 namespace cql3 {
 class query_processor;
@@ -29,7 +30,7 @@ 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&, cache&) {
+    allow_all_authenticator(cql3::query_processor&, ::service::raft_group0_client&, ::service::migration_manager&, cache&, utils::alien_worker&) {
     }
 
     virtual future<> start() override {

auth/cache.cc

@@ -8,7 +8,6 @@
 
 #include "auth/cache.hh"
 #include "auth/common.hh"
-#include "auth/role_or_anonymous.hh"
 #include "auth/roles-metadata.hh"
 #include "cql3/query_processor.hh"
 #include "cql3/untyped_result_set.hh"
@@ -16,35 +15,16 @@
 #include "db/system_keyspace.hh"
 #include "schema/schema.hh"
 #include <iterator>
-#include <seastar/core/abort_source.hh>
 #include <seastar/coroutine/maybe_yield.hh>
 #include <seastar/core/format.hh>
-#include <seastar/core/metrics.hh>
-#include <seastar/core/do_with.hh>
 
 namespace auth {
 
 logging::logger logger("auth-cache");
 
-cache::cache(cql3::query_processor& qp, abort_source& as) noexcept
+cache::cache(cql3::query_processor& qp) noexcept
     : _current_version(0)
-    , _qp(qp)
-    , _loading_sem(1)
-    , _as(as)
-    , _permission_loader(nullptr)
-    , _permission_loader_sem(8) {
-    namespace sm = seastar::metrics;
-    _metrics.add_group("auth_cache", {
-        sm::make_gauge("roles", [this] { return _roles.size(); },
-                sm::description("Number of roles currently cached")),
-        sm::make_gauge("permissions", [this] {
-            return _cached_permissions_count;
-        }, sm::description("Total number of permission sets currently cached across all roles"))
-    });
-}
-
-void cache::set_permission_loader(permission_loader_func loader) {
-    _permission_loader = std::move(loader);
+    , _qp(qp) {
 }
 
 lw_shared_ptr<const cache::role_record> cache::get(const role_name_t& role) const noexcept {
@@ -55,83 +35,6 @@ lw_shared_ptr<const cache::role_record> cache::get(const role_name_t& role) cons
     return it->second;
 }
 
-future<permission_set> cache::get_permissions(const role_or_anonymous& role, const resource& r) {
-    std::unordered_map<resource, permission_set>* perms_cache;
-    lw_shared_ptr<role_record> role_ptr;
-
-    if (is_anonymous(role)) {
-        perms_cache = &_anonymous_permissions;
-    } else {
-        const auto& role_name = *role.name;
-        auto role_it = _roles.find(role_name);
-        if (role_it == _roles.end()) {
-            // Role might have been deleted but there are some connections
-            // left which reference it. They should no longer have access to anything.
-            return make_ready_future<permission_set>(permissions::NONE);
-        }
-        role_ptr = role_it->second;
-        perms_cache = &role_ptr->cached_permissions;
-    }
-
-    if (auto it = perms_cache->find(r); it != perms_cache->end()) {
-        return make_ready_future<permission_set>(it->second);
-    }
-    // keep alive role_ptr as it holds perms_cache (except anonymous)
-    return do_with(std::move(role_ptr), [this, &role, &r, perms_cache] (auto& role_ptr) {
-        return load_permissions(role, r, perms_cache);
-    });
-}
-
-future<permission_set> cache::load_permissions(const role_or_anonymous& role, const resource& r, std::unordered_map<resource, permission_set>* perms_cache) {
-    SCYLLA_ASSERT(_permission_loader);
-    auto units = co_await get_units(_permission_loader_sem, 1, _as);
-
-    // Check again, perhaps we were blocked and other call loaded
-    // the permissions already. This is a protection against misses storm.
-    if (auto it = perms_cache->find(r); it != perms_cache->end()) {
-        co_return it->second;
-    }
-    auto perms = co_await _permission_loader(role, r);
-    add_permissions(*perms_cache, r, perms);
-    co_return perms;
-}
-
-future<> cache::prune(const resource& r) {
-    auto units = co_await get_units(_loading_sem, 1, _as);
-    _anonymous_permissions.erase(r);
-    for (auto& it : _roles) {
-        // Prunning can run concurrently with other functions but it
-        // can only cause cached_permissions extra reload via get_permissions.
-        remove_permissions(it.second->cached_permissions, r);
-        co_await coroutine::maybe_yield();
-    }
-}
-
-future<> cache::reload_all_permissions() noexcept {
-    SCYLLA_ASSERT(_permission_loader);
-    auto units = co_await get_units(_loading_sem, 1, _as);
-    auto copy_keys = [] (const std::unordered_map<resource, permission_set>& m) {
-        std::vector<resource> keys;
-        keys.reserve(m.size());
-        for (const auto& [res, _] : m) {
-            keys.push_back(res);
-        }
-        return keys;
-    };
-    const role_or_anonymous anon;
-    for (const auto& res : copy_keys(_anonymous_permissions)) {
-        _anonymous_permissions[res] = co_await _permission_loader(anon, res);
-    }
-    for (auto& [role, entry] : _roles) {
-        auto& perms_cache = entry->cached_permissions;
-        auto r = role_or_anonymous(role);
-        for (const auto& res : copy_keys(perms_cache)) {
-            perms_cache[res] = co_await _permission_loader(r, res);
-        }
-    }
-    logger.debug("Reloaded auth cache with {} entries", _roles.size());
-}
-
 future<lw_shared_ptr<cache::role_record>> cache::fetch_role(const role_name_t& role) const {
     auto rec = make_lw_shared<role_record>();
     rec->version = _current_version;
@@ -199,7 +102,7 @@ future<lw_shared_ptr<cache::role_record>> cache::fetch_role(const role_name_t& r
 future<> cache::prune_all() noexcept {
     for (auto it = _roles.begin(); it != _roles.end(); ) {
         if (it->second->version != _current_version) {
-            remove_role(it++);
+            _roles.erase(it++);
             co_await coroutine::maybe_yield();
         } else {
             ++it;
@@ -213,8 +116,6 @@ future<> cache::load_all() {
         co_return;
     }
     SCYLLA_ASSERT(this_shard_id() == 0);
-    auto units = co_await get_units(_loading_sem, 1, _as);
-
     ++_current_version;
 
     logger.info("Loading all roles");
@@ -223,7 +124,7 @@ future<> cache::load_all() {
         const auto name = r.get_as<sstring>("role");
         auto role = co_await fetch_role(name);
         if (role) {
-            add_role(name, role);
+            _roles[name] = role;
         }
         co_return stop_iteration::no;
     };
@@ -236,74 +137,36 @@ future<> cache::load_all() {
         co_await distribute_role(name, role);
     }
     co_await container().invoke_on_others([this](cache& c) -> future<> {
-        auto units = co_await get_units(c._loading_sem, 1, c._as);
         c._current_version = _current_version;
         co_await c.prune_all();
     });
 }
 
-future<> cache::gather_inheriting_roles(std::unordered_set<role_name_t>& roles, lw_shared_ptr<cache::role_record> role, const role_name_t& name) {
-    if (!role) {
-        // Role might have been removed or not yet added, either way
-        // their members will be handled by another top call to this function.
-        co_return;
-    }
-    for (const auto& member_name : role->members) {
-        bool is_new = roles.insert(member_name).second;
-        if (!is_new) {
-            continue;
-        }
-        lw_shared_ptr<cache::role_record> member_role;
-        auto r = _roles.find(member_name);
-        if (r != _roles.end()) {
-            member_role = r->second;
-        }
-        co_await gather_inheriting_roles(roles, member_role, member_name);
-    }
-}
-
 future<> cache::load_roles(std::unordered_set<role_name_t> roles) {
     if (legacy_mode(_qp)) {
         co_return;
     }
-    SCYLLA_ASSERT(this_shard_id() == 0);
-    auto units = co_await get_units(_loading_sem, 1, _as);
-
-    std::unordered_set<role_name_t> roles_to_clear_perms;
     for (const auto& name : roles) {
         logger.info("Loading role {}", name);
         auto role = co_await fetch_role(name);
          if (role) {
-            add_role(name, role);
-            co_await gather_inheriting_roles(roles_to_clear_perms, role, name);
+            _roles[name] = role;
         } else {
-            if (auto it = _roles.find(name); it != _roles.end()) {
-                auto old_role = it->second;
-                remove_role(it);
-                co_await gather_inheriting_roles(roles_to_clear_perms, old_role, name);
-            }
+            _roles.erase(name);
         }
         co_await distribute_role(name, role);
     }
-
-    co_await container().invoke_on_all([&roles_to_clear_perms] (cache& c) -> future<> {
-        for (const auto& name : roles_to_clear_perms) {
-            c.clear_role_permissions(name);
-            co_await coroutine::maybe_yield();
-        }
-    });
 }
 
 future<> cache::distribute_role(const role_name_t& name, lw_shared_ptr<role_record> role) {
     auto role_ptr = role.get();
-    co_await container().invoke_on_others([&name, role_ptr](cache& c) -> future<> {
-        auto units = co_await get_units(c._loading_sem, 1, c._as);
+    co_await container().invoke_on_others([&name, role_ptr](cache& c) {
         if (!role_ptr) {
-            c.remove_role(name);
-            co_return;
+            c._roles.erase(name);
+            return;
         }
         auto role_copy = make_lw_shared<role_record>(*role_ptr);
-        c.add_role(name, std::move(role_copy));
+        c._roles[name] = std::move(role_copy);
     });
 }
 
@@ -314,40 +177,4 @@ bool cache::includes_table(const table_id& id) noexcept {
             || id == db::system_keyspace::role_permissions()->id();
 }
 
-void cache::add_role(const role_name_t& name, lw_shared_ptr<role_record> role) {
-    if (auto it = _roles.find(name); it != _roles.end()) {
-        _cached_permissions_count -= it->second->cached_permissions.size();
-    }
-    _cached_permissions_count += role->cached_permissions.size();
-    _roles[name] = std::move(role);
-}
-
-void cache::remove_role(const role_name_t& name) {
-    if (auto it = _roles.find(name); it != _roles.end()) {
-        remove_role(it);
-    }
-}
-
-void cache::remove_role(roles_map::iterator it) {
-    _cached_permissions_count -= it->second->cached_permissions.size();
-    _roles.erase(it);
-}
-
-void cache::clear_role_permissions(const role_name_t& name) {
-    if (auto it = _roles.find(name); it != _roles.end()) {
-        _cached_permissions_count -= it->second->cached_permissions.size();
-        it->second->cached_permissions.clear();
-    }
-}
-
-void cache::add_permissions(std::unordered_map<resource, permission_set>& cache, const resource& r, permission_set perms) {
-    if (cache.emplace(r, perms).second) {
-        ++_cached_permissions_count;
-    }
-}
-
-void cache::remove_permissions(std::unordered_map<resource, permission_set>& cache, const resource& r) {
-    _cached_permissions_count -= cache.erase(r);
-}
-
 } // namespace auth

auth/cache.hh

@@ -8,7 +8,6 @@
 
 #pragma once
 
-#include <seastar/core/abort_source.hh>
 #include <unordered_set>
 #include <unordered_map>
 
@@ -16,15 +15,11 @@
 #include <seastar/core/future.hh>
 #include <seastar/core/sharded.hh>
 #include <seastar/core/shared_ptr.hh>
-#include <seastar/core/semaphore.hh>
-#include <seastar/core/metrics_registration.hh>
 
 #include <absl/container/flat_hash_map.h>
 
 #include "auth/permission.hh"
 #include "auth/common.hh"
-#include "auth/resource.hh"
-#include "auth/role_or_anonymous.hh"
 
 namespace cql3 { class query_processor; }
 
@@ -34,7 +29,6 @@ class cache : public peering_sharded_service<cache> {
 public:
     using role_name_t = sstring;
     using version_tag_t = char;
-    using permission_loader_func = std::function<future<permission_set>(const role_or_anonymous&, const resource&)>;
 
 	struct role_record {
         bool can_login = false;
@@ -44,19 +38,11 @@ public:
         sstring salted_hash;
         std::unordered_map<sstring, sstring> attributes;
         std::unordered_map<sstring, permission_set> permissions;
-    private:
-        friend cache;
-        // cached permissions include effects of role's inheritance
-        std::unordered_map<resource, permission_set> cached_permissions;
         version_tag_t version; // used for seamless cache reloads
     };
 
-    explicit cache(cql3::query_processor& qp, abort_source& as) noexcept;
+    explicit cache(cql3::query_processor& qp) noexcept;
     lw_shared_ptr<const role_record> get(const role_name_t& role) const noexcept;
-    void set_permission_loader(permission_loader_func loader);
-    future<permission_set> get_permissions(const role_or_anonymous& role, const resource& r);
-    future<> prune(const resource& r);
-    future<> reload_all_permissions() noexcept;
     future<> load_all();
     future<> load_roles(std::unordered_set<role_name_t> roles);
     static bool includes_table(const table_id&) noexcept;
@@ -64,31 +50,12 @@ public:
 private:
     using roles_map = absl::flat_hash_map<role_name_t, lw_shared_ptr<role_record>>;
     roles_map _roles;
-    // anonymous permissions map exists mainly due to compatibility with
-    // higher layers which use role_or_anonymous to get permissions.
-    std::unordered_map<resource, permission_set> _anonymous_permissions;
     version_tag_t _current_version;
     cql3::query_processor& _qp;
-    semaphore _loading_sem; // protects iteration of _roles map
-    abort_source& _as;
-    permission_loader_func _permission_loader;
-    semaphore _permission_loader_sem; // protects against reload storms on a single role change
-    metrics::metric_groups _metrics;
-    size_t _cached_permissions_count = 0;
 
     future<lw_shared_ptr<role_record>> fetch_role(const role_name_t& role) const;
     future<> prune_all() noexcept;
     future<> distribute_role(const role_name_t& name, const lw_shared_ptr<role_record> role);
-    future<> gather_inheriting_roles(std::unordered_set<role_name_t>& roles, lw_shared_ptr<cache::role_record> role, const role_name_t& name);
-
-    void add_role(const role_name_t& name, lw_shared_ptr<role_record> role);
-    void remove_role(const role_name_t& name);
-    void remove_role(roles_map::iterator it);
-    void clear_role_permissions(const role_name_t& name);
-    void add_permissions(std::unordered_map<resource, permission_set>& cache, const resource& r, permission_set perms);
-    void remove_permissions(std::unordered_map<resource, permission_set>& cache, const resource& r);
-
-    future<permission_set> load_permissions(const role_or_anonymous& role, const resource& r, std::unordered_map<resource, permission_set>* perms_cache);
 };
 
 } // namespace auth

auth/certificate_authenticator.cc

@@ -35,13 +35,14 @@ static const class_registrator<auth::authenticator
     , cql3::query_processor&
     , ::service::raft_group0_client&
     , ::service::migration_manager&
-    , auth::cache&> cert_auth_reg(CERT_AUTH_NAME);
+    , auth::cache&
+    , utils::alien_worker&> 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&, auth::cache&)
+auth::certificate_authenticator::certificate_authenticator(cql3::query_processor& qp, ::service::raft_group0_client&, ::service::migration_manager&, auth::cache&, utils::alien_worker&)
     : _queries([&] {
         auto& conf = qp.db().get_config();
         auto queries = conf.auth_certificate_role_queries();
@@ -76,9 +77,9 @@ auth::certificate_authenticator::certificate_authenticator(cql3::query_processor
                         throw std::invalid_argument(fmt::format("Invalid source: {}", map.at(cfg_source_attr)));
                     }
                     continue;
-                } catch (const std::out_of_range&) {
+                } catch (std::out_of_range&) {
                     // just fallthrough
-                } catch (const boost::regex_error&) {
+                } catch (boost::regex_error&) {
                     std::throw_with_nested(std::invalid_argument(fmt::format("Invalid query expression: {}", map.at(cfg_query_attr))));
                 }
             }

auth/certificate_authenticator.hh

@@ -10,6 +10,7 @@
 #pragma once
 
 #include "auth/authenticator.hh"
+#include "utils/alien_worker.hh"
 #include <boost/regex_fwd.hpp>  // IWYU pragma: keep
 
 namespace cql3 {
@@ -33,7 +34,7 @@ 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&, cache&);
+    certificate_authenticator(cql3::query_processor&, ::service::raft_group0_client&, ::service::migration_manager&, cache&, utils::alien_worker&);
     ~certificate_authenticator();
 
     future<> start() override;

auth/common.cc

@@ -94,7 +94,7 @@ static future<> create_legacy_metadata_table_if_missing_impl(
         try {
             co_return co_await mm.announce(co_await ::service::prepare_new_column_family_announcement(qp.proxy(), table, ts),
                     std::move(group0_guard), format("auth: create {} metadata table", table->cf_name()));
-        } catch (const exceptions::already_exists_exception&) {}
+        } catch (exceptions::already_exists_exception&) {}
     }
 }
 

auth/default_authorizer.cc

@@ -256,7 +256,7 @@ future<> default_authorizer::revoke_all(std::string_view role_name, ::service::g
         } else {
             co_await collect_mutations(_qp, mc, query, {sstring(role_name)});
         }
-    } catch (const exceptions::request_execution_exception& e) {
+    } catch (exceptions::request_execution_exception& e) {
         alogger.warn("CassandraAuthorizer failed to revoke all permissions of {}: {}", role_name, e);
     }
 }
@@ -293,13 +293,13 @@ future<> default_authorizer::revoke_all_legacy(const resource& resource) {
                                 [resource](auto ep) {
                     try {
                         std::rethrow_exception(ep);
-                    } catch (const exceptions::request_execution_exception& e) {
+                    } catch (exceptions::request_execution_exception& e) {
                         alogger.warn("CassandraAuthorizer failed to revoke all permissions on {}: {}", resource, e);
                     }
 
                 });
             });
-        } catch (const exceptions::request_execution_exception& e) {
+        } catch (exceptions::request_execution_exception& e) {
             alogger.warn("CassandraAuthorizer failed to revoke all permissions on {}: {}", resource, e);
             return make_ready_future();
         }

auth/ldap_role_manager.cc

@@ -88,16 +88,10 @@ static const class_registrator<
 
 ldap_role_manager::ldap_role_manager(
         std::string_view query_template, std::string_view target_attr, std::string_view bind_name, std::string_view bind_password,
-        uint32_t permissions_update_interval_in_ms,
-        utils::observer<uint32_t>  permissions_update_interval_in_ms_observer,
         cql3::query_processor& qp, ::service::raft_group0_client& rg0c, ::service::migration_manager& mm, cache& cache)
         : _std_mgr(qp, rg0c, mm, cache), _group0_client(rg0c), _query_template(query_template), _target_attr(target_attr), _bind_name(bind_name)
         , _bind_password(bind_password)
-        , _permissions_update_interval_in_ms(permissions_update_interval_in_ms)
-        , _permissions_update_interval_in_ms_observer(std::move(permissions_update_interval_in_ms_observer))
-        , _connection_factory(bind(std::mem_fn(&ldap_role_manager::reconnect), std::ref(*this)))
-        , _cache(cache)
-        , _cache_pruner(make_ready_future<>()) {
+        , _connection_factory(bind(std::mem_fn(&ldap_role_manager::reconnect), std::ref(*this))) {
 }
 
 ldap_role_manager::ldap_role_manager(cql3::query_processor& qp, ::service::raft_group0_client& rg0c, ::service::migration_manager& mm, cache& cache)
@@ -106,8 +100,6 @@ ldap_role_manager::ldap_role_manager(cql3::query_processor& qp, ::service::raft_
             qp.db().get_config().ldap_attr_role(),
             qp.db().get_config().ldap_bind_dn(),
             qp.db().get_config().ldap_bind_passwd(),
-            qp.db().get_config().permissions_update_interval_in_ms(),
-            qp.db().get_config().permissions_update_interval_in_ms.observe([this] (const uint32_t& v) { _permissions_update_interval_in_ms = v; }),
             qp,
             rg0c,
             mm,
@@ -127,22 +119,6 @@ future<> ldap_role_manager::start() {
         return make_exception_future(
                 std::runtime_error(fmt::format("error getting LDAP server address from template {}", _query_template)));
     }
-    _cache_pruner = futurize_invoke([this] () -> future<> {
-        while (true) {
-            try {
-                co_await seastar::sleep_abortable(std::chrono::milliseconds(_permissions_update_interval_in_ms), _as);
-            } catch (const seastar::sleep_aborted&) {
-                co_return; // ignore
-            }
-            co_await _cache.container().invoke_on_all([] (cache& c) -> future<> {
-                try {
-                    co_await c.reload_all_permissions();
-                } catch (...) {
-                    mylog.warn("Cache reload all permissions failed: {}", std::current_exception());
-                }
-            });
-        }
-    });
     return _std_mgr.start();
 }
 
@@ -199,11 +175,7 @@ future<conn_ptr> ldap_role_manager::reconnect() {
 
 future<> ldap_role_manager::stop() {
     _as.request_abort();
-    return std::move(_cache_pruner).then([this] {
-        return _std_mgr.stop();
-    }).then([this] {
-        return _connection_factory.stop();
-    });
+    return _std_mgr.stop().then([this] { return _connection_factory.stop(); });
 }
 
 future<> ldap_role_manager::create(std::string_view name, const role_config& config, ::service::group0_batch& mc) {

auth/ldap_role_manager.hh

@@ -10,7 +10,6 @@
 #pragma once
 
 #include <seastar/core/abort_source.hh>
-#include <seastar/core/future.hh>
 #include <stdexcept>
 
 #include "ent/ldap/ldap_connection.hh"
@@ -35,22 +34,14 @@ class ldap_role_manager : public role_manager {
     seastar::sstring _target_attr; ///< LDAP entry attribute containing the Scylla role name.
     seastar::sstring _bind_name; ///< Username for LDAP simple bind.
     seastar::sstring _bind_password; ///< Password for LDAP simple bind.
-
-    uint32_t _permissions_update_interval_in_ms;
-    utils::observer<uint32_t> _permissions_update_interval_in_ms_observer;
-
     mutable ldap_reuser _connection_factory; // Potentially modified by query_granted().
     seastar::abort_source _as;
-    cache& _cache;
-    seastar::future<> _cache_pruner;
   public:
     ldap_role_manager(
             std::string_view query_template, ///< LDAP query template as described in Scylla documentation.
             std::string_view target_attr, ///< LDAP entry attribute containing the Scylla role name.
             std::string_view bind_name, ///< LDAP bind credentials.
             std::string_view bind_password, ///< LDAP bind credentials.
-            uint32_t permissions_update_interval_in_ms,
-            utils::observer<uint32_t> permissions_update_interval_in_ms_observer,
             cql3::query_processor& qp, ///< Passed to standard_role_manager.
             ::service::raft_group0_client& rg0c, ///< Passed to standard_role_manager.
             ::service::migration_manager& mm, ///< Passed to standard_role_manager.

auth/password_authenticator.cc

@@ -49,7 +49,8 @@ static const class_registrator<
         cql3::query_processor&,
         ::service::raft_group0_client&,
         ::service::migration_manager&,
-        cache&> password_auth_reg("org.apache.cassandra.auth.PasswordAuthenticator");
+        cache&,
+        utils::alien_worker&> password_auth_reg("org.apache.cassandra.auth.PasswordAuthenticator");
 
 static thread_local auto rng_for_salt = std::default_random_engine(std::random_device{}());
 
@@ -63,13 +64,14 @@ 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, cache& cache)
+password_authenticator::password_authenticator(cql3::query_processor& qp, ::service::raft_group0_client& g0, ::service::migration_manager& mm, cache& cache, utils::alien_worker& hashing_worker)
     : _qp(qp)
     , _group0_client(g0)
     , _migration_manager(mm)
     , _cache(cache)
     , _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) {
@@ -328,18 +330,20 @@ future<authenticated_user> password_authenticator::authenticate(
             }
             salted_hash = role->salted_hash;
         }
-        const bool password_match = co_await passwords::check(password, *salted_hash);
+        const bool password_match = co_await _hashing_worker.submit<bool>([password = std::move(password), salted_hash] {
+            return passwords::check(password, *salted_hash);
+        });
         if (!password_match) {
             throw exceptions::authentication_exception("Username and/or password are incorrect");
         }
         co_return username;
-    } catch (const std::system_error &) {
+    } catch (std::system_error &) {
         std::throw_with_nested(exceptions::authentication_exception("Could not verify password"));
-    } catch (const exceptions::request_execution_exception& e) {
+    } catch (exceptions::request_execution_exception& e) {
         std::throw_with_nested(exceptions::authentication_exception(e.what()));
-    } catch (const exceptions::authentication_exception& e) {
+    } catch (exceptions::authentication_exception& e) {
         std::throw_with_nested(e);
-    } catch (const exceptions::unavailable_exception& e) {
+    } catch (exceptions::unavailable_exception& e) {
         std::throw_with_nested(exceptions::authentication_exception(e.get_message()));
     } catch (...) {
         std::throw_with_nested(exceptions::authentication_exception("authentication failed"));

auth/password_authenticator.hh

@@ -18,6 +18,7 @@
 #include "auth/passwords.hh"
 #include "auth/cache.hh"
 #include "service/raft/raft_group0_client.hh"
+#include "utils/alien_worker.hh"
 
 namespace db {
     class config;
@@ -48,12 +49,13 @@ class password_authenticator : public authenticator {
     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&, cache&);
+    password_authenticator(cql3::query_processor&, ::service::raft_group0_client&, ::service::migration_manager&, cache&, utils::alien_worker&);
 
     ~password_authenticator();
 

auth/passwords.cc

@@ -7,8 +7,6 @@
  */
 
 #include "auth/passwords.hh"
-#include "utils/crypt_sha512.hh"
-#include <seastar/core/coroutine.hh>
 
 #include <cerrno>
 
@@ -23,46 +21,25 @@ static thread_local crypt_data tlcrypt = {};
 
 namespace detail {
 
-void verify_hashing_output(const char * res) {
-    if (!res || (res[0] == '*')) {
-        throw std::system_error(errno, std::system_category());
-    }
-}
-
 void verify_scheme(scheme scheme) {
     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);
-    try {
-        verify_hashing_output(e);
-    } catch (const std::system_error& ex) {
-        throw no_supported_schemes();
+
+    if (e && (e[0] != '*')) {
+        return;
     }
+
+    throw no_supported_schemes();
 }
 
 sstring hash_with_salt(const sstring& pass, const sstring& salt) {
     auto res = crypt_r(pass.c_str(), salt.c_str(), &tlcrypt);
-    verify_hashing_output(res);
-    return res;
-}
-
-seastar::future<sstring> hash_with_salt_async(const sstring& pass, const sstring& salt) {
-    sstring res;
-    // Only SHA-512 hashes for passphrases shorter than 256 bytes can be computed using
-    // the __crypt_sha512 method. For other computations, we fall back to the
-    // crypt_r implementation from `<crypt.h>`, which can stall.
-    if (salt.starts_with(prefix_for_scheme(scheme::sha_512)) && pass.size() <= 255) {
-        char buf[128];
-        const char * output_ptr = co_await __crypt_sha512(pass.c_str(), salt.c_str(), buf);
-        verify_hashing_output(output_ptr);
-        res = output_ptr;
-    } else {
-        const char * output_ptr = crypt_r(pass.c_str(), salt.c_str(), &tlcrypt);
-        verify_hashing_output(output_ptr);
-        res = output_ptr;
+    if (!res || (res[0] == '*')) {
+        throw std::system_error(errno, std::system_category());
     }
-    co_return res;
+    return res;
 }
 
 std::string_view prefix_for_scheme(scheme c) noexcept {
@@ -81,9 +58,8 @@ no_supported_schemes::no_supported_schemes()
         : std::runtime_error("No allowed hashing schemes are supported on this system") {
 }
 
-seastar::future<bool> check(const sstring& pass, const sstring& salted_hash) {
-    const auto pwd_hash = co_await detail::hash_with_salt_async(pass, salted_hash);
-    co_return pwd_hash == salted_hash;
+bool check(const sstring& pass, const sstring& salted_hash) {
+    return detail::hash_with_salt(pass, salted_hash) == salted_hash;
 }
 
 } // namespace auth::passwords

auth/passwords.hh

@@ -11,7 +11,6 @@
 #include <random>
 #include <stdexcept>
 
-#include <seastar/core/future.hh>
 #include <seastar/core/sstring.hh>
 
 #include "seastarx.hh"
@@ -76,23 +75,11 @@ sstring generate_salt(RandomNumberEngine& g, scheme scheme) {
 
 ///
 /// Hash a password combined with an implementation-specific salt string.
-/// Deprecated in favor of `hash_with_salt_async`. This function is still used
-/// when generating password hashes for storage to ensure that
-/// `hash_with_salt` and `hash_with_salt_async` produce identical results,
-/// preserving backward compatibility.
 ///
 /// \throws \ref std::system_error when an unexpected implementation-specific error occurs.
 ///
 sstring hash_with_salt(const sstring& pass, const sstring& salt);
 
-///
-/// Async version of `hash_with_salt` that returns a future.
-/// If possible, hashing uses `coroutine::maybe_yield` to prevent reactor stalls.
-///
-/// \throws \ref std::system_error when an unexpected implementation-specific error occurs.
-///
-seastar::future<sstring> hash_with_salt_async(const sstring& pass, const sstring& salt);
-
 } // namespace detail
 
 ///
@@ -120,6 +107,6 @@ sstring hash(const sstring& pass, RandomNumberEngine& g, scheme scheme) {
 ///
 /// \throws \ref std::system_error when an unexpected implementation-specific error occurs.
 ///
-seastar::future<bool> check(const sstring& pass, const sstring& salted_hash);
+bool check(const sstring& pass, const sstring& salted_hash);
 
 } // namespace auth::passwords

auth/permissions_cache.cc

@@ -0,0 +1,38 @@
+/*
+ * Copyright (C) 2017-present ScyllaDB
+ */
+
+/*
+ * SPDX-License-Identifier: LicenseRef-ScyllaDB-Source-Available-1.0
+ */
+
+#include "auth/permissions_cache.hh"
+
+#include <fmt/ranges.h>
+#include "auth/authorizer.hh"
+#include "auth/service.hh"
+
+namespace auth {
+
+permissions_cache::permissions_cache(const utils::loading_cache_config& c, service& ser, logging::logger& log)
+        : _cache(c, log, [&ser, &log](const key_type& k) {
+              log.debug("Refreshing permissions for {}", k.first);
+              return ser.get_uncached_permissions(k.first, k.second);
+          }) {
+}
+
+bool permissions_cache::update_config(utils::loading_cache_config c) {
+    return _cache.update_config(std::move(c));
+}
+
+void permissions_cache::reset() {
+    _cache.reset();
+}
+
+future<permission_set> permissions_cache::get(const role_or_anonymous& maybe_role, const resource& r) {
+    return do_with(key_type(maybe_role, r), [this](const auto& k) {
+        return _cache.get(k);
+    });
+}
+
+}

auth/permissions_cache.hh

@@ -0,0 +1,66 @@
+/*
+ * Copyright (C) 2017-present ScyllaDB
+ */
+
+/*
+ * SPDX-License-Identifier: LicenseRef-ScyllaDB-Source-Available-1.0
+ */
+
+#pragma once
+
+#include <iostream>
+#include <utility>
+
+#include <fmt/core.h>
+#include <seastar/core/future.hh>
+
+#include "auth/permission.hh"
+#include "auth/resource.hh"
+#include "auth/role_or_anonymous.hh"
+#include "utils/log.hh"
+#include "utils/hash.hh"
+#include "utils/loading_cache.hh"
+
+namespace std {
+
+inline std::ostream& operator<<(std::ostream& os, const pair<auth::role_or_anonymous, auth::resource>& p) {
+    fmt::print(os, "{{role: {}, resource: {}}}", p.first, p.second);
+    return os;
+}
+
+}
+
+namespace db {
+class config;
+}
+
+namespace auth {
+
+class service;
+
+class permissions_cache final {
+    using cache_type = utils::loading_cache<
+            std::pair<role_or_anonymous, resource>,
+            permission_set,
+            1,
+            utils::loading_cache_reload_enabled::yes,
+            utils::simple_entry_size<permission_set>,
+            utils::tuple_hash>;
+
+    using key_type = typename cache_type::key_type;
+
+    cache_type _cache;
+
+public:
+    explicit permissions_cache(const utils::loading_cache_config&, service&, logging::logger&);
+
+    future <> stop() {
+        return _cache.stop();
+    }
+
+    bool update_config(utils::loading_cache_config);
+    void reset();
+    future<permission_set> get(const role_or_anonymous&, const resource&);
+};
+
+}

auth/saslauthd_authenticator.cc

@@ -35,9 +35,10 @@ static const class_registrator<
         cql3::query_processor&,
         ::service::raft_group0_client&,
         ::service::migration_manager&,
-        cache&> saslauthd_auth_reg("com.scylladb.auth.SaslauthdAuthenticator");
+        cache&,
+        utils::alien_worker&> saslauthd_auth_reg("com.scylladb.auth.SaslauthdAuthenticator");
 
-saslauthd_authenticator::saslauthd_authenticator(cql3::query_processor& qp, ::service::raft_group0_client&, ::service::migration_manager&, cache&)
+saslauthd_authenticator::saslauthd_authenticator(cql3::query_processor& qp, ::service::raft_group0_client&, ::service::migration_manager&, cache&, utils::alien_worker&)
     : _socket_path(qp.db().get_config().saslauthd_socket_path())
 {}
 

auth/saslauthd_authenticator.hh

@@ -12,6 +12,7 @@
 
 #include "auth/authenticator.hh"
 #include "auth/cache.hh"
+#include "utils/alien_worker.hh"
 
 namespace cql3 {
 class query_processor;
@@ -29,7 +30,7 @@ namespace auth {
 class saslauthd_authenticator : public authenticator {
     sstring _socket_path; ///< Path to the domain socket on which saslauthd is listening.
 public:
-    saslauthd_authenticator(cql3::query_processor&, ::service::raft_group0_client&, ::service::migration_manager&, cache&);
+    saslauthd_authenticator(cql3::query_processor&, ::service::raft_group0_client&, ::service::migration_manager&, cache&,utils::alien_worker&);
 
     future<> start() override;
 

auth/service.cc

@@ -64,11 +64,11 @@ static const sstring superuser_col_name("super");
 static logging::logger log("auth_service");
 
 class auth_migration_listener final : public ::service::migration_listener {
-    service& _service;
+    authorizer& _authorizer;
     cql3::query_processor& _qp;
 
 public:
-    explicit auth_migration_listener(service& s, cql3::query_processor& qp) : _service(s),  _qp(qp) {
+    explicit auth_migration_listener(authorizer& a, cql3::query_processor& qp) : _authorizer(a),  _qp(qp) {
     }
 
 private:
@@ -92,14 +92,14 @@ private:
             return;
         }
         // Do it in the background.
-        (void)do_with(auth::make_data_resource(ks_name), ::service::group0_batch::unused(), [this] (auto& r, auto& mc) mutable {
-            return _service.revoke_all(r, mc);
+        (void)do_with(::service::group0_batch::unused(), [this, &ks_name] (auto& mc) mutable {
+            return _authorizer.revoke_all(auth::make_data_resource(ks_name), mc);
         }).handle_exception([] (std::exception_ptr e) {
             log.error("Unexpected exception while revoking all permissions on dropped keyspace: {}", e);
         });
 
-        (void)do_with(auth::make_functions_resource(ks_name), ::service::group0_batch::unused(), [this] (auto& r, auto& mc) mutable {
-            return _service.revoke_all(r, mc);
+        (void)do_with(::service::group0_batch::unused(), [this, &ks_name] (auto& mc) mutable {
+            return _authorizer.revoke_all(auth::make_functions_resource(ks_name), mc);
         }).handle_exception([] (std::exception_ptr e) {
             log.error("Unexpected exception while revoking all permissions on functions in dropped keyspace: {}", e);
         });
@@ -111,8 +111,9 @@ private:
             return;
         }
         // Do it in the background.
-        (void)do_with(auth::make_data_resource(ks_name, cf_name), ::service::group0_batch::unused(), [this] (auto& r, auto& mc) mutable {
-            return _service.revoke_all(r, mc);
+        (void)do_with(::service::group0_batch::unused(), [this, &ks_name, &cf_name] (auto& mc) mutable {
+            return _authorizer.revoke_all(
+                    auth::make_data_resource(ks_name, cf_name), mc);
         }).handle_exception([] (std::exception_ptr e) {
             log.error("Unexpected exception while revoking all permissions on dropped table: {}", e);
         });
@@ -125,8 +126,9 @@ private:
             return;
         }
         // Do it in the background.
-        (void)do_with(auth::make_functions_resource(ks_name, function_name), ::service::group0_batch::unused(), [this] (auto& r, auto& mc) mutable {
-            return _service.revoke_all(r, mc);
+        (void)do_with(::service::group0_batch::unused(), [this, &ks_name, &function_name] (auto& mc) mutable {
+            return _authorizer.revoke_all(
+                    auth::make_functions_resource(ks_name, function_name), mc);
         }).handle_exception([] (std::exception_ptr e) {
             log.error("Unexpected exception while revoking all permissions on dropped function: {}", e);
         });
@@ -136,8 +138,9 @@ private:
             // in non legacy path revoke is part of schema change statement execution
             return;
         }
-        (void)do_with(auth::make_functions_resource(ks_name, aggregate_name), ::service::group0_batch::unused(), [this] (auto& r, auto& mc) mutable {
-            return _service.revoke_all(r, mc);
+        (void)do_with(::service::group0_batch::unused(), [this, &ks_name, &aggregate_name] (auto& mc) mutable {
+            return _authorizer.revoke_all(
+                    auth::make_functions_resource(ks_name, aggregate_name), mc);
         }).handle_exception([] (std::exception_ptr e) {
             log.error("Unexpected exception while revoking all permissions on dropped aggregate: {}", e);
         });
@@ -154,6 +157,7 @@ static future<> validate_role_exists(const service& ser, std::string_view role_n
 }
 
 service::service(
+        utils::loading_cache_config c,
         cache& cache,
         cql3::query_processor& qp,
         ::service::raft_group0_client& g0,
@@ -162,31 +166,41 @@ service::service(
         std::unique_ptr<authenticator> a,
         std::unique_ptr<role_manager> r,
         maintenance_socket_enabled used_by_maintenance_socket)
-            : _cache(cache)
+            : _loading_cache_config(std::move(c))
+            , _permissions_cache(nullptr)
+            , _cache(cache)
             , _qp(qp)
             , _group0_client(g0)
             , _mnotifier(mn)
             , _authorizer(std::move(z))
             , _authenticator(std::move(a))
             , _role_manager(std::move(r))
-            , _migration_listener(std::make_unique<auth_migration_listener>(*this, qp))
+            , _migration_listener(std::make_unique<auth_migration_listener>(*_authorizer, qp))
+            , _permissions_cache_cfg_cb([this] (uint32_t) { (void) _permissions_cache_config_action.trigger_later(); })
+            , _permissions_cache_config_action([this] { update_cache_config(); return make_ready_future<>(); })
+            , _permissions_cache_max_entries_observer(_qp.db().get_config().permissions_cache_max_entries.observe(_permissions_cache_cfg_cb))
+            , _permissions_cache_update_interval_in_ms_observer(_qp.db().get_config().permissions_update_interval_in_ms.observe(_permissions_cache_cfg_cb))
+            , _permissions_cache_validity_in_ms_observer(_qp.db().get_config().permissions_validity_in_ms.observe(_permissions_cache_cfg_cb))
             , _used_by_maintenance_socket(used_by_maintenance_socket) {}
 
 service::service(
+        utils::loading_cache_config c,
         cql3::query_processor& qp,
         ::service::raft_group0_client& g0,
         ::service::migration_notifier& mn,
         ::service::migration_manager& mm,
         const service_config& sc,
         maintenance_socket_enabled used_by_maintenance_socket,
-        cache& cache)
+        cache& cache,
+        utils::alien_worker& hashing_worker)
             : service(
+                      std::move(c),
                       cache,
                       qp,
                       g0,
                       mn,
                       create_object<authorizer>(sc.authorizer_java_name, qp, g0, mm),
-                      create_object<authenticator>(sc.authenticator_java_name, qp, g0, mm, cache),
+                      create_object<authenticator>(sc.authenticator_java_name, qp, g0, mm, cache, hashing_worker),
                       create_object<role_manager>(sc.role_manager_java_name, qp, g0, mm, cache),
                       used_by_maintenance_socket) {
 }
@@ -212,7 +226,7 @@ future<> service::create_legacy_keyspace_if_missing(::service::migration_manager
             try {
                 co_return co_await mm.announce(::service::prepare_new_keyspace_announcement(db.real_database(), ksm, ts),
                         std::move(group0_guard), seastar::format("auth_service: create {} keyspace", meta::legacy::AUTH_KS));
-            } catch (const ::service::group0_concurrent_modification&) {
+            } catch (::service::group0_concurrent_modification&) {
                 log.info("Concurrent operation is detected while creating {} keyspace, retrying.", meta::legacy::AUTH_KS);
             }
         }
@@ -244,14 +258,7 @@ future<> service::start(::service::migration_manager& mm, db::system_keyspace& s
         co_await _role_manager->ensure_superuser_is_created();
     }
     co_await when_all_succeed(_authorizer->start(), _authenticator->start()).discard_result();
-    if (!_used_by_maintenance_socket) {
-        // Maintenance socket mode can't cache permissions because it has
-        // different authorizer. We can't mix cached permissions, they could be
-        // different in normal mode.
-        _cache.set_permission_loader(std::bind(
-                &service::get_uncached_permissions,
-                this, std::placeholders::_1, std::placeholders::_2));
-    }
+    _permissions_cache = std::make_unique<permissions_cache>(_loading_cache_config, *this, log);
     co_await once_among_shards([this] {
         _mnotifier.register_listener(_migration_listener.get());
         return make_ready_future<>();
@@ -263,7 +270,9 @@ future<> service::stop() {
     // Only one of the shards has the listener registered, but let's try to
     // unregister on each one just to make sure.
     return _mnotifier.unregister_listener(_migration_listener.get()).then([this] {
-        _cache.set_permission_loader(nullptr);
+        if (_permissions_cache) {
+            return _permissions_cache->stop();
+        }
         return make_ready_future<>();
     }).then([this] {
         return when_all_succeed(_role_manager->stop(), _authorizer->stop(), _authenticator->stop()).discard_result();
@@ -275,8 +284,21 @@ future<> service::ensure_superuser_is_created() {
     co_await _authenticator->ensure_superuser_is_created();
 }
 
+void service::update_cache_config() {
+    auto db = _qp.db();
+
+    utils::loading_cache_config perm_cache_config;
+    perm_cache_config.max_size = db.get_config().permissions_cache_max_entries();
+    perm_cache_config.expiry = std::chrono::milliseconds(db.get_config().permissions_validity_in_ms());
+    perm_cache_config.refresh = std::chrono::milliseconds(db.get_config().permissions_update_interval_in_ms());
+
+    if (!_permissions_cache->update_config(std::move(perm_cache_config))) {
+        log.error("Failed to apply permissions cache changes. Please read the documentation of these parameters");
+    }
+}
 
 void service::reset_authorization_cache() {
+    _permissions_cache->reset();
     _qp.reset_cache();
 }
 
@@ -301,10 +323,7 @@ service::get_uncached_permissions(const role_or_anonymous& maybe_role, const res
 }
 
 future<permission_set> service::get_permissions(const role_or_anonymous& maybe_role, const resource& r) const {
-    if (legacy_mode(_qp) || _used_by_maintenance_socket) {
-        return get_uncached_permissions(maybe_role, r);
-    }
-    return _cache.get_permissions(maybe_role, r);
+    return _permissions_cache->get(maybe_role, r);
 }
 
 future<bool> service::has_superuser(std::string_view role_name, const role_set& roles) const {
@@ -429,11 +448,6 @@ future<bool> service::exists(const resource& r) const {
     return make_ready_future<bool>(false);
 }
 
-future<> service::revoke_all(const resource& r, ::service::group0_batch& mc) const {
-    co_await _authorizer->revoke_all(r, mc);
-    co_await _cache.prune(r);
-}
-
 future<std::vector<cql3::description>> service::describe_roles(bool with_hashed_passwords) {
     std::vector<cql3::description> result{};
 
@@ -788,7 +802,7 @@ future<> revoke_permissions(
 }
 
 future<> revoke_all(const service& ser, const resource& r, ::service::group0_batch& mc) {
-    return ser.revoke_all(r, mc);
+    return ser.underlying_authorizer().revoke_all(r, mc);
 }
 
 future<std::vector<permission_details>> list_filtered_permissions(
@@ -863,6 +877,13 @@ future<> migrate_to_auth_v2(db::system_keyspace& sys_ks, ::service::raft_group0_
                 continue; // some tables might not have been created if they were not used
             }
 
+            // use longer than usual timeout as we scan the whole table
+            // but not infinite or very long as we want to fail reasonably fast
+            const auto t = 5min;
+            const timeout_config tc{t, t, t, t, t, t, t};
+            ::service::client_state cs(::service::client_state::internal_tag{}, tc);
+            ::service::query_state qs(cs, empty_service_permit());
+
             std::vector<sstring> col_names;
             for (const auto& col : schema->all_columns()) {
                 col_names.push_back(col.name_as_cql_string());
@@ -872,51 +893,38 @@ future<> migrate_to_auth_v2(db::system_keyspace& sys_ks, ::service::raft_group0_
                 val_binders_str += ", ?";
             }
 
-            std::vector<mutation> collected;
-            // use longer than usual timeout as we scan the whole table
-            // but not infinite or very long as we want to fail reasonably fast
-            const auto t = 5min;
-            const timeout_config tc{t, t, t, t, t, t, t};
-            ::service::client_state cs(::service::client_state::internal_tag{}, tc);
-            ::service::query_state qs(cs, empty_service_permit());
-
             co_await qp.query_internal(
-                seastar::format("SELECT * FROM {}.{}", meta::legacy::AUTH_KS, cf_name),
-                db::consistency_level::ALL,
-                {},
-                1000,
-                [&qp, &cf_name, &col_names, &val_binders_str, &schema, ts, &collected] (const cql3::untyped_result_set::row& row) -> future<stop_iteration> {
-                    std::vector<data_value_or_unset> values;
-                    for (const auto& col : schema->all_columns()) {
-                        if (row.has(col.name_as_text())) {
-                            values.push_back(
-                                    col.type->deserialize(row.get_blob_unfragmented(col.name_as_text())));
-                        } else {
-                            values.push_back(unset_value{});
-                        }
+                    seastar::format("SELECT * FROM {}.{}", meta::legacy::AUTH_KS, cf_name),
+                    db::consistency_level::ALL,
+                    qs,
+                    {},
+                    1000,
+                    [&](const cql3::untyped_result_set_row& row) -> future<stop_iteration> {
+                std::vector<data_value_or_unset> values;
+                for (const auto& col : schema->all_columns()) {
+                    if (row.has(col.name_as_text())) {
+                        values.push_back(
+                                col.type->deserialize(row.get_blob_unfragmented(col.name_as_text())));
+                    } else {
+                        values.push_back(unset_value{});
                     }
-                    auto muts = co_await qp.get_mutations_internal(
-                            seastar::format("INSERT INTO {}.{} ({}) VALUES ({})",
-                                    db::system_keyspace::NAME,
-                                    cf_name,
-                                    fmt::join(col_names, ", "),
-                                    val_binders_str),
-                            internal_distributed_query_state(),
-                            ts,
-                            std::move(values));
-                    if (muts.size() != 1) {
-                        on_internal_error(log,
-                                format("expecting single insert mutation, got {}", muts.size()));
-                    }
-
-                    collected.push_back(std::move(muts[0]));
-                    co_return stop_iteration::no;
-                },
-                std::move(qs));
-
-            for (auto& m : collected) {
-                co_yield std::move(m);
-            }
+                }
+                auto muts = co_await qp.get_mutations_internal(
+                        seastar::format("INSERT INTO {}.{} ({}) VALUES ({})",
+                                db::system_keyspace::NAME,
+                                cf_name,
+                                fmt::join(col_names, ", "),
+                                val_binders_str),
+                        internal_distributed_query_state(),
+                        ts,
+                        std::move(values));
+                if (muts.size() != 1) {
+                    on_internal_error(log,
+                            format("expecting single insert mutation, got {}", muts.size()));
+                }
+                co_yield std::move(muts[0]);
+                co_return stop_iteration::no;
+            });
         }
         co_yield co_await sys_ks.make_auth_version_mutation(ts,
                 db::system_keyspace::auth_version_t::v2);
@@ -925,7 +933,7 @@ future<> migrate_to_auth_v2(db::system_keyspace& sys_ks, ::service::raft_group0_
             start_operation_func,
             std::move(gen),
             as,
-            std::nullopt);
+            get_raft_timeout());
 }
 
 }

auth/service.hh

@@ -20,12 +20,14 @@
 #include "auth/authenticator.hh"
 #include "auth/authorizer.hh"
 #include "auth/permission.hh"
+#include "auth/permissions_cache.hh"
 #include "auth/cache.hh"
 #include "auth/role_manager.hh"
 #include "auth/common.hh"
 #include "cql3/description.hh"
 #include "seastarx.hh"
 #include "service/raft/raft_group0_client.hh"
+#include "utils/alien_worker.hh"
 #include "utils/observable.hh"
 #include "utils/serialized_action.hh"
 #include "service/maintenance_mode.hh"
@@ -74,6 +76,8 @@ public:
 /// peering_sharded_service inheritance is needed to be able to access shard local authentication service
 /// given an object from another shard. Used for bouncing lwt requests to correct shard.
 class service final : public seastar::peering_sharded_service<service> {
+    utils::loading_cache_config _loading_cache_config;
+    std::unique_ptr<permissions_cache> _permissions_cache;
     cache& _cache;
 
     cql3::query_processor& _qp;
@@ -91,12 +95,20 @@ class service final : public seastar::peering_sharded_service<service> {
     // Only one of these should be registered, so we end up with some unused instances. Not the end of the world.
     std::unique_ptr<::service::migration_listener> _migration_listener;
 
+    std::function<void(uint32_t)> _permissions_cache_cfg_cb;
+    serialized_action _permissions_cache_config_action;
+
+    utils::observer<uint32_t> _permissions_cache_max_entries_observer;
+    utils::observer<uint32_t> _permissions_cache_update_interval_in_ms_observer;
+    utils::observer<uint32_t> _permissions_cache_validity_in_ms_observer;
+
     maintenance_socket_enabled _used_by_maintenance_socket;
 
     abort_source _as;
 
 public:
     service(
+            utils::loading_cache_config,
             cache& cache,
             cql3::query_processor&,
             ::service::raft_group0_client&,
@@ -112,13 +124,15 @@ public:
     /// of the instances themselves.
     ///
     service(
+            utils::loading_cache_config,
             cql3::query_processor&,
             ::service::raft_group0_client&,
             ::service::migration_notifier&,
             ::service::migration_manager&,
             const service_config&,
             maintenance_socket_enabled,
-            cache&);
+            cache&,
+            utils::alien_worker&);
 
     future<> start(::service::migration_manager&, db::system_keyspace&);
 
@@ -126,6 +140,8 @@ public:
 
     future<> ensure_superuser_is_created();
 
+    void update_cache_config();
+
     void reset_authorization_cache();
 
     ///
@@ -167,13 +183,6 @@ public:
 
     future<bool> exists(const resource&) const;
 
-    ///
-    /// Revoke all permissions granted to any role for a particular resource.
-    ///
-    /// \throws \ref unsupported_authorization_operation if revoking permissions is not supported.
-    ///
-    future<> revoke_all(const resource&, ::service::group0_batch&) const;
-
     ///
     /// Produces descriptions that can be used to restore the state of auth. That encompasses
     /// roles, role grants, and permission grants.

auth/standard_role_manager.cc

@@ -52,6 +52,13 @@ static const class_registrator<
         ::service::migration_manager&,
         cache&> registration("org.apache.cassandra.auth.CassandraRoleManager");
 
+struct record final {
+    sstring name;
+    bool is_superuser;
+    bool can_login;
+    role_set member_of;
+};
+
 static db::consistency_level consistency_for_role(std::string_view role_name) noexcept {
     if (role_name == meta::DEFAULT_SUPERUSER_NAME) {
         return db::consistency_level::QUORUM;
@@ -60,13 +67,13 @@ static db::consistency_level consistency_for_role(std::string_view role_name) no
     return db::consistency_level::LOCAL_ONE;
 }
 
-future<std::optional<standard_role_manager::record>> standard_role_manager::legacy_find_record(std::string_view role_name) {
+static future<std::optional<record>> find_record(cql3::query_processor& qp, std::string_view role_name) {
     const sstring query = seastar::format("SELECT * FROM {}.{} WHERE {} = ?",
-            get_auth_ks_name(_qp),
+            get_auth_ks_name(qp),
             meta::roles_table::name,
             meta::roles_table::role_col_name);
 
-    const auto results = co_await _qp.execute_internal(
+    const auto results = co_await qp.execute_internal(
             query,
             consistency_for_role(role_name),
             internal_distributed_query_state(),
@@ -86,25 +93,8 @@ future<std::optional<standard_role_manager::record>> standard_role_manager::lega
                         : role_set())});
 }
 
-future<std::optional<standard_role_manager::record>> standard_role_manager::find_record(std::string_view role_name) {
-    if (legacy_mode(_qp)) {
-        return legacy_find_record(role_name);
-    }
-    auto name = sstring(role_name);
-    auto role = _cache.get(name);
-    if (!role) {
-        return make_ready_future<std::optional<record>>(std::nullopt);
-    }
-    return make_ready_future<std::optional<record>>(std::make_optional(record{
-        .name = std::move(name),
-        .is_superuser = role->is_superuser,
-        .can_login = role->can_login,
-        .member_of = role->member_of
-    }));
-}
-
-future<standard_role_manager::record> standard_role_manager::require_record(std::string_view role_name) {
-    return find_record(role_name).then([role_name](std::optional<record> mr) {
+static future<record> require_record(cql3::query_processor& qp, std::string_view role_name) {
+    return find_record(qp, role_name).then([role_name](std::optional<record> mr) {
         if (!mr) {
             throw nonexistant_role(role_name);
         }
@@ -202,7 +192,7 @@ future<> standard_role_manager::legacy_create_default_role_if_missing() {
                 {_superuser},
                 cql3::query_processor::cache_internal::no).discard_result();
         log.info("Created default superuser role '{}'.", _superuser);
-    } catch (const exceptions::unavailable_exception& e) {
+    } catch(const exceptions::unavailable_exception& e) {
         log.warn("Skipped default role setup: some nodes were not ready; will retry");
         throw e;
     }
@@ -396,7 +386,7 @@ standard_role_manager::alter(std::string_view role_name, const role_config_updat
         return fmt::to_string(fmt::join(assignments, ", "));
     };
 
-    return require_record(role_name).then([this, role_name, &u, &mc](record) {
+    return require_record(_qp, role_name).then([this, role_name, &u, &mc](record) {
         if (!u.is_superuser && !u.can_login) {
             return make_ready_future<>();
         }
@@ -630,17 +620,18 @@ standard_role_manager::revoke(std::string_view revokee_name, std::string_view ro
     });
 }
 
-future<> standard_role_manager::collect_roles(
+static future<> collect_roles(
+        cql3::query_processor& qp,
         std::string_view grantee_name,
         bool recurse,
         role_set& roles) {
-    return require_record(grantee_name).then([this, &roles, recurse](standard_role_manager::record r) {
-        return do_with(std::move(r.member_of), [this, &roles, recurse](const role_set& memberships) {
-            return do_for_each(memberships.begin(), memberships.end(), [this, &roles, recurse](const sstring& role_name) {
+    return require_record(qp, grantee_name).then([&qp, &roles, recurse](record r) {
+        return do_with(std::move(r.member_of), [&qp, &roles, recurse](const role_set& memberships) {
+            return do_for_each(memberships.begin(), memberships.end(), [&qp, &roles, recurse](const sstring& role_name) {
                 roles.insert(role_name);
 
                 if (recurse) {
-                    return collect_roles(role_name, true, roles);
+                    return collect_roles(qp, role_name, true, roles);
                 }
 
                 return make_ready_future<>();
@@ -655,7 +646,7 @@ future<role_set> standard_role_manager::query_granted(std::string_view grantee_n
     return do_with(
             role_set{sstring(grantee_name)},
             [this, grantee_name, recurse](role_set& roles) {
-        return collect_roles(grantee_name, recurse, roles).then([&roles] { return roles; });
+        return collect_roles(_qp, grantee_name, recurse, roles).then([&roles] { return roles; });
     });
 }
 
@@ -715,21 +706,27 @@ future<role_set> standard_role_manager::query_all(::service::query_state& qs) {
 }
 
 future<bool> standard_role_manager::exists(std::string_view role_name) {
-    return find_record(role_name).then([](std::optional<record> mr) {
+    return find_record(_qp, role_name).then([](std::optional<record> mr) {
         return static_cast<bool>(mr);
     });
 }
 
 future<bool> standard_role_manager::is_superuser(std::string_view role_name) {
-    return require_record(role_name).then([](record r) {
+    return require_record(_qp, role_name).then([](record r) {
         return r.is_superuser;
     });
 }
 
 future<bool> standard_role_manager::can_login(std::string_view role_name) {
-    return require_record(role_name).then([](record r) {
-        return r.can_login;
-    });
+    if (legacy_mode(_qp)) {
+       const auto r = co_await require_record(_qp, role_name);
+       co_return r.can_login;
+    }
+    auto role = _cache.get(sstring(role_name));
+    if (!role) {
+        throw nonexistant_role(role_name);
+    }
+    co_return role->can_login;
 }
 
 future<std::optional<sstring>> standard_role_manager::get_attribute(std::string_view role_name, std::string_view attribute_name, ::service::query_state& qs) {

auth/standard_role_manager.hh

@@ -90,12 +90,6 @@ public:
 
 private:
     enum class membership_change { add, remove };
-    struct record final {
-        sstring name;
-        bool is_superuser;
-        bool can_login;
-        role_set member_of;
-    };
 
     future<> create_legacy_metadata_tables_if_missing() const;
 
@@ -113,14 +107,6 @@ private:
     future<> legacy_modify_membership(std::string_view role_name, std::string_view grantee_name, membership_change);
 
     future<> modify_membership(std::string_view role_name, std::string_view grantee_name, membership_change, ::service::group0_batch& mc);
-
-    future<std::optional<record>> legacy_find_record(std::string_view role_name);
-    future<std::optional<record>> find_record(std::string_view role_name);
-    future<record> require_record(std::string_view role_name);
-    future<> collect_roles(
-            std::string_view grantee_name,
-            bool recurse,
-            role_set& roles);
 };
 
 } // namespace auth

auth/transitional.cc

@@ -38,8 +38,8 @@ class transitional_authenticator : public authenticator {
 public:
     static const sstring PASSWORD_AUTHENTICATOR_NAME;
 
-    transitional_authenticator(cql3::query_processor& qp, ::service::raft_group0_client& g0, ::service::migration_manager& mm, cache& cache)
-            : transitional_authenticator(std::make_unique<password_authenticator>(qp, g0, mm, cache)) {
+    transitional_authenticator(cql3::query_processor& qp, ::service::raft_group0_client& g0, ::service::migration_manager& mm, cache& cache, utils::alien_worker& hashing_worker)
+            : transitional_authenticator(std::make_unique<password_authenticator>(qp, g0, mm, cache, hashing_worker)) {
     }
     transitional_authenticator(std::unique_ptr<authenticator> a)
             : _authenticator(std::move(a)) {
@@ -81,7 +81,7 @@ public:
         }).handle_exception([](auto ep) {
             try {
                 std::rethrow_exception(ep);
-            } catch (const exceptions::authentication_exception&) {
+            } catch (exceptions::authentication_exception&) {
                 // return anon user
                 return make_ready_future<authenticated_user>(anonymous_user());
             }
@@ -126,7 +126,7 @@ public:
             virtual bytes evaluate_response(bytes_view client_response) override {
                 try {
                     return _sasl->evaluate_response(client_response);
-                } catch (const exceptions::authentication_exception&) {
+                } catch (exceptions::authentication_exception&) {
                     _complete = true;
                     return {};
                 }
@@ -141,7 +141,7 @@ public:
                     return _sasl->get_authenticated_user().handle_exception([](auto ep) {
                         try {
                             std::rethrow_exception(ep);
-                        } catch (const exceptions::authentication_exception&) {
+                        } catch (exceptions::authentication_exception&) {
                             // return anon user
                             return make_ready_future<authenticated_user>(anonymous_user());
                         }
@@ -241,7 +241,8 @@ static const class_registrator<
         cql3::query_processor&,
         ::service::raft_group0_client&,
         ::service::migration_manager&,
-        auth::cache&> transitional_authenticator_reg(auth::PACKAGE_NAME + "TransitionalAuthenticator");
+        auth::cache&,
+        utils::alien_worker&> transitional_authenticator_reg(auth::PACKAGE_NAME + "TransitionalAuthenticator");
 
 static const class_registrator<
         auth::authorizer,

cdc/generation.cc

@@ -204,7 +204,7 @@ future<topology_description> topology_description::clone_async() const {
 
     for (const auto& entry : _entries) {
         vec.push_back(entry);
-        co_await coroutine::maybe_yield();
+        co_await seastar::maybe_yield();
     }
 
     co_return topology_description{std::move(vec)};
@@ -814,7 +814,8 @@ generation_service::generation_service(
             config cfg, gms::gossiper& g, sharded<db::system_distributed_keyspace>& sys_dist_ks,
             sharded<db::system_keyspace>& sys_ks,
             abort_source& abort_src, const locator::shared_token_metadata& stm, gms::feature_service& f,
-            replica::database& db)
+            replica::database& db,
+            std::function<bool()> raft_topology_change_enabled)
         : _cfg(std::move(cfg))
         , _gossiper(g)
         , _sys_dist_ks(sys_dist_ks)
@@ -823,6 +824,7 @@ generation_service::generation_service(
         , _token_metadata(stm)
         , _feature_service(f)
         , _db(db)
+        , _raft_topology_change_enabled(std::move(raft_topology_change_enabled))
 {
 }
 
@@ -876,7 +878,16 @@ future<> generation_service::on_join(gms::inet_address ep, locator::host_id id,
 future<> generation_service::on_change(gms::inet_address ep, locator::host_id id, const gms::application_state_map& states, gms::permit_id pid) {
     assert_shard_zero(__PRETTY_FUNCTION__);
 
-    return make_ready_future<>();
+    if (_raft_topology_change_enabled()) {
+        return make_ready_future<>();
+    }
+
+    return on_application_state_change(ep, id, states, gms::application_state::CDC_GENERATION_ID, pid, [this] (gms::inet_address ep, locator::host_id id, const gms::versioned_value& v, gms::permit_id) {
+        auto gen_id = gms::versioned_value::cdc_generation_id_from_string(v.value());
+        cdc_log.debug("Endpoint: {}, CDC generation ID change: {}", ep, gen_id);
+
+        return legacy_handle_cdc_generation(gen_id);
+    });
 }
 
 future<> generation_service::check_and_repair_cdc_streams() {

cdc/generation_service.hh

@@ -79,12 +79,17 @@ private:
     std::optional<cdc::generation_id> _gen_id;
     future<> _cdc_streams_rewrite_complete = make_ready_future<>();
 
+    /* Returns true if raft topology changes are enabled.
+     * Can only be called from shard 0.
+     */
+    std::function<bool()> _raft_topology_change_enabled;
 public:
     generation_service(config cfg, gms::gossiper&,
             sharded<db::system_distributed_keyspace>&,
             sharded<db::system_keyspace>& sys_ks,
             abort_source&, const locator::shared_token_metadata&,
-            gms::feature_service&, replica::database& db);
+            gms::feature_service&, replica::database& db,
+            std::function<bool()> raft_topology_change_enabled);
 
     future<> stop();
     ~generation_service();

cdc/split.cc

@@ -15,7 +15,7 @@
 #include "mutation/tombstone.hh"
 #include "schema/schema.hh"
 
-#include <seastar/core/sstring.hh>
+#include "seastar/core/sstring.hh"
 #include "types/concrete_types.hh"
 #include "types/types.hh"
 #include "types/user.hh"

client_data.hh

@@ -10,9 +10,7 @@
 #include <seastar/net/inet_address.hh>
 #include <seastar/core/sstring.hh>
 #include "seastarx.hh"
-#include "utils/loading_shared_values.hh"
 
-#include <list>
 #include <optional>
 
 enum class client_type {
@@ -29,20 +27,6 @@ enum class client_connection_stage {
     ready,
 };
 
-// We implement a keys cache using a map-like utils::loading_shared_values container by storing empty values.
-struct options_cache_value_type {};
-using client_options_cache_type = utils::loading_shared_values<sstring, options_cache_value_type>;
-using client_options_cache_entry_type = client_options_cache_type::entry_ptr;
-using client_options_cache_key_type = client_options_cache_type::key_type;
-
-// This struct represents a single OPTION key-value pair from the client's connection options.
-// Both key and value are represented by corresponding "references" to their cached values.
-// Each "reference" is effectively a lw_shared_ptr value.
-struct client_option_key_value_cached_entry {
-    client_options_cache_entry_type key;
-    client_options_cache_entry_type value;
-};
-
 sstring to_string(client_connection_stage ct);
 
 // Representation of a row in `system.clients'. std::optionals are for nullable cells.
@@ -53,8 +37,8 @@ struct client_data {
     client_connection_stage connection_stage = client_connection_stage::established;
     int32_t shard_id;  /// ID of server-side shard which is processing the connection.
 
-    std::optional<client_options_cache_entry_type> driver_name;
-    std::optional<client_options_cache_entry_type> driver_version;
+    std::optional<sstring> driver_name;
+    std::optional<sstring> driver_version;
     std::optional<sstring> hostname;
     std::optional<int32_t> protocol_version;
     std::optional<sstring> ssl_cipher_suite;
@@ -62,7 +46,6 @@ struct client_data {
     std::optional<sstring> ssl_protocol;
     std::optional<sstring> username;
     std::optional<sstring> scheduling_group_name;
-    std::list<client_option_key_value_cached_entry> client_options;
 
     sstring stage_str() const { return to_string(connection_stage); }
     sstring client_type_str() const { return to_string(ct); }

cmake/mode.common.cmake

@@ -125,6 +125,10 @@ if(target_arch)
   add_compile_options("-march=${target_arch}")
 endif()
 
+if(CMAKE_CXX_COMPILER_ID STREQUAL "Clang")
+  add_compile_options("SHELL:-Xclang -fexperimental-assignment-tracking=disabled")
+endif()
+
 function(maybe_limit_stack_usage_in_KB stack_usage_threshold_in_KB config)
   math(EXPR _stack_usage_threshold_in_bytes "${stack_usage_threshold_in_KB} * 1024")
   set(_stack_usage_threshold_flag "-Wstack-usage=${_stack_usage_threshold_in_bytes}")

compaction/compaction_group_view.hh

@@ -12,7 +12,6 @@
 #include <seastar/core/condition-variable.hh>
 
 #include "schema/schema_fwd.hh"
-#include "sstables/open_info.hh"
 #include "compaction_descriptor.hh"
 
 class reader_permit;
@@ -45,7 +44,7 @@ public:
     virtual compaction_strategy_state& get_compaction_strategy_state() noexcept = 0;
     virtual reader_permit make_compaction_reader_permit() const = 0;
     virtual sstables::sstables_manager& get_sstables_manager() noexcept = 0;
-    virtual sstables::shared_sstable make_sstable(sstables::sstable_state) const = 0;
+    virtual sstables::shared_sstable make_sstable() const = 0;
     virtual sstables::sstable_writer_config configure_writer(sstring origin) const = 0;
     virtual api::timestamp_type min_memtable_timestamp() const = 0;
     virtual api::timestamp_type min_memtable_live_timestamp() const = 0;

compaction/compaction_manager.cc

@@ -416,9 +416,7 @@ future<compaction_result> compaction_task_executor::compact_sstables(compaction_
         descriptor.enable_garbage_collection(co_await sstable_set_for_tombstone_gc(t));
     }
     descriptor.creator = [&t] (shard_id) {
-        // All compaction types going through this path will work on normal input sstables only.
-        // Off-strategy, for example, waits until the sstables move out of staging state.
-        return t.make_sstable(sstables::sstable_state::normal);
+        return t.make_sstable();
     };
     descriptor.replacer = [this, &t, &on_replace, offstrategy] (compaction_completion_desc desc) {
         t.get_compaction_strategy().notify_completion(t, desc.old_sstables, desc.new_sstables);
@@ -1519,9 +1517,7 @@ future<> compaction_manager::maybe_wait_for_sstable_count_reduction(compaction_g
             | std::views::transform(std::mem_fn(&sstables::sstable::run_identifier))
             | std::ranges::to<std::unordered_set>());
     };
-    const auto injected_threshold = utils::get_local_injector().inject_parameter<size_t>("set_sstable_count_reduction_threshold");
-    const auto threshold = injected_threshold.value_or(size_t(std::max(schema->max_compaction_threshold(), 32)));
-
+    const auto threshold = size_t(std::max(schema->max_compaction_threshold(), 32));
     auto count = co_await num_runs_for_compaction();
     if (count <= threshold) {
         cmlog.trace("No need to wait for sstable count reduction in {}: {} <= {}",
@@ -1536,7 +1532,9 @@ future<> compaction_manager::maybe_wait_for_sstable_count_reduction(compaction_g
     auto& cstate = get_compaction_state(&t);
     try {
         while (can_perform_regular_compaction(t) && co_await num_runs_for_compaction() > threshold) {
-            co_await cstate.compaction_done.when();
+            co_await cstate.compaction_done.wait([this, &t] {
+                return !can_perform_regular_compaction(t);
+            });
         }
     } catch (const broken_condition_variable&) {
         co_return;
@@ -1849,10 +1847,6 @@ protected:
                 throw make_compaction_stopped_exception();
             }
         }, false);
-        if (utils::get_local_injector().is_enabled("split_sstable_force_stop_exception")) {
-            throw make_compaction_stopped_exception();
-        }
-
         co_return co_await do_rewrite_sstable(std::move(sst));
     }
 };
@@ -2290,16 +2284,12 @@ future<compaction_manager::compaction_stats_opt> compaction_manager::perform_spl
 }
 
 future<std::vector<sstables::shared_sstable>>
-compaction_manager::maybe_split_new_sstable(sstables::shared_sstable sst, compaction_group_view& t, compaction_type_options::split opt) {
+compaction_manager::maybe_split_sstable(sstables::shared_sstable sst, compaction_group_view& t, compaction_type_options::split opt) {
     if (!split_compaction_task_executor::sstable_needs_split(sst, opt)) {
         co_return std::vector<sstables::shared_sstable>{sst};
     }
-    // Throw an error if split cannot be performed due to e.g. out of space prevention.
-    // We don't want to prevent split because compaction is temporarily disabled on a view only for synchronization,
-    // which is unneeded against new sstables that aren't part of any set yet, so never use can_proceed(&t) here.
-    if (is_disabled()) {
-        co_return coroutine::exception(std::make_exception_ptr(std::runtime_error(format("Cannot split {} because manager has compaction disabled, " \
-                                                                                         "reason might be out of space prevention", sst->get_filename()))));
+    if (!can_proceed(&t)) {
+        co_return std::vector<sstables::shared_sstable>{sst};
     }
     std::vector<sstables::shared_sstable> ret;
 
@@ -2307,11 +2297,8 @@ compaction_manager::maybe_split_new_sstable(sstables::shared_sstable sst, compac
     compaction_progress_monitor monitor;
     compaction_data info = create_compaction_data();
     compaction_descriptor desc = split_compaction_task_executor::make_descriptor(sst, opt);
-    desc.creator = [&t, sst] (shard_id _) {
-        // NOTE: preserves the sstable state, since we want the output to be on the same state as the original.
-        // For example, if base table has views, it's important that sstable produced by repair will be
-        // in the staging state.
-        return t.make_sstable(sst->state());
+    desc.creator = [&t] (shard_id _) {
+        return t.make_sstable();
     };
     desc.replacer = [&] (compaction_completion_desc d) {
         std::move(d.new_sstables.begin(), d.new_sstables.end(), std::back_inserter(ret));

compaction/compaction_manager.hh

@@ -376,8 +376,7 @@ public:
     // Splits a single SSTable by segregating all its data according to the classifier.
     // If SSTable doesn't need split, the same input SSTable is returned as output.
     // If SSTable needs split, then output SSTables are returned and the input SSTable is deleted.
-    // Exception is thrown if the input sstable cannot be split due to e.g. out of space prevention.
-    future<std::vector<sstables::shared_sstable>> maybe_split_new_sstable(sstables::shared_sstable sst, compaction_group_view& t, compaction_type_options::split opt);
+    future<std::vector<sstables::shared_sstable>> maybe_split_sstable(sstables::shared_sstable sst, compaction_group_view& t, compaction_type_options::split opt);
 
     // Run a custom job for a given table, defined by a function
     // it completes when future returned by job is ready or returns immediately

conf/scylla.yaml

@@ -299,11 +299,13 @@ batch_size_fail_threshold_in_kb: 1024
 # max_hint_window_in_ms: 10800000 # 3 hours
 
 
-# Validity period for authorized statements cache. Defaults to 10000, set to 0 to disable.
+# Validity period for permissions cache (fetching permissions can be an
+# expensive operation depending on the authorizer, CassandraAuthorizer is
+# one example). Defaults to 10000, set to 0 to disable.
 # Will be disabled automatically for AllowAllAuthorizer.
 # permissions_validity_in_ms: 10000
 
-# Refresh interval for authorized statements cache.
+# Refresh interval for permissions cache (if enabled).
 # After this interval, cache entries become eligible for refresh. Upon next
 # access, an async reload is scheduled and the old value returned until it
 # completes. If permissions_validity_in_ms is non-zero, then this also must have
@@ -564,16 +566,15 @@ commitlog_total_space_in_mb: -1
 # prometheus_address: 1.2.3.4
 
 # audit settings
-# Table audit is enabled by default.
+# By default, Scylla does not audit anything.
 # 'audit' config option controls if and where to output audited events:
-#   - "none": auditing is disabled
-#   - "table": save audited events in audit.audit_log column family (default)
+#   - "none": auditing is disabled (default)
+#   - "table": save audited events in audit.audit_log column family
 #   - "syslog": send audited events via syslog (depends on OS, but usually to /dev/log)
-audit: "table"
+# audit: "none"
 #
 # List of statement categories that should be audited.
-# Possible categories are: QUERY, DML, DCL, DDL, AUTH, ADMIN
-audit_categories: "DCL,AUTH,ADMIN"
+# audit_categories: "DCL,DDL,AUTH"
 #
 # List of tables that should be audited.
 # audit_tables: "<keyspace_name>.<table_name>,<keyspace_name>.<table_name>"
@@ -874,16 +875,7 @@ maintenance_socket: ignore
 # The `tablets` option cannot be changed using `ALTER KEYSPACE`.
 tablets_mode_for_new_keyspaces: enabled
 
-# Require every tablet-enabled keyspace to be RF-rack-valid.
-#
-# A tablet-enabled keyspace is RF-rack-valid when, for each data center,
-# its replication factor (RF) is 0, 1, or exactly equal to the number of
-# racks in that data center. Setting the RF to the number of racks ensures
-# that a single rack failure never results in data unavailability.
-#
-# When set to true, CREATE KEYSPACE and ALTER KEYSPACE statements that
-# would produce an RF-rack-invalid keyspace are rejected.
-# When set to false, such statements are allowed but emit a warning.
+# Enforce RF-rack-valid keyspaces.
 rf_rack_valid_keyspaces: false
 
 #

configure.py

@@ -368,87 +368,6 @@ def find_ninja():
     sys.exit(1)
 
 
-def find_compiler(name):
-    """
-    Find a compiler by name, skipping ccache wrapper directories.
-
-    This is useful when using sccache to avoid double-caching through ccache.
-
-    Args:
-        name: The compiler name (e.g., 'clang++', 'clang', 'gcc')
-
-    Returns:
-        Path to the compiler, skipping ccache directories, or None if not found.
-    """
-    ccache_dirs = {'/usr/lib/ccache', '/usr/lib64/ccache'}
-    for path_dir in os.environ.get('PATH', '').split(os.pathsep):
-        # Skip ccache wrapper directories
-        if os.path.realpath(path_dir) in ccache_dirs or path_dir in ccache_dirs:
-            continue
-        candidate = os.path.join(path_dir, name)
-        if os.path.isfile(candidate) and os.access(candidate, os.X_OK):
-            return candidate
-    return None
-
-
-def resolve_compilers_for_compiler_cache(args, compiler_cache):
-    """
-    When using a compiler cache, resolve compiler paths to avoid ccache directories.
-
-    This prevents double-caching when ccache symlinks are in PATH.
-
-    Args:
-        args: The argument namespace with cc and cxx attributes.
-        compiler_cache: Path to the compiler cache binary, or None.
-    """
-    if not compiler_cache:
-        return
-    if not os.path.isabs(args.cxx):
-        real_cxx = find_compiler(args.cxx)
-        if real_cxx:
-            args.cxx = real_cxx
-    if not os.path.isabs(args.cc):
-        real_cc = find_compiler(args.cc)
-        if real_cc:
-            args.cc = real_cc
-
-
-def find_compiler_cache(preference):
-    """
-    Find a compiler cache based on the preference.
-
-    Args:
-        preference: One of 'auto', 'sccache', 'ccache', 'none', or a path to a binary.
-
-    Returns:
-        Path to the compiler cache binary, or None if not found/disabled.
-    """
-    if preference == 'none':
-        return None
-
-    if preference == 'auto':
-        # Prefer sccache over ccache
-        for cache in ['sccache', 'ccache']:
-            path = which(cache)
-            if path:
-                return path
-        return None
-
-    if preference in ('sccache', 'ccache'):
-        path = which(preference)
-        if path:
-            return path
-        print(f"Warning: {preference} not found on PATH, disabling compiler cache")
-        return None
-
-    # Assume it's a path to a binary
-    if os.path.isfile(preference) and os.access(preference, os.X_OK):
-        return preference
-
-    print(f"Warning: compiler cache '{preference}' not found or not executable, disabling compiler cache")
-    return None
-
-
 modes = {
     'debug': {
         'cxxflags': '-DDEBUG -DSANITIZE -DDEBUG_LSA_SANITIZER -DSCYLLA_ENABLE_ERROR_INJECTION',
@@ -725,9 +644,29 @@ raft_tests = set([
 vector_search_tests = set([
     'test/vector_search/vector_store_client_test',
     'test/vector_search/load_balancer_test',
-    'test/vector_search/client_test',
-    'test/vector_search/filter_test',
-    'test/vector_search/rescoring_test'
+    'test/vector_search/client_test'
+])
+
+vector_search_validator_bin = 'vector-search-validator/bin/vector-search-validator'
+vector_search_validator_deps = set([
+    'test/vector_search_validator/build-validator',
+    'test/vector_search_validator/Cargo.toml',
+    'test/vector_search_validator/crates/validator/Cargo.toml',
+    'test/vector_search_validator/crates/validator/src/main.rs',
+    'test/vector_search_validator/crates/validator-scylla/Cargo.toml',
+    'test/vector_search_validator/crates/validator-scylla/src/lib.rs',
+    'test/vector_search_validator/crates/validator-scylla/src/cql.rs',
+])
+
+vector_store_bin = 'vector-search-validator/bin/vector-store'
+vector_store_deps = set([
+    'test/vector_search_validator/build-env',
+    'test/vector_search_validator/build-vector-store',
+])
+
+vector_search_validator_bins = set([
+    vector_search_validator_bin,
+    vector_store_bin,
 ])
 
 wasms = set([
@@ -763,7 +702,7 @@ other = set([
     'iotune',
 ])
 
-all_artifacts = apps | cpp_apps | tests | other | wasms
+all_artifacts = apps | cpp_apps | tests | other | wasms | vector_search_validator_bins
 
 arg_parser = argparse.ArgumentParser('Configure scylla', add_help=False, formatter_class=argparse.ArgumentDefaultsHelpFormatter)
 arg_parser.add_argument('--out', dest='buildfile', action='store', default='build.ninja',
@@ -793,11 +732,6 @@ arg_parser.add_argument('--compiler', action='store', dest='cxx', default='clang
                         help='C++ compiler path')
 arg_parser.add_argument('--c-compiler', action='store', dest='cc', default='clang',
                         help='C compiler path')
-arg_parser.add_argument('--compiler-cache', action='store', dest='compiler_cache', default='auto',
-                        help='Compiler cache to use: auto (default, prefers sccache), sccache, ccache, none, or a path to a binary')
-# Workaround for https://github.com/mozilla/sccache/issues/2575
-arg_parser.add_argument('--sccache-rust', action=argparse.BooleanOptionalAction, default=False,
-                        help='Use sccache for rust code (if sccache is selected as compiler cache). Doesn\'t work with distributed builds.')
 add_tristate(arg_parser, name='dpdk', dest='dpdk', default=False,
                         help='Use dpdk (from seastar dpdk sources)')
 arg_parser.add_argument('--dpdk-target', action='store', dest='dpdk_target', default='',
@@ -925,10 +859,10 @@ scylla_core = (['message/messaging_service.cc',
                 'utils/alien_worker.cc',
                 'utils/array-search.cc',
                 'utils/base64.cc',
-                'utils/crypt_sha512.cc',
                 'utils/logalloc.cc',
                 'utils/large_bitset.cc',
-                'test/lib/limiting_data_source.cc',
+                'utils/buffer_input_stream.cc',
+                'utils/limiting_data_source.cc',
                 'utils/updateable_value.cc',
                 'message/dictionary_service.cc',
                 'utils/directories.cc',
@@ -1016,10 +950,6 @@ scylla_core = (['message/messaging_service.cc',
                 'cql3/functions/aggregate_fcts.cc',
                 'cql3/functions/castas_fcts.cc',
                 'cql3/functions/error_injection_fcts.cc',
-                'cql3/statements/strong_consistency/modification_statement.cc',
-                'cql3/statements/strong_consistency/select_statement.cc',
-                'cql3/statements/strong_consistency/statement_helpers.cc',
-                'cql3/functions/vector_similarity_fcts.cc',
                 'cql3/statements/cf_prop_defs.cc',
                 'cql3/statements/cf_statement.cc',
                 'cql3/statements/authentication_statement.cc',
@@ -1044,8 +974,8 @@ scylla_core = (['message/messaging_service.cc',
                 'cql3/statements/raw/parsed_statement.cc',
                 'cql3/statements/property_definitions.cc',
                 'cql3/statements/update_statement.cc',
-                'cql3/statements/broadcast_modification_statement.cc',
-                'cql3/statements/broadcast_select_statement.cc',
+                'cql3/statements/strongly_consistent_modification_statement.cc',
+                'cql3/statements/strongly_consistent_select_statement.cc',
                 'cql3/statements/delete_statement.cc',
                 'cql3/statements/prune_materialized_view_statement.cc',
                 'cql3/statements/batch_statement.cc',
@@ -1077,7 +1007,6 @@ scylla_core = (['message/messaging_service.cc',
                 'cql3/statements/list_service_level_attachments_statement.cc',
                 'cql3/statements/list_effective_service_level_statement.cc',
                 'cql3/statements/describe_statement.cc',
-                'cql3/statements/view_prop_defs.cc',
                 'cql3/update_parameters.cc',
                 'cql3/util.cc',
                 'cql3/ut_name.cc',
@@ -1174,7 +1103,6 @@ scylla_core = (['message/messaging_service.cc',
                 'utils/gz/crc_combine.cc',
                 'utils/gz/crc_combine_table.cc',
                 'utils/http.cc',
-                'utils/http_client_error_processing.cc',
                 'utils/rest/client.cc',
                 'utils/s3/aws_error.cc',
                 'utils/s3/client.cc',
@@ -1192,7 +1120,6 @@ scylla_core = (['message/messaging_service.cc',
                 'utils/azure/identity/default_credentials.cc',
                 'utils/gcp/gcp_credentials.cc',
                 'utils/gcp/object_storage.cc',
-                'utils/gcp/object_storage_retry_strategy.cc',
                 'gms/version_generator.cc',
                 'gms/versioned_value.cc',
                 'gms/gossiper.cc',
@@ -1230,7 +1157,6 @@ scylla_core = (['message/messaging_service.cc',
                 'locator/topology.cc',
                 'locator/util.cc',
                 'service/client_state.cc',
-                'service/client_routes.cc',
                 'service/storage_service.cc',
                 'service/session.cc',
                 'service/task_manager_module.cc',
@@ -1277,6 +1203,7 @@ scylla_core = (['message/messaging_service.cc',
                 'auth/passwords.cc',
                 'auth/password_authenticator.cc',
                 'auth/permission.cc',
+                'auth/permissions_cache.cc',
                 'auth/service.cc',
                 'auth/standard_role_manager.cc',
                 'auth/ldap_role_manager.cc',
@@ -1337,9 +1264,6 @@ scylla_core = (['message/messaging_service.cc',
                 'lang/wasm.cc',
                 'lang/wasm_alien_thread_runner.cc',
                 'lang/wasm_instance_cache.cc',
-                'service/strong_consistency/groups_manager.cc',
-                'service/strong_consistency/coordinator.cc',
-                'service/strong_consistency/state_machine.cc',
                 'service/raft/group0_state_id_handler.cc',
                 'service/raft/group0_state_machine.cc',
                 'service/raft/group0_state_machine_merger.cc',
@@ -1361,6 +1285,7 @@ scylla_core = (['message/messaging_service.cc',
                 'service/topology_state_machine.cc',
                 'service/topology_mutation.cc',
                 'service/topology_coordinator.cc',
+                'node_ops/node_ops_ctl.cc',
                 'node_ops/task_manager_module.cc',
                 'reader_concurrency_semaphore_group.cc',
                 'utils/disk_space_monitor.cc',
@@ -1368,7 +1293,6 @@ scylla_core = (['message/messaging_service.cc',
                 'vector_search/dns.cc',
                 'vector_search/client.cc',
                 'vector_search/clients.cc',
-                'vector_search/filter.cc',
                 'vector_search/truststore.cc'
                 ] + [Antlr3Grammar('cql3/Cql.g')] \
                   + scylla_raft_core
@@ -1393,8 +1317,6 @@ api = ['api/api.cc',
        'api/storage_proxy.cc',
        Json2Code('api/api-doc/cache_service.json'),
        'api/cache_service.cc',
-       Json2Code('api/api-doc/client_routes.json'),
-       'api/client_routes.cc',
        Json2Code('api/api-doc/collectd.json'),
        'api/collectd.cc',
        Json2Code('api/api-doc/endpoint_snitch_info.json'),
@@ -1444,7 +1366,6 @@ alternator = [
        'alternator/auth.cc',
        'alternator/streams.cc',
        'alternator/ttl.cc',
-       'alternator/http_compression.cc'
 ]
 
 idls = ['idl/gossip_digest.idl.hh',
@@ -1478,7 +1399,6 @@ idls = ['idl/gossip_digest.idl.hh',
         'idl/hinted_handoff.idl.hh',
         'idl/storage_proxy.idl.hh',
         'idl/sstables.idl.hh',
-        'idl/strong_consistency/state_machine.idl.hh',
         'idl/group0_state_machine.idl.hh',
         'idl/mapreduce_request.idl.hh',
         'idl/replica_exception.idl.hh',
@@ -1537,7 +1457,6 @@ scylla_perfs = ['test/perf/perf_alternator.cc',
                 'test/perf/perf_fast_forward.cc',
                 'test/perf/perf_row_cache_update.cc',
                 'test/perf/perf_simple_query.cc',
-                'test/perf/perf_cql_raw.cc',
                 'test/perf/perf_sstable.cc',
                 'test/perf/perf_tablets.cc',
                 'test/perf/tablet_load_balancing.cc',
@@ -1560,6 +1479,7 @@ deps = {
 
 pure_boost_tests = set([
     'test/boost/anchorless_list_test',
+    'test/boost/auth_passwords_test',
     'test/boost/auth_resource_test',
     'test/boost/big_decimal_test',
     'test/boost/caching_options_test',
@@ -1645,7 +1565,6 @@ for t in sorted(perf_tests):
 
 deps['test/boost/combined_tests'] += [
     'test/boost/aggregate_fcts_test.cc',
-    'test/boost/auth_cache_test.cc',
     'test/boost/auth_test.cc',
     'test/boost/batchlog_manager_test.cc',
     'test/boost/cache_algorithm_test.cc',
@@ -1693,7 +1612,6 @@ deps['test/boost/combined_tests'] += [
     'test/boost/schema_registry_test.cc',
     'test/boost/secondary_index_test.cc',
     'test/boost/sessions_test.cc',
-    'test/boost/simple_value_with_expiry_test.cc',
     'test/boost/sstable_compaction_test.cc',
     'test/boost/sstable_compressor_factory_test.cc',
     'test/boost/sstable_compression_config_test.cc',
@@ -1776,20 +1694,6 @@ deps['test/raft/discovery_test'] =  ['test/raft/discovery_test.cc',
 deps['test/vector_search/vector_store_client_test'] =  ['test/vector_search/vector_store_client_test.cc'] + scylla_tests_dependencies
 deps['test/vector_search/load_balancer_test'] = ['test/vector_search/load_balancer_test.cc'] + scylla_tests_dependencies
 deps['test/vector_search/client_test'] = ['test/vector_search/client_test.cc'] + scylla_tests_dependencies
-deps['test/vector_search/filter_test'] = ['test/vector_search/filter_test.cc'] + scylla_tests_dependencies
-deps['test/vector_search/rescoring_test'] = ['test/vector_search/rescoring_test.cc'] + scylla_tests_dependencies
-
-boost_tests_prefixes = ["test/boost/", "test/vector_search/", "test/raft/", "test/manual/", "test/ldap/"]
-
-# We need to link these files to all Boost tests to make sure that
-# we can execute `--list_json_content` on them. That will produce
-# a similar result as calling `--list_content={HRF,DOT}`.
-# Unfortunately, to be able to do that, we're forced to link the
-# relevant code by hand.
-for key in deps.keys():
-    for prefix in boost_tests_prefixes:
-        if key.startswith(prefix):
-            deps[key] += ["test/lib/boost_tree_lister_injector.cc", "test/lib/boost_test_tree_lister.cc"]
 
 wasm_deps = {}
 
@@ -2095,7 +1999,7 @@ def semicolon_separated(*flags):
 def real_relpath(path, start):
     return os.path.relpath(os.path.realpath(path), os.path.realpath(start))
 
-def configure_seastar(build_dir, mode, mode_config, compiler_cache=None):
+def configure_seastar(build_dir, mode, mode_config):
     seastar_cxx_ld_flags = mode_config['cxx_ld_flags']
     # We want to "undo" coverage for seastar if we have it enabled.
     if args.coverage:
@@ -2142,10 +2046,6 @@ def configure_seastar(build_dir, mode, mode_config, compiler_cache=None):
         '-DSeastar_IO_URING=ON',
     ]
 
-    if compiler_cache:
-        seastar_cmake_args += [f'-DCMAKE_CXX_COMPILER_LAUNCHER={compiler_cache}',
-                               f'-DCMAKE_C_COMPILER_LAUNCHER={compiler_cache}']
-
     if args.stack_guards is not None:
         stack_guards = 'ON' if args.stack_guards else 'OFF'
         seastar_cmake_args += ['-DSeastar_STACK_GUARDS={}'.format(stack_guards)]
@@ -2177,7 +2077,7 @@ def configure_seastar(build_dir, mode, mode_config, compiler_cache=None):
     subprocess.check_call(seastar_cmd, shell=False, cwd=cmake_dir)
 
 
-def configure_abseil(build_dir, mode, mode_config, compiler_cache=None):
+def configure_abseil(build_dir, mode, mode_config):
     abseil_cflags = mode_config['lib_cflags']
     cxx_flags = mode_config['cxxflags']
     if '-DSANITIZE' in cxx_flags:
@@ -2203,10 +2103,6 @@ def configure_abseil(build_dir, mode, mode_config, compiler_cache=None):
         '-DABSL_PROPAGATE_CXX_STD=ON',
     ]
 
-    if compiler_cache:
-        abseil_cmake_args += [f'-DCMAKE_CXX_COMPILER_LAUNCHER={compiler_cache}',
-                              f'-DCMAKE_C_COMPILER_LAUNCHER={compiler_cache}']
-
     cmake_args = abseil_cmake_args[:]
     abseil_build_dir = os.path.join(build_dir, mode, 'abseil')
     abseil_cmd = ['cmake', '-G', 'Ninja', real_relpath('abseil', abseil_build_dir)] + cmake_args
@@ -2352,6 +2248,15 @@ def get_extra_cxxflags(mode, mode_config, cxx, debuginfo):
     if debuginfo and mode_config['can_have_debug_info']:
         cxxflags += ['-g', '-gz']
 
+    if 'clang' in cxx:
+        # Since AssignmentTracking was enabled by default in clang
+        # (llvm/llvm-project@de6da6ad55d3ca945195d1cb109cb8efdf40a52a)
+        # coroutine frame debugging info (`coro_frame_ty`) is broken.
+        #
+        # It seems that we aren't losing much by disabling AssigmentTracking,
+        # so for now we choose to disable it to get `coro_frame_ty` back.
+        cxxflags.append('-Xclang -fexperimental-assignment-tracking=disabled')
+
     return cxxflags
 
 
@@ -2379,15 +2284,10 @@ def write_build_file(f,
                      scylla_product,
                      scylla_version,
                      scylla_release,
-                     compiler_cache,
                      args):
     use_precompiled_header = not args.disable_precompiled_header
     warnings = get_warning_options(args.cxx)
     rustc_target = pick_rustc_target('wasm32-wasi', 'wasm32-wasip1')
-    # If compiler cache is available, prefix the compiler with it
-    cxx_with_cache = f'{compiler_cache} {args.cxx}' if compiler_cache else args.cxx
-    # For Rust, sccache is used via RUSTC_WRAPPER environment variable
-    rustc_wrapper = f'RUSTC_WRAPPER={compiler_cache} ' if compiler_cache and 'sccache' in compiler_cache and args.sccache_rust else ''
     f.write(textwrap.dedent('''\
         configure_args = {configure_args}
         builddir = {outdir}
@@ -2450,7 +2350,7 @@ def write_build_file(f,
             command = clang --target=wasm32 --no-standard-libraries -Wl,--export-all -Wl,--no-entry $in -o $out
             description = C2WASM $out
         rule rust2wasm
-            command = {rustc_wrapper}cargo build --target={rustc_target} --example=$example --locked --manifest-path=test/resource/wasm/rust/Cargo.toml --target-dir=$builddir/wasm/ $
+            command = cargo build --target={rustc_target} --example=$example --locked --manifest-path=test/resource/wasm/rust/Cargo.toml --target-dir=$builddir/wasm/ $
                 && wasm-opt -Oz $builddir/wasm/{rustc_target}/debug/examples/$example.wasm -o $builddir/wasm/$example.wasm $
                 && wasm-strip $builddir/wasm/$example.wasm
             description = RUST2WASM $out
@@ -2466,7 +2366,7 @@ def write_build_file(f,
           command = llvm-profdata merge $in -output=$out
         ''').format(configure_args=configure_args,
                     outdir=outdir,
-                    cxx=cxx_with_cache,
+                    cxx=args.cxx,
                     user_cflags=user_cflags,
                     warnings=warnings,
                     defines=defines,
@@ -2474,7 +2374,6 @@ def write_build_file(f,
                     user_ldflags=user_ldflags,
                     libs=libs,
                     rustc_target=rustc_target,
-                    rustc_wrapper=rustc_wrapper,
                     link_pool_depth=link_pool_depth,
                     seastar_path=args.seastar_path,
                     ninja=ninja,
@@ -2559,15 +2458,16 @@ def write_build_file(f,
               description = TEST {mode}
             # This rule is unused for PGO stages. They use the rust lib from the parent mode.
             rule rust_lib.{mode}
-              command = CARGO_BUILD_DEP_INFO_BASEDIR='.' {rustc_wrapper}cargo build --locked --manifest-path=rust/Cargo.toml --target-dir=$builddir/{mode} --profile=rust-{mode} $
+              command = CARGO_BUILD_DEP_INFO_BASEDIR='.' cargo build --locked --manifest-path=rust/Cargo.toml --target-dir=$builddir/{mode} --profile=rust-{mode} $
                         && touch $out
               description = RUST_LIB $out
-            ''').format(mode=mode, antlr3_exec=args.antlr3_exec, fmt_lib=fmt_lib, test_repeat=args.test_repeat, test_timeout=args.test_timeout, rustc_wrapper=rustc_wrapper, **modeval))
+            ''').format(mode=mode, antlr3_exec=args.antlr3_exec, fmt_lib=fmt_lib, test_repeat=args.test_repeat, test_timeout=args.test_timeout, **modeval))
         f.write(
-            'build {mode}-build: phony {artifacts} {wasms}\n'.format(
+            'build {mode}-build: phony {artifacts} {wasms} {vector_search_validator_bins}\n'.format(
                 mode=mode,
-                artifacts=str.join(' ', ['$builddir/' + mode + '/' + x for x in sorted(build_artifacts - wasms)]),
+                artifacts=str.join(' ', ['$builddir/' + mode + '/' + x for x in sorted(build_artifacts - wasms - vector_search_validator_bins)]),
                 wasms = str.join(' ', ['$builddir/' + x for x in sorted(build_artifacts & wasms)]),
+                vector_search_validator_bins=str.join(' ', ['$builddir/' + x for x in sorted(build_artifacts & vector_search_validator_bins)]),
             )
         )
         if profile_recipe := modes[mode].get('profile_recipe'):
@@ -2597,7 +2497,7 @@ def write_build_file(f,
                 continue
             profile_dep = modes[mode].get('profile_target', "")
 
-            if binary in other or binary in wasms:
+            if binary in other or binary in wasms or binary in vector_search_validator_bins:
                 continue
             srcs = deps[binary]
             # 'scylla'
@@ -2625,7 +2525,7 @@ def write_build_file(f,
                 # In debug/sanitize modes, we compile with fsanitizers,
                 # so must use the same options during the link:
                 if '-DSANITIZE' in modes[mode]['cxxflags']:
-                    f.write('   libs = -fsanitize=address -fsanitize=undefined -lubsan\n')
+                    f.write('   libs = -fsanitize=address -fsanitize=undefined\n')
                 else:
                     f.write('   libs =\n')
                 f.write(f'build $builddir/{mode}/{binary}.stripped: strip $builddir/{mode}/{binary}\n')
@@ -2708,10 +2608,11 @@ def write_build_file(f,
         )
 
         f.write(
-            'build {mode}-test: test.{mode} {test_executables} $builddir/{mode}/scylla {wasms}\n'.format(
+            'build {mode}-test: test.{mode} {test_executables} $builddir/{mode}/scylla {wasms} {vector_search_validator_bins} \n'.format(
                 mode=mode,
                 test_executables=' '.join(['$builddir/{}/{}'.format(mode, binary) for binary in sorted(tests)]),
                 wasms=' '.join([f'$builddir/{binary}' for binary in sorted(wasms)]),
+                vector_search_validator_bins=' '.join([f'$builddir/{binary}' for binary in sorted(vector_search_validator_bins)]),
             )
         )
         f.write(
@@ -2798,35 +2699,38 @@ def write_build_file(f,
 
         seastar_dep = f'$builddir/{mode}/seastar/libseastar.{seastar_lib_ext}'
         seastar_testing_dep = f'$builddir/{mode}/seastar/libseastar_testing.{seastar_lib_ext}'
-        f.write(f'build {seastar_dep}: ninja $builddir/{mode}/seastar/build.ninja | always {profile_dep}\n')
+        f.write('build {seastar_dep}: ninja $builddir/{mode}/seastar/build.ninja | always {profile_dep}\n'
+                .format(**locals()))
         f.write('  pool = submodule_pool\n')
-        f.write(f'  subdir = $builddir/{mode}/seastar\n')
-        f.write('  target = seastar\n')
-        f.write(f'build {seastar_testing_dep}: ninja $builddir/{mode}/seastar/build.ninja | always {profile_dep}\n')
+        f.write('  subdir = $builddir/{mode}/seastar\n'.format(**locals()))
+        f.write('  target = seastar\n'.format(**locals()))
+        f.write('build {seastar_testing_dep}: ninja $builddir/{mode}/seastar/build.ninja | always {profile_dep}\n'
+                .format(**locals()))
         f.write('  pool = submodule_pool\n')
-        f.write(f'  subdir = $builddir/{mode}/seastar\n')
-        f.write('  target = seastar_testing\n')
-        f.write(f'  profile_dep = {profile_dep}\n')
+        f.write('  subdir = $builddir/{mode}/seastar\n'.format(**locals()))
+        f.write('  target = seastar_testing\n'.format(**locals()))
+        f.write('  profile_dep = {profile_dep}\n'.format(**locals()))
 
         for lib in abseil_libs:
-            f.write(f'build $builddir/{mode}/abseil/{lib}: ninja $builddir/{mode}/abseil/build.ninja | always {profile_dep}\n')
-            f.write(f'  pool = submodule_pool\n')
-            f.write(f'  subdir = $builddir/{mode}/abseil\n')
-            f.write(f'  target = {lib}\n')
-            f.write(f'  profile_dep = {profile_dep}\n')
+            f.write('build $builddir/{mode}/abseil/{lib}: ninja $builddir/{mode}/abseil/build.ninja | always {profile_dep}\n'.format(**locals()))
+            f.write('  pool = submodule_pool\n')
+            f.write('  subdir = $builddir/{mode}/abseil\n'.format(**locals()))
+            f.write('  target = {lib}\n'.format(**locals()))
+            f.write('  profile_dep = {profile_dep}\n'.format(**locals()))
 
         f.write(f'build $builddir/{mode}/stdafx.hh.pch: cxx_build_precompiled_header.{mode} stdafx.hh | {profile_dep} {seastar_dep} {abseil_dep} {gen_headers_dep} {pch_dep}\n')
 
-        f.write(f'build $builddir/{mode}/seastar/apps/iotune/iotune: ninja $builddir/{mode}/seastar/build.ninja | $builddir/{mode}/seastar/libseastar.{seastar_lib_ext}\n')
+        f.write('build $builddir/{mode}/seastar/apps/iotune/iotune: ninja $builddir/{mode}/seastar/build.ninja | $builddir/{mode}/seastar/libseastar.{seastar_lib_ext}\n'
+                .format(**locals()))
         f.write('  pool = submodule_pool\n')
-        f.write(f'  subdir = $builddir/{mode}/seastar\n')
-        f.write('  target = iotune\n')
-        f.write(f'  profile_dep = {profile_dep}\n')
-        f.write(textwrap.dedent(f'''\
+        f.write('  subdir = $builddir/{mode}/seastar\n'.format(**locals()))
+        f.write('  target = iotune\n'.format(**locals()))
+        f.write('  profile_dep = {profile_dep}\n'.format(**locals()))
+        f.write(textwrap.dedent('''\
             build $builddir/{mode}/iotune: copy $builddir/{mode}/seastar/apps/iotune/iotune
             build $builddir/{mode}/iotune.stripped: strip $builddir/{mode}/iotune
             build $builddir/{mode}/iotune.debug: phony $builddir/{mode}/iotune.stripped
-            '''))
+            ''').format(**locals()))
         if args.dist_only:
             include_scylla_and_iotune = ''
             include_scylla_and_iotune_stripped = ''
@@ -2835,16 +2739,16 @@ def write_build_file(f,
             include_scylla_and_iotune = f'$builddir/{mode}/scylla $builddir/{mode}/iotune $builddir/{mode}/patchelf'
             include_scylla_and_iotune_stripped = f'$builddir/{mode}/scylla.stripped $builddir/{mode}/iotune.stripped $builddir/{mode}/patchelf.stripped'
             include_scylla_and_iotune_debug = f'$builddir/{mode}/scylla.debug $builddir/{mode}/iotune.debug'
-        f.write(f'build $builddir/{mode}/dist/tar/{scylla_product}-unstripped-{scylla_version}-{scylla_release}.{arch}.tar.gz: package {include_scylla_and_iotune} $builddir/SCYLLA-RELEASE-FILE $builddir/SCYLLA-VERSION-FILE $builddir/debian/debian $builddir/node_exporter/node_exporter | always\n')
-        f.write(f'  mode = {mode}\n')
-        f.write(f'build $builddir/{mode}/dist/tar/{scylla_product}-{scylla_version}-{scylla_release}.{arch}.tar.gz: stripped_package {include_scylla_and_iotune_stripped} $builddir/SCYLLA-RELEASE-FILE $builddir/SCYLLA-VERSION-FILE $builddir/debian/debian $builddir/node_exporter/node_exporter.stripped | always\n')
-        f.write(f'  mode = {mode}\n')
-        f.write(f'build $builddir/{mode}/dist/tar/{scylla_product}-debuginfo-{scylla_version}-{scylla_release}.{arch}.tar.gz: debuginfo_package {include_scylla_and_iotune_debug} $builddir/SCYLLA-RELEASE-FILE $builddir/SCYLLA-VERSION-FILE $builddir/debian/debian $builddir/node_exporter/node_exporter.debug | always\n')
-        f.write(f'  mode = {mode}\n')
-        f.write(f'build $builddir/{mode}/dist/tar/{scylla_product}-package.tar.gz: copy $builddir/{mode}/dist/tar/{scylla_product}-{scylla_version}-{scylla_release}.{arch}.tar.gz\n')
-        f.write(f'  mode = {mode}\n')
-        f.write(f'build $builddir/{mode}/dist/tar/{scylla_product}-{arch}-package.tar.gz: copy $builddir/{mode}/dist/tar/{scylla_product}-{scylla_version}-{scylla_release}.{arch}.tar.gz\n')
-        f.write(f'  mode = {mode}\n')
+        f.write('build $builddir/{mode}/dist/tar/{scylla_product}-unstripped-{scylla_version}-{scylla_release}.{arch}.tar.gz: package {include_scylla_and_iotune} $builddir/SCYLLA-RELEASE-FILE $builddir/SCYLLA-VERSION-FILE $builddir/debian/debian $builddir/node_exporter/node_exporter | always\n'.format(**locals()))
+        f.write('  mode = {mode}\n'.format(**locals()))
+        f.write('build $builddir/{mode}/dist/tar/{scylla_product}-{scylla_version}-{scylla_release}.{arch}.tar.gz: stripped_package {include_scylla_and_iotune_stripped} $builddir/SCYLLA-RELEASE-FILE $builddir/SCYLLA-VERSION-FILE $builddir/debian/debian $builddir/node_exporter/node_exporter.stripped | always\n'.format(**locals()))
+        f.write('  mode = {mode}\n'.format(**locals()))
+        f.write('build $builddir/{mode}/dist/tar/{scylla_product}-debuginfo-{scylla_version}-{scylla_release}.{arch}.tar.gz: debuginfo_package {include_scylla_and_iotune_debug} $builddir/SCYLLA-RELEASE-FILE $builddir/SCYLLA-VERSION-FILE $builddir/debian/debian $builddir/node_exporter/node_exporter.debug | always\n'.format(**locals()))
+        f.write('  mode = {mode}\n'.format(**locals()))
+        f.write('build $builddir/{mode}/dist/tar/{scylla_product}-package.tar.gz: copy $builddir/{mode}/dist/tar/{scylla_product}-{scylla_version}-{scylla_release}.{arch}.tar.gz\n'.format(**locals()))
+        f.write('  mode = {mode}\n'.format(**locals()))
+        f.write('build $builddir/{mode}/dist/tar/{scylla_product}-{arch}-package.tar.gz: copy $builddir/{mode}/dist/tar/{scylla_product}-{scylla_version}-{scylla_release}.{arch}.tar.gz\n'.format(**locals()))
+        f.write('  mode = {mode}\n'.format(**locals()))
 
         f.write(f'build $builddir/dist/{mode}/redhat: rpmbuild $builddir/{mode}/dist/tar/{scylla_product}-unstripped-{scylla_version}-{scylla_release}.{arch}.tar.gz\n')
         f.write(f'  mode = {mode}\n')
@@ -2879,6 +2783,19 @@ def write_build_file(f,
             'build compiler-training: phony {}\n'.format(' '.join(['{mode}-compiler-training'.format(mode=mode) for mode in default_modes]))
     )
 
+    f.write(textwrap.dedent(f'''\
+        rule build-vector-search-validator
+            command = test/vector_search_validator/build-validator $builddir
+        rule build-vector-store
+            command = test/vector_search_validator/build-vector-store $builddir
+        '''))
+    f.write(
+            'build $builddir/{vector_search_validator_bin}: build-vector-search-validator {}\n'.format(' '.join([dep for dep in sorted(vector_search_validator_deps)]), vector_search_validator_bin=vector_search_validator_bin)
+    )
+    f.write(
+            'build $builddir/{vector_store_bin}: build-vector-store {}\n'.format(' '.join([dep for dep in sorted(vector_store_deps)]), vector_store_bin=vector_store_bin)
+    )
+
     f.write(textwrap.dedent(f'''\
         build dist-unified-tar: phony {' '.join([f'$builddir/{mode}/dist/tar/{scylla_product}-unified-{scylla_version}-{scylla_release}.{arch}.tar.gz' for mode in default_modes])}
         build dist-unified: phony dist-unified-tar
@@ -3004,9 +2921,6 @@ def create_build_system(args):
 
     os.makedirs(outdir, exist_ok=True)
 
-    compiler_cache = find_compiler_cache(args.compiler_cache)
-    resolve_compilers_for_compiler_cache(args, compiler_cache)
-
     scylla_product, scylla_version, scylla_release = generate_version(args.date_stamp)
 
     for mode, mode_config in build_modes.items():
@@ -3023,8 +2937,8 @@ def create_build_system(args):
         # {outdir}/{mode}/seastar/build.ninja, and
         # {outdir}/{mode}/seastar/seastar.pc is queried for building flags
         for mode, mode_config in build_modes.items():
-            configure_seastar(outdir, mode, mode_config, compiler_cache)
-            configure_abseil(outdir, mode, mode_config, compiler_cache)
+            configure_seastar(outdir, mode, mode_config)
+            configure_abseil(outdir, mode, mode_config)
         user_cflags += ' -isystem abseil'
 
     for mode, mode_config in build_modes.items():
@@ -3047,7 +2961,6 @@ def create_build_system(args):
                          scylla_product,
                          scylla_version,
                          scylla_release,
-                         compiler_cache,
                          args)
     generate_compdb('compile_commands.json', ninja, args.buildfile, selected_modes)
 
@@ -3090,10 +3003,6 @@ def configure_using_cmake(args):
     selected_modes = args.selected_modes or default_modes
     selected_configs = ';'.join(build_modes[mode].cmake_build_type for mode
                                 in selected_modes)
-
-    compiler_cache = find_compiler_cache(args.compiler_cache)
-    resolve_compilers_for_compiler_cache(args, compiler_cache)
-
     settings = {
         'CMAKE_CONFIGURATION_TYPES': selected_configs,
         'CMAKE_CROSS_CONFIGS': selected_configs,
@@ -3111,14 +3020,6 @@ def configure_using_cmake(args):
         'Scylla_WITH_DEBUG_INFO' : 'ON' if args.debuginfo else 'OFF',
         'Scylla_USE_PRECOMPILED_HEADER': 'OFF' if args.disable_precompiled_header else 'ON',
     }
-
-    if compiler_cache:
-        settings['CMAKE_CXX_COMPILER_LAUNCHER'] = compiler_cache
-        settings['CMAKE_C_COMPILER_LAUNCHER'] = compiler_cache
-        # For Rust, sccache is used via RUSTC_WRAPPER
-        if 'sccache' in compiler_cache and args.sccache_rust:
-            settings['Scylla_RUSTC_WRAPPER'] = compiler_cache
-
     if args.date_stamp:
         settings['Scylla_DATE_STAMP'] = args.date_stamp
     if args.staticboost:
@@ -3150,7 +3051,7 @@ def configure_using_cmake(args):
 
     if not args.dist_only:
         for mode in selected_modes:
-            configure_seastar(build_dir, build_modes[mode].cmake_build_type, modes[mode], compiler_cache)
+            configure_seastar(build_dir, build_modes[mode].cmake_build_type, modes[mode])
 
     cmake_command = ['cmake']
     cmake_command += [f'-D{var}={value}' for var, value in settings.items()]

cql3/CMakeLists.txt

@@ -47,10 +47,6 @@ target_sources(cql3
     functions/aggregate_fcts.cc
     functions/castas_fcts.cc
     functions/error_injection_fcts.cc
-    statements/strong_consistency/select_statement.cc
-    statements/strong_consistency/modification_statement.cc
-    statements/strong_consistency/statement_helpers.cc
-    functions/vector_similarity_fcts.cc
     statements/cf_prop_defs.cc
     statements/cf_statement.cc
     statements/authentication_statement.cc
@@ -75,8 +71,8 @@ target_sources(cql3
     statements/raw/parsed_statement.cc
     statements/property_definitions.cc
     statements/update_statement.cc
-    statements/broadcast_modification_statement.cc
-    statements/broadcast_select_statement.cc
+    statements/strongly_consistent_modification_statement.cc
+    statements/strongly_consistent_select_statement.cc
     statements/delete_statement.cc
     statements/prune_materialized_view_statement.cc
     statements/batch_statement.cc
@@ -108,7 +104,6 @@ target_sources(cql3
     statements/list_service_level_attachments_statement.cc
     statements/list_effective_service_level_statement.cc
     statements/describe_statement.cc
-    statements/view_prop_defs.cc
     update_parameters.cc
     util.cc
     ut_name.cc

cql3/Cql.g

@@ -389,10 +389,8 @@ selectStatement returns [std::unique_ptr<raw::select_statement> expr]
         bool is_ann_ordering = false;
     }
     : K_SELECT (
-                ( (K_JSON K_DISTINCT)=> K_JSON { statement_subtype = raw::select_statement::parameters::statement_subtype::JSON; }
-                | (K_JSON selectClause K_FROM)=> K_JSON { statement_subtype = raw::select_statement::parameters::statement_subtype::JSON; }
-                )?
-                ( (K_DISTINCT selectClause K_FROM)=> K_DISTINCT { is_distinct = true; } )?
+                ( K_JSON { statement_subtype = raw::select_statement::parameters::statement_subtype::JSON; } )?
+                ( K_DISTINCT { is_distinct = true; } )?
                 sclause=selectClause
                )
       K_FROM (
@@ -427,7 +425,6 @@ selector returns [shared_ptr<raw_selector> s]
 
 unaliasedSelector returns [uexpression tmp]
     :  ( c=cident                                  { tmp = unresolved_identifier{std::move(c)}; }
-       | v=value                                   { tmp = std::move(v); }
        | K_COUNT '(' countArgument ')'             { tmp = make_count_rows_function_expression(); }
        | K_WRITETIME '(' c=cident ')'              { tmp = column_mutation_attribute{column_mutation_attribute::attribute_kind::writetime,
                                                                                               unresolved_identifier{std::move(c)}}; }
@@ -450,7 +447,9 @@ selectionFunctionArgs returns [std::vector<expression> a]
 
 countArgument
     : '*'
-    /* COUNT(1) is also allowed, it is recognized via the general function(args) path */
+    | i=INTEGER { if (i->getText() != "1") {
+                    add_recognition_error("Only COUNT(1) is supported, got COUNT(" + i->getText() + ")");
+                } }
     ;
 
 whereClause returns [uexpression clause]
@@ -874,8 +873,8 @@ cfamDefinition[cql3::statements::create_table_statement::raw_statement& expr]
     ;
 
 cfamColumns[cql3::statements::create_table_statement::raw_statement& expr]
-    @init { bool is_static=false, is_ttl=false; }
-    : k=ident v=comparatorType (K_TTL {is_ttl = true;})? (K_STATIC {is_static = true;})? { $expr.add_definition(k, v, is_static, is_ttl); }
+    @init { bool is_static=false; }
+    : k=ident v=comparatorType (K_STATIC {is_static = true;})? { $expr.add_definition(k, v, is_static); }
         (K_PRIMARY K_KEY { $expr.add_key_aliases(std::vector<shared_ptr<cql3::column_identifier>>{k}); })?
     | K_PRIMARY K_KEY '(' pkDef[expr] (',' c=ident { $expr.add_column_alias(c); } )* ')'
     ;
@@ -886,10 +885,6 @@ pkDef[cql3::statements::create_table_statement::raw_statement& expr]
     | '(' k1=ident { l.push_back(k1); } ( ',' kn=ident { l.push_back(kn); } )* ')' { $expr.add_key_aliases(l); }
     ;
 
-cfamProperties[cql3::statements::cf_properties& expr]
-    : cfamProperty[expr] (K_AND cfamProperty[expr])*
-    ;
-
 cfamProperty[cql3::statements::cf_properties& expr]
     : property[*$expr.properties()]
     | K_COMPACT K_STORAGE { $expr.set_compact_storage(); }
@@ -927,22 +922,16 @@ typeColumns[create_type_statement& expr]
  */
 createIndexStatement returns [std::unique_ptr<create_index_statement> expr]
     @init {
-        auto idx_props = make_shared<index_specific_prop_defs>();
-        auto props = index_prop_defs();
+        auto props = make_shared<index_prop_defs>();
         bool if_not_exists = false;
         auto name = ::make_shared<cql3::index_name>();
         std::vector<::shared_ptr<index_target::raw>> targets;
     }
-    : K_CREATE (K_CUSTOM { idx_props->is_custom = true; })? K_INDEX (K_IF K_NOT K_EXISTS { if_not_exists = true; } )?
+    : K_CREATE (K_CUSTOM { props->is_custom = true; })? K_INDEX (K_IF K_NOT K_EXISTS { if_not_exists = true; } )?
         (idxName[*name])? K_ON cf=columnFamilyName '(' (target1=indexIdent { targets.emplace_back(target1); } (',' target2=indexIdent { targets.emplace_back(target2); } )*)? ')'
-        (K_USING cls=STRING_LITERAL { idx_props->custom_class = sstring{$cls.text}; })?
-        (K_WITH cfamProperties[props])?
-      {
-        props.extract_index_specific_properties_to(*idx_props);
-        view_prop_defs view_props = std::move(props).into_view_prop_defs();
-
-        $expr = std::make_unique<create_index_statement>(cf, name, targets, std::move(idx_props), std::move(view_props), if_not_exists);
-      }
+        (K_USING cls=STRING_LITERAL { props->custom_class = sstring{$cls.text}; })?
+        (K_WITH properties[*props])?
+      { $expr = std::make_unique<create_index_statement>(cf, name, targets, props, if_not_exists); }
     ;
 
 indexIdent returns [::shared_ptr<index_target::raw> id]
@@ -1042,7 +1031,6 @@ alterTableStatement returns [std::unique_ptr<alter_table_statement::raw_statemen
         std::vector<alter_table_statement::column_change> column_changes;
         std::vector<std::pair<shared_ptr<cql3::column_identifier::raw>, shared_ptr<cql3::column_identifier::raw>>> renames;
         auto attrs = std::make_unique<cql3::attributes::raw>();
-        shared_ptr<cql3::column_identifier::raw> ttl_change;
     }
     : K_ALTER K_COLUMNFAMILY cf=columnFamilyName
           ( K_ALTER id=cident K_TYPE v=comparatorType { type = alter_table_statement::type::alter; column_changes.emplace_back(alter_table_statement::column_change{id, v}); }
@@ -1061,11 +1049,9 @@ alterTableStatement returns [std::unique_ptr<alter_table_statement::raw_statemen
           | K_RENAME                                  { type = alter_table_statement::type::rename; }
                id1=cident K_TO toId1=cident { renames.emplace_back(id1, toId1); }
                ( K_AND idn=cident K_TO toIdn=cident { renames.emplace_back(idn, toIdn); } )*
-          | K_TTL                                     { type = alter_table_statement::type::ttl; }
-               ( id=cident { ttl_change = id; } | K_NULL )
           )
     {
-        $expr = std::make_unique<alter_table_statement::raw_statement>(std::move(cf), type, std::move(column_changes), std::move(props), std::move(renames), std::move(attrs), std::move(ttl_change));
+        $expr = std::make_unique<alter_table_statement::raw_statement>(std::move(cf), type, std::move(column_changes), std::move(props), std::move(renames), std::move(attrs));
     }
     ;
 
@@ -1093,9 +1079,9 @@ alterTypeStatement returns [std::unique_ptr<alter_type_statement> expr]
  */
 alterViewStatement returns [std::unique_ptr<alter_view_statement> expr]
     @init {
-        auto props = cql3::statements::view_prop_defs();
+        auto props = cql3::statements::cf_prop_defs();
     }
-    : K_ALTER K_MATERIALIZED K_VIEW cf=columnFamilyName K_WITH properties[*props.properties()]
+    : K_ALTER K_MATERIALIZED K_VIEW cf=columnFamilyName K_WITH properties[props]
     {
         $expr = std::make_unique<alter_view_statement>(std::move(cf), std::move(props));
     }
@@ -2074,21 +2060,7 @@ vector_type returns [shared_ptr<cql3::cql3_type::raw> pt]
         {
             if ($d.text[0] == '-')
                 throw exceptions::invalid_request_exception("Vectors must have a dimension greater than 0");
-            unsigned long parsed_dimension;
-            try {
-                parsed_dimension = std::stoul($d.text);
-            } catch (const std::exception& e) {
-                throw exceptions::invalid_request_exception(format("Invalid vector dimension: {}", $d.text));
-            }
-            static_assert(sizeof(unsigned long) >= sizeof(vector_dimension_t));
-            if (parsed_dimension == 0) {
-                throw exceptions::invalid_request_exception("Vectors must have a dimension greater than 0");
-            }
-            if (parsed_dimension > cql3::cql3_type::MAX_VECTOR_DIMENSION) {
-                throw exceptions::invalid_request_exception(
-                        format("Vectors must have a dimension less than or equal to {}", cql3::cql3_type::MAX_VECTOR_DIMENSION));
-            }
-            $pt = cql3::cql3_type::raw::vector(t, static_cast<vector_dimension_t>(parsed_dimension));
+            $pt = cql3::cql3_type::raw::vector(t, std::stoul($d.text));
         }
     ;
 

cql3/assignment_testable.hh

@@ -25,11 +25,6 @@ public:
         NOT_ASSIGNABLE,
     };
 
-    struct vector_test_result {
-        test_result result;
-        std::optional<vector_dimension_t> dimension_opt;
-    };
-
     static bool is_assignable(test_result tr) {
         return tr != test_result::NOT_ASSIGNABLE;
     }
@@ -49,8 +44,6 @@ public:
      */
     virtual test_result test_assignment(data_dictionary::database db, const sstring& keyspace, const schema* schema_opt, const column_specification& receiver) const = 0;
 
-    virtual vector_test_result test_assignment_any_size_float_vector() const = 0;
-
     virtual std::optional<data_type> assignment_testable_type_opt() const = 0;
 
     // for error reporting

cql3/cql3_type.cc

@@ -307,14 +307,17 @@ public:
 
 class cql3_type::raw_vector : public raw {
     shared_ptr<raw> _type;
-    vector_dimension_t _dimension;
+    size_t _dimension;
+
+    // This limitation is acquired from the maximum number of dimensions in OpenSearch. 
+    static constexpr size_t MAX_VECTOR_DIMENSION = 16000;
 
     virtual sstring to_string() const override {
         return seastar::format("vector<{}, {}>", _type, _dimension);
     }
 
 public:
-    raw_vector(shared_ptr<raw> type, vector_dimension_t dimension)
+    raw_vector(shared_ptr<raw> type, size_t dimension)
             : _type(std::move(type)), _dimension(dimension) {
     }
 
@@ -414,7 +417,7 @@ cql3_type::raw::tuple(std::vector<shared_ptr<raw>> ts) {
 }
 
 shared_ptr<cql3_type::raw>
-cql3_type::raw::vector(shared_ptr<raw> t, vector_dimension_t dimension) {
+cql3_type::raw::vector(shared_ptr<raw> t, size_t dimension) {
     return ::make_shared<raw_vector>(std::move(t), dimension);
 }
 

cql3/cql3_type.hh

@@ -39,9 +39,6 @@ public:
     data_type get_type() const { return _type; }
     const sstring& to_string() const { return _type->cql3_type_name(); }
 
-    // This limitation is acquired from the maximum number of dimensions in OpenSearch.
-    static constexpr vector_dimension_t MAX_VECTOR_DIMENSION = 16000;
-
     // For UserTypes, we need to know the current keyspace to resolve the
     // actual type used, so Raw is a "not yet prepared" CQL3Type.
     class raw {
@@ -67,7 +64,7 @@ public:
         static shared_ptr<raw> list(shared_ptr<raw> t);
         static shared_ptr<raw> set(shared_ptr<raw> t);
         static shared_ptr<raw> tuple(std::vector<shared_ptr<raw>> ts);
-        static shared_ptr<raw> vector(shared_ptr<raw> t, vector_dimension_t dimension);
+        static shared_ptr<raw> vector(shared_ptr<raw> t, size_t dimension);
         static shared_ptr<raw> frozen(shared_ptr<raw> t);
         friend sstring format_as(const raw& r) {
             return r.to_string();

cql3/expr/prepare_expr.cc

@@ -10,7 +10,6 @@
 #include "expr-utils.hh"
 #include "evaluate.hh"
 #include "cql3/functions/functions.hh"
-#include "cql3/functions/aggregate_fcts.hh"
 #include "cql3/functions/castas_fcts.hh"
 #include "cql3/functions/scalar_function.hh"
 #include "cql3/column_identifier.hh"
@@ -502,8 +501,8 @@ vector_validate_assignable_to(const collection_constructor& c, data_dictionary::
         throw exceptions::invalid_request_exception(format("Invalid vector type literal for {} of type {}", *receiver.name, receiver.type->as_cql3_type()));
     }
 
-    vector_dimension_t expected_size = vt->get_dimension();
-    if (expected_size == 0) {
+    size_t expected_size = vt->get_dimension();
+    if (!expected_size) {
         throw exceptions::invalid_request_exception(format("Invalid vector type literal for {}: type {} expects at least one element",
                                                             *receiver.name, receiver.type->as_cql3_type()));
     }
@@ -1048,47 +1047,8 @@ prepare_function_args_for_type_inference(std::span<const expression> args, data_
     return partially_prepared_args;
 }
 
-// Special case for count(1) - recognize it as the countRows() function. Note it is quite
-// artificial and we might relax it to the more general count(expression) later.
-static
-std::optional<expression>
-try_prepare_count_rows(const expr::function_call& fc, data_dictionary::database db, const sstring& keyspace, const schema* schema_opt, lw_shared_ptr<column_specification> receiver) {
-    return std::visit(overloaded_functor{
-        [&] (const functions::function_name& name) -> std::optional<expression> {
-            auto native_name = name;
-            if (!native_name.has_keyspace()) {
-                native_name = name.as_native_function();
-            }
-            // Collapse count(1) into countRows()
-            if (native_name == functions::function_name::native_function("count")) {
-                if (fc.args.size() == 1) {
-                    if (auto uc_arg = expr::as_if<expr::untyped_constant>(&fc.args[0])) {
-                        if (uc_arg->partial_type == expr::untyped_constant::type_class::integer
-                                && uc_arg->raw_text == "1") {
-                            return expr::function_call{
-                                .func = functions::aggregate_fcts::make_count_rows_function(),
-                                .args = {},
-                            };
-                        } else {
-                            throw exceptions::invalid_request_exception(format("count() expects a column or the literal 1 as an argument", fc.args[0]));
-                        }
-                    }
-                }
-            }
-            return std::nullopt;
-        },
-        [] (const shared_ptr<functions::function>&) -> std::optional<expression> {
-            // Already prepared, nothing to do
-            return std::nullopt;
-        },
-    }, fc.func);
-}
-
 std::optional<expression>
 prepare_function_call(const expr::function_call& fc, data_dictionary::database db, const sstring& keyspace, const schema* schema_opt, lw_shared_ptr<column_specification> receiver) {
-    if (auto prepared = try_prepare_count_rows(fc, db, keyspace, schema_opt, receiver)) {
-        return prepared;
-    }
     // Try to extract a column family name from the available information.
     // Most functions can be prepared without information about the column family, usually just the keyspace is enough.
     // One exception is the token() function - in order to prepare system.token() we have to know the partition key of the table,
@@ -1474,112 +1434,6 @@ test_assignment(const expression& expr, data_dictionary::database db, const sstr
     }, expr);
 }
 
-template <cql3_type::kind... Kinds>
-assignment_testable::vector_test_result
-test_assignment_any_size_float_vector(const expression& expr) {
-    using test_result = assignment_testable::vector_test_result;
-    const test_result NOT_ASSIGNABLE = {assignment_testable::test_result::NOT_ASSIGNABLE, std::nullopt};
-    const test_result WEAKLY_ASSIGNABLE = {assignment_testable::test_result::WEAKLY_ASSIGNABLE, std::nullopt};
-    auto is_float_or_bind = [] (const expression& e) {
-        return expr::visit(overloaded_functor{
-            [] (const bind_variable&) {
-                return true;
-            },
-            [] (const untyped_constant& uc) {
-                return uc.partial_type == untyped_constant::type_class::floating_point
-                    || uc.partial_type == untyped_constant::type_class::integer;
-            },
-            [] (const constant& value) {
-                auto kind = value.type->as_cql3_type().get_kind();
-                return cql3_type::kind_enum_set::frozen<Kinds...>().contains(kind);
-            },
-            [] (const auto&) {
-                return false;
-            },
-        }, e);
-    };
-    auto validate_assignment = [&] (const data_type& dt) -> test_result {
-         auto vt = dynamic_pointer_cast<const vector_type_impl>(dt->underlying_type());
-            if (!vt) {
-                return NOT_ASSIGNABLE;
-            }
-            auto elem_kind = vt->get_elements_type()->as_cql3_type().get_kind();
-            if (cql3_type::kind_enum_set::frozen<Kinds...>().contains(elem_kind)) {
-                return {assignment_testable::test_result::WEAKLY_ASSIGNABLE, vt->get_dimension()};
-            }
-            return NOT_ASSIGNABLE;
-    };
-    return expr::visit(overloaded_functor{
-        [&] (const constant& value) -> test_result {
-            return validate_assignment(value.type);
-        },
-        [&] (const binary_operator&) -> test_result {
-            return NOT_ASSIGNABLE;
-        },
-        [&] (const conjunction&) -> test_result {
-            return NOT_ASSIGNABLE;
-        },
-        [&] (const column_value& col_val) -> test_result {
-            return validate_assignment(col_val.col->type);
-        },
-        [&] (const subscript&) -> test_result {
-            return NOT_ASSIGNABLE;
-        },
-        [&] (const unresolved_identifier& ui) -> test_result {
-            return NOT_ASSIGNABLE;
-        },
-        [&] (const column_mutation_attribute& cma) -> test_result {
-            return NOT_ASSIGNABLE;
-        },
-        [&] (const function_call& fc) -> test_result {
-            return NOT_ASSIGNABLE;
-        },
-        [&] (const cast& c) -> test_result {
-            return NOT_ASSIGNABLE;
-        },
-        [&] (const field_selection& fs) -> test_result {
-            return NOT_ASSIGNABLE;
-        },
-        [&] (const bind_variable& bv) -> test_result {
-            return WEAKLY_ASSIGNABLE;
-        },
-        [&] (const untyped_constant& uc) -> test_result {
-            return uc.partial_type == untyped_constant::type_class::null
-                ? WEAKLY_ASSIGNABLE
-                : NOT_ASSIGNABLE;
-        },
-        [&] (const tuple_constructor& tc) -> test_result {
-            return NOT_ASSIGNABLE;
-        },
-        [&] (const collection_constructor& c) -> test_result {
-            switch (c.style) {
-            case collection_constructor::style_type::list_or_vector: {
-                if(std::ranges::all_of(c.elements, is_float_or_bind)) {
-                    return {assignment_testable::test_result::WEAKLY_ASSIGNABLE, c.elements.size()};
-                }
-                return NOT_ASSIGNABLE;
-            }
-            case collection_constructor::style_type::set: return NOT_ASSIGNABLE;
-            case collection_constructor::style_type::map: return NOT_ASSIGNABLE;
-            case collection_constructor::style_type::vector:
-                on_internal_error(expr_logger, "vector style type found in test_assignment, should have been introduced post-prepare");
-            }
-            on_internal_error(expr_logger, fmt::format("unexpected collection_constructor style {}", static_cast<unsigned>(c.style)));
-        },
-        [&] (const usertype_constructor& uc) -> test_result {
-            return NOT_ASSIGNABLE;
-        },
-        [&] (const temporary& t) -> test_result {
-            return NOT_ASSIGNABLE;
-        },
-    }, expr);
-}
-
-assignment_testable::vector_test_result
-test_assignment_any_size_float_vector(const expression& expr) {
-    return test_assignment_any_size_float_vector<cql3_type::kind::FLOAT, cql3_type::kind::DOUBLE>(expr);
-}
-
 expression
 prepare_expression(const expression& expr, data_dictionary::database db, const sstring& keyspace, const schema* schema_opt, lw_shared_ptr<column_specification> receiver) {
     auto e_opt = try_prepare_expression(expr, db, keyspace, schema_opt, std::move(receiver));
@@ -1613,9 +1467,6 @@ public:
     virtual test_result test_assignment(data_dictionary::database db, const sstring& keyspace, const schema* schema_opt, const column_specification& receiver) const override {
         return expr::test_assignment(_e, db, keyspace, schema_opt, receiver);
     }
-    virtual vector_test_result test_assignment_any_size_float_vector() const override {
-        return expr::test_assignment_any_size_float_vector(_e);
-    }
     virtual sstring assignment_testable_source_context() const override {
         return fmt::format("{}", _e);
     }

cql3/functions/functions.cc

@@ -16,7 +16,6 @@
 #include "cql3/functions/user_function.hh"
 #include "cql3/functions/user_aggregate.hh"
 #include "cql3/functions/uuid_fcts.hh"
-#include "cql3/functions/vector_similarity_fcts.hh"
 #include "data_dictionary/data_dictionary.hh"
 #include "as_json_function.hh"
 #include "cql3/prepare_context.hh"
@@ -399,14 +398,6 @@ functions::get(data_dictionary::database db,
         }
     });
 
-    const auto func_name = name.has_keyspace() ? name : name.as_native_function();
-    if (SIMILARITY_FUNCTIONS.contains(func_name)) {
-        auto arg_types = retrieve_vector_arg_types(func_name, provided_args);
-        auto fun = ::make_shared<vector_similarity_fct>(func_name.name, arg_types);
-        validate_types(db, keyspace, schema.get(), fun, provided_args, receiver_ks, receiver_cf);
-        return fun;
-    }
-
     if (name.has_keyspace()
                 ? name == TOKEN_FUNCTION_NAME
                 : name.name == TOKEN_FUNCTION_NAME.name) {

cql3/functions/vector_similarity_fcts.cc

@@ -1,184 +0,0 @@
-/*
- * Copyright (C) 2025-present ScyllaDB
- */
-
-/*
- * SPDX-License-Identifier: LicenseRef-ScyllaDB-Source-Available-1.0
- */
-
-#include "vector_similarity_fcts.hh"
-#include "types/types.hh"
-#include "types/vector.hh"
-#include "exceptions/exceptions.hh"
-#include <span>
-#include <bit>
-
-namespace cql3 {
-namespace functions {
-
-namespace detail {
-
-std::vector<float> extract_float_vector(const bytes_opt& param, vector_dimension_t dimension) {
-    if (!param) {
-        throw exceptions::invalid_request_exception("Cannot extract float vector from null parameter");
-    }
-
-    const size_t expected_size = dimension * sizeof(float);
-    if (param->size() != expected_size) {
-        throw exceptions::invalid_request_exception(
-            fmt::format("Invalid vector size: expected {} bytes for {} floats, got {} bytes",
-                       expected_size, dimension, param->size()));
-    }
-
-    std::vector<float> result;
-    result.reserve(dimension);
-
-    bytes_view view(*param);
-    for (size_t i = 0; i < dimension; ++i) {
-        // read_simple handles network byte order (big-endian) conversion
-        uint32_t raw = read_simple<uint32_t>(view);
-        result.push_back(std::bit_cast<float>(raw));
-    }
-
-    return result;
-}
-
-} // namespace detail
-
-namespace {
-
-// The computations of similarity scores match the exact formulas of Cassandra's (jVector's) implementation to ensure compatibility.
-// There exist tests checking the compliance of the results.
-// Reference:
-// https://github.com/datastax/jvector/blob/f967f1c9249035b63b55a566fac7d4dc38380349/jvector-base/src/main/java/io/github/jbellis/jvector/vector/VectorSimilarityFunction.java#L36-L69
-
-// You should only use this function if you need to preserve the original vectors and cannot normalize
-// them in advance.
-float compute_cosine_similarity(std::span<const float> v1, std::span<const float> v2) {
-    double dot_product = 0.0;
-    double squared_norm_a = 0.0;
-    double squared_norm_b = 0.0;
-
-    for (size_t i = 0; i < v1.size(); ++i) {
-        double a = v1[i];
-        double b = v2[i];
-
-        dot_product += a * b;
-        squared_norm_a += a * a;
-        squared_norm_b += b * b;
-    }
-
-    if (squared_norm_a == 0 || squared_norm_b == 0) {
-        return std::numeric_limits<float>::quiet_NaN();
-    }
-
-    // The cosine similarity is in the range [-1, 1].
-    // It is mapped to a similarity score in the range [0, 1] (-1 -> 0, 1 -> 1)
-    // for consistency with other similarity functions.
-    return (1 + (dot_product / (std::sqrt(squared_norm_a * squared_norm_b)))) / 2;
-}
-
-float compute_euclidean_similarity(std::span<const float> v1, std::span<const float> v2) {
-    double sum = 0.0;
-
-    for (size_t i = 0; i < v1.size(); ++i) {
-        double a = v1[i];
-        double b = v2[i];
-
-        double diff = a - b;
-        sum += diff * diff;
-    }
-
-    // The squared Euclidean (L2) distance is of range [0, inf).
-    // It is mapped to a similarity score in the range (0, 1] (0 -> 1, inf -> 0)
-    // for consistency with other similarity functions.
-    return (1 / (1 + sum));
-}
-
-// Assumes that both vectors are L2-normalized.
-// This similarity is intended as an optimized way to perform cosine similarity calculation.
-float compute_dot_product_similarity(std::span<const float> v1, std::span<const float> v2) {
-    double dot_product = 0.0;
-
-    for (size_t i = 0; i < v1.size(); ++i) {
-        double a = v1[i];
-        double b = v2[i];
-        dot_product += a * b;
-    }
-
-    // The dot product is in the range [-1, 1] for L2-normalized vectors.
-    // It is mapped to a similarity score in the range [0, 1] (-1 -> 0, 1 -> 1)
-    // for consistency with other similarity functions.
-    return ((1 + dot_product) / 2);
-}
-
-} // namespace
-
-thread_local const std::unordered_map<function_name, similarity_function_t> SIMILARITY_FUNCTIONS = {
-        {SIMILARITY_COSINE_FUNCTION_NAME, compute_cosine_similarity},
-        {SIMILARITY_EUCLIDEAN_FUNCTION_NAME, compute_euclidean_similarity},
-        {SIMILARITY_DOT_PRODUCT_FUNCTION_NAME, compute_dot_product_similarity},
-};
-
-std::vector<data_type> retrieve_vector_arg_types(const function_name& name, const std::vector<shared_ptr<assignment_testable>>& provided_args) {
-    if (provided_args.size() != 2) {
-        throw exceptions::invalid_request_exception(fmt::format("Invalid number of arguments for function {}(vector<float, n>, vector<float, n>)", name));
-    }
-
-    auto [first_result, first_dim_opt] = provided_args[0]->test_assignment_any_size_float_vector();
-    auto [second_result, second_dim_opt] = provided_args[1]->test_assignment_any_size_float_vector();
-
-    auto invalid_type_error_message = [&name](const shared_ptr<assignment_testable>& arg) {
-        auto type = arg->assignment_testable_type_opt();
-        const auto& source_context = arg->assignment_testable_source_context();
-        if (type) {
-            return fmt::format("Function {} requires a float vector argument, but found {} of type {}", name, source_context, type.value()->cql3_type_name());
-        } else {
-            return fmt::format("Function {} requires a float vector argument, but found {}", name, source_context);
-        }
-    };
-
-    if (!is_assignable(first_result)) {
-        throw exceptions::invalid_request_exception(invalid_type_error_message(provided_args[0]));
-    }
-    if (!is_assignable(second_result)) {
-        throw exceptions::invalid_request_exception(invalid_type_error_message(provided_args[1]));
-    }
-
-    if (!first_dim_opt && !second_dim_opt) {
-        throw exceptions::invalid_request_exception(fmt::format("Cannot infer type of argument {} for function {}(vector<float, n>, vector<float, n>)",
-                provided_args[0]->assignment_testable_source_context(), name));
-    }
-    if (first_dim_opt && second_dim_opt) {
-        if (*first_dim_opt != *second_dim_opt) {
-            throw exceptions::invalid_request_exception(fmt::format(
-                    "All arguments must have the same vector dimensions, but found vector<float, {}> and vector<float, {}>", *first_dim_opt, *second_dim_opt));
-        }
-    }
-
-    vector_dimension_t dimension = first_dim_opt ? *first_dim_opt : *second_dim_opt;
-    auto type = vector_type_impl::get_instance(float_type, dimension);
-    return {type, type};
-}
-
-bytes_opt vector_similarity_fct::execute(std::span<const bytes_opt> parameters) {
-    if (std::any_of(parameters.begin(), parameters.end(), [](const auto& param) {
-            return !param;
-        })) {
-        return std::nullopt;
-    }
-
-    // Extract dimension from the vector type
-    const auto& type = static_cast<const vector_type_impl&>(*arg_types()[0]);
-    vector_dimension_t dimension = type.get_dimension();
-
-    // Optimized path: extract floats directly from bytes, bypassing data_value overhead
-    std::vector<float> v1 = detail::extract_float_vector(parameters[0], dimension);
-    std::vector<float> v2 = detail::extract_float_vector(parameters[1], dimension);
-
-    float result = SIMILARITY_FUNCTIONS.at(_name)(v1, v2);
-    return float_type->decompose(result);
-}
-
-} // namespace functions
-} // namespace cql3