25.8 Stable: Bump to 25.8.24

.github/workflows/regression-reusable-suite.yml

@@ -81,7 +81,7 @@ jobs:
       AWS_DEFAULT_REGION: ${{ secrets.AWS_DEFAULT_REGION }}
       # Docker credentials
       DOCKER_USERNAME: ${{ secrets.DOCKER_USERNAME }}
-      DOCKER_PASSWORD: ${{ secrets.DOCKER_PASSWORD }}
+      DOCKER_PASSWORD: ${{ secrets.DOCKER_TOKEN }}
       # Database credentials
       CHECKS_DATABASE_HOST: ${{ secrets.CHECKS_DATABASE_HOST }}
       CHECKS_DATABASE_USER: ${{ secrets.CLICKHOUSE_TEST_STAT_LOGIN }}

.github/workflows/regression.yml

@@ -64,8 +64,8 @@ name: Regression test workflow - Release
       DOCKER_USERNAME:
         description: username of the docker user.
         required: true
-      DOCKER_PASSWORD:
-        description: password to the docker user.
+      DOCKER_TOKEN:
+        description: token of the docker user.
         required: true
       REGRESSION_AWS_S3_BUCKET:
         description: aws s3 bucket used for regression tests.
@@ -96,7 +96,7 @@ env:
   AWS_SECRET_ACCESS_KEY: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
   AWS_DEFAULT_REGION: ${{ secrets.AWS_DEFAULT_REGION }}
   DOCKER_USERNAME: ${{ secrets.DOCKER_USERNAME }}
-  DOCKER_PASSWORD: ${{ secrets.DOCKER_PASSWORD }}
+  DOCKER_PASSWORD: ${{ secrets.DOCKER_TOKEN }}
   CHECKS_DATABASE_HOST: ${{ secrets.CHECKS_DATABASE_HOST }}
   CHECKS_DATABASE_USER: ${{ secrets.CLICKHOUSE_TEST_STAT_LOGIN }}
   CHECKS_DATABASE_PASSWORD:  ${{ secrets.CLICKHOUSE_TEST_STAT_PASSWORD }}

cmake/autogenerated_versions.txt

@@ -2,13 +2,13 @@
 
 # NOTE: VERSION_REVISION has nothing common with DBMS_TCP_PROTOCOL_VERSION,
 # only DBMS_TCP_PROTOCOL_VERSION should be incremented on protocol changes.
-SET(VERSION_REVISION 54523)
+SET(VERSION_REVISION 54524)
 SET(VERSION_MAJOR 25)
 SET(VERSION_MINOR 8)
-SET(VERSION_PATCH 23)
-SET(VERSION_GITHASH 6d29497525664acca46a1d1cd0d5787e9ad0857d)
-SET(VERSION_DESCRIBE v25.8.23.10001.altinitytest)
-SET(VERSION_STRING 25.8.23.10001.altinitytest)
+SET(VERSION_PATCH 24)
+SET(VERSION_GITHASH 443620d152ed64184d6f221c988c569334311055)
+SET(VERSION_DESCRIBE v25.8.24.10001.altinitytest)
+SET(VERSION_STRING 25.8.24.10001.altinitytest)
 # end of autochange
 
 SET(VERSION_TWEAK 10001)

contrib/libarchive-cmake/config.h

@@ -1031,6 +1031,9 @@ typedef uint64_t uintmax_t;
 /* Define to 1 if you have the <stdint.h> header file. */
 #define HAVE_STDINT_H 1
 
+/* Define to 1 if you have the <stdio.h> header file. */
+#define HAVE_STDIO_H 1
+
 /* Define to 1 if you have the <stdlib.h> header file. */
 #define HAVE_STDLIB_H 1
 

src/AggregateFunctions/QuantileTiming.h

@@ -391,6 +391,8 @@ namespace detail
                     readBinaryLittleEndian(index, buf);
                     if (index == BIG_THRESHOLD)
                         break;
+                    if (index - SMALL_THRESHOLD >= BIG_SIZE)
+                        throw Exception(ErrorCodes::INCORRECT_DATA, "Incorrect index {} in 'large' kind of quantileTiming deserialization", index);
 
                     UInt64 elem_count = 0;
                     readBinaryLittleEndian(elem_count, buf);

src/Columns/ColumnObject.cpp

@@ -531,6 +531,7 @@ bool ColumnObject::tryInsert(const Field & x)
         for (const auto & path : new_dynamic_paths)
         {
             dynamic_paths_ptrs.erase(path);
+            sorted_dynamic_paths.erase(path);
             dynamic_paths.erase(path);
         }
 
@@ -566,6 +567,7 @@ bool ColumnObject::tryInsert(const Field & x)
         }
         else if (auto * dynamic_path_column = tryToAddNewDynamicPath(path))
         {
+            new_dynamic_paths.insert(String(path));
             if (!dynamic_path_column->tryInsert(value_field))
             {
                 restore_sizes();

src/Columns/tests/gtest_column_object.cpp

@@ -444,3 +444,50 @@ TEST(ColumnObject, RepairDuplicatesInDynamicPathsAndSharedData)
     ASSERT_EQ((*column_object)[2], (Object{{"b", Field(1u)}, {"c", Field(1u)}, {"d", Field(1u)}}));
     ASSERT_EQ((*column_object)[3], (Object{{"d", Field(1u)}}));
 }
+
+TEST(ColumnObject, TryInsertRestoresSortedDynamicPaths)
+{
+    /// "b" is a typed UInt32 path; everything else becomes a dynamic path.
+    auto type = DataTypeFactory::instance().get("JSON(max_dynamic_types=10, max_dynamic_paths=10, b UInt32)");
+    auto col = type->createColumn();
+    auto & col_object = assert_cast<ColumnObject &>(*col);
+    const auto & dynamic_paths = col_object.getDynamicPaths();
+
+    /// One valid row so we have something to serialize later.
+    col_object.insert(Object{{"b", Field{5u}}});
+    ASSERT_EQ(col_object.size(), 1u);
+    ASSERT_EQ(dynamic_paths.size(), 0u);
+
+    /// tryInsert with Object{"a_new": 1, "b": "not_a_number"}.
+    /// Fields are processed in alphabetical order ("a_new" before "b"), so:
+    ///   1. "a_new" is new → tryToAddNewDynamicPath succeeds, "a_new" is added to all
+    ///      three structures including sorted_dynamic_paths.
+    ///   2. ColumnDynamic::tryInsert(1u) for "a_new" succeeds.
+    ///   3. ColumnUInt32::tryInsert(String) for "b" returns false.
+    ///   4. restore_sizes() is called.
+    ///      BUG 1: new_dynamic_paths was never populated, so the loop that removes
+    ///             newly-added paths is a no-op — "a_new" is left in dynamic_paths,
+    ///             dynamic_paths_ptrs, and sorted_dynamic_paths with a rolled-back size.
+    ///      BUG 2: even after populating new_dynamic_paths, the old code erased from
+    ///             dynamic_paths before sorted_dynamic_paths, leaving a dangling view.
+    ///      FIX:   record the path in new_dynamic_paths immediately after
+    ///             tryToAddNewDynamicPath succeeds, and erase sorted_dynamic_paths
+    ///             before dynamic_paths.
+    bool result = col_object.tryInsert(Object{{"a_new", Field{1u}}, {"b", Field{String("not_a_number")}}});
+    ASSERT_FALSE(result);
+
+    ASSERT_EQ(col_object.size(), 1u);
+    ASSERT_EQ(dynamic_paths.size(), 0u);   /// "a_new" must be fully rolled back
+
+    /// serializeValueIntoArena iterates sorted_dynamic_paths.
+    /// Without the fix the stale "a_new" entry causes undefined behavior here (see comment above).
+    Arena arena;
+    const char * begin = nullptr;
+    auto ref = col_object.serializeValueIntoArena(0, arena, begin, nullptr);
+
+    /// Round-trip sanity check.
+    ReadBufferFromMemory buf(ref.data, ref.size);
+    col_object.deserializeAndInsertFromArena(buf, nullptr);
+    ASSERT_EQ(col_object.size(), 2u);
+    ASSERT_EQ(col_object[1], col_object[0]);
+}

src/Coordination/KeeperStorage.cpp

@@ -1789,8 +1789,7 @@ std::list<KeeperStorageBase::Delta> preprocess(
     ProfileEvents::increment(ProfileEvents::KeeperGetRequest);
 
     if (zk_request.path == Coordination::keeper_api_feature_flags_path
-        || zk_request.path == Coordination::keeper_config_path
-        || zk_request.path == Coordination::keeper_availability_zone_path)
+        || zk_request.path == Coordination::keeper_config_path)
         return {};
 
     if (!storage.uncommitted_state.getNode(zk_request.path))

src/Core/PostgreSQL/insertPostgreSQLValue.cpp

@@ -149,6 +149,9 @@ void insertPostgreSQLValue(
                 {
                     max_dimension = std::max(max_dimension, dimension);
 
+                    if (dimension == 0)
+                        throw Exception(ErrorCodes::BAD_ARGUMENTS, "Unexpected array closing bracket");
+
                     --dimension;
                     if (dimension == 0)
                         break;

src/Core/ServerSettings.cpp

@@ -1041,16 +1041,16 @@ The policy on how to perform a scheduling of CPU slots specified by `concurrent_
     Disable all insert/alter/delete queries. This setting will be enabled if someone needs read-only nodes to prevent insertion and mutation affect reading performance.
     )", 0) \
     DECLARE(UInt64, parts_kill_delay_period, 30, R"(
-    Period to completely remove parts for SharedMergeTree. Only available in ClickHouse Cloud
+    Period to completely remove parts for SharedMergeTree.
     )", 0) \
     DECLARE(UInt64, parts_kill_delay_period_random_add, 10, R"(
-    Add uniformly distributed value from 0 to x seconds to kill_delay_period to avoid thundering herd effect and subsequent DoS of ZooKeeper in case of very large number of tables. Only available in ClickHouse Cloud
+    Add uniformly distributed value from 0 to x seconds to kill_delay_period to avoid thundering herd effect and subsequent DoS of ZooKeeper in case of very large number of tables.
     )", 0) \
     DECLARE(UInt64, parts_killer_pool_size, 128, R"(
-    Threads for cleanup of shared merge tree outdated threads. Only available in ClickHouse Cloud
+    Threads for cleanup of shared merge tree outdated threads.
     )", 0) \
     DECLARE(UInt64, keeper_multiread_batch_size, 10'000, R"(
-    Maximum size of batch for MultiRead request to [Zoo]Keeper that support batching. If set to 0, batching is disabled. Available only in ClickHouse Cloud.
+    Maximum size of batch for MultiRead request to [Zoo]Keeper that support batching. If set to 0, batching is disabled.
     )", 0) \
     DECLARE(String, license_key, "", "License key for ClickHouse Enterprise Edition", 0) \
     DECLARE(NonZeroUInt64, prefetch_threadpool_pool_size, 100, R"(Size of background pool for prefetches for remote object storages)", 0) \
@@ -1123,7 +1123,7 @@ The policy on how to perform a scheduling of CPU slots specified by `concurrent_
     <process_query_plan_packet>true</process_query_plan_packet>
     ```
     )", 0) \
-    DECLARE(Bool, storage_shared_set_join_use_inner_uuid, true, "If enabled, an inner UUID is generated during the creation of SharedSet and SharedJoin. ClickHouse Cloud only", 0) \
+    DECLARE(Bool, storage_shared_set_join_use_inner_uuid, true, "If enabled, an inner UUID is generated during the creation of SharedSet and SharedJoin.", 0) \
     DECLARE(UInt64, startup_mv_delay_ms, 0, R"(Debug parameter to simulate materizlied view creation delay)", 0) \
     DECLARE(UInt64, os_cpu_busy_time_threshold, 1'000'000, "Threshold of OS CPU busy time in microseconds (OSCPUVirtualTimeMicroseconds metric) to consider CPU doing some useful work, no CPU overload would be considered if busy time was below this value.", 0) \
     DECLARE(Float, min_os_cpu_wait_time_ratio_to_drop_connection, 0, R"(
@@ -1283,7 +1283,7 @@ void ServerSettings::dumpToSystemServerSettingsColumns(ServerSettingColumnsParam
             {"mark_cache_size", {std::to_string(context->getMarkCache()->maxSizeInBytes()), ChangeableWithoutRestart::Yes}},
             {"uncompressed_cache_size", {std::to_string(context->getUncompressedCache()->maxSizeInBytes()), ChangeableWithoutRestart::Yes}},
             {"index_mark_cache_size", {std::to_string(context->getIndexMarkCache()->maxSizeInBytes()), ChangeableWithoutRestart::Yes}},
-            {"index_uncompressed_cache_size", {std::to_string(context->getIndexUncompressedCache()->maxSizeInBytes()), ChangeableWithoutRestart::Yes}},
+            {"index_uncompressed_cache_size", {std::to_string(context->getIndexUncompressedCache(/*only_if_enabled=*/ false)->maxSizeInBytes()), ChangeableWithoutRestart::Yes}},
             {"mmap_cache_size", {std::to_string(context->getMMappedFileCache()->maxSizeInBytes()), ChangeableWithoutRestart::Yes}},
             {"query_condition_cache_size", {std::to_string(context->getQueryConditionCache()->maxSizeInBytes()), ChangeableWithoutRestart::Yes}},