Merge branch 'master' into use_minmax_index

.clang-tidy

@@ -16,7 +16,6 @@ Checks: '-*,
     modernize-make-unique,
     modernize-raw-string-literal,
     modernize-redundant-void-arg,
-    modernize-replace-auto-ptr,
     modernize-replace-random-shuffle,
     modernize-use-bool-literals,
     modernize-use-nullptr,
@@ -145,6 +144,7 @@ Checks: '-*,
     clang-analyzer-cplusplus.SelfAssignment,
     clang-analyzer-deadcode.DeadStores,
     clang-analyzer-cplusplus.Move,
+    clang-analyzer-optin.cplusplus.UninitializedObject,
     clang-analyzer-optin.cplusplus.VirtualCall,
     clang-analyzer-security.insecureAPI.UncheckedReturn,
     clang-analyzer-security.insecureAPI.bcmp,
@@ -164,6 +164,8 @@ Checks: '-*,
     clang-analyzer-unix.cstring.NullArg,
 
     boost-use-to-string,
+
+    alpha.security.cert.env.InvalidPtr,
 '
 WarningsAsErrors: '*'
 

.github/PULL_REQUEST_TEMPLATE.md

@@ -1,4 +1,4 @@
-Changelog category (leave one):
+### Changelog category (leave one):
 - New Feature
 - Improvement
 - Bug Fix (user-visible misbehaviour in official stable or prestable release)
@@ -9,7 +9,7 @@ Changelog category (leave one):
 - Not for changelog (changelog entry is not required)
 
 
-Changelog entry (a user-readable short description of the changes that goes to CHANGELOG.md):
+### Changelog entry (a user-readable short description of the changes that goes to CHANGELOG.md):
 ...
 
 

.github/workflows/backport_branches.yml

@@ -341,10 +341,15 @@ jobs:
     steps:
       - name: Set envs
         run: |
+          DEPENDENCIES=$(cat << 'EOF' | jq '. | length'
+          ${{ toJSON(needs) }}
+          EOF
+          )
+          echo "DEPENDENCIES=$DEPENDENCIES" >> "$GITHUB_ENV"
           cat >> "$GITHUB_ENV" << 'EOF'
-          TEMP_PATH=${{runner.temp}}/report_check
-          REPORTS_PATH=${{runner.temp}}/reports_dir
           CHECK_NAME=ClickHouse build check (actions)
+          REPORTS_PATH=${{runner.temp}}/reports_dir
+          TEMP_PATH=${{runner.temp}}/report_check
           EOF
       - name: Download json reports
         uses: actions/download-artifact@v2
@@ -360,7 +365,7 @@ jobs:
           sudo rm -fr "$TEMP_PATH"
           mkdir -p "$TEMP_PATH"
           cd "$GITHUB_WORKSPACE/tests/ci"
-          python3 build_report_check.py "$CHECK_NAME"
+          python3 build_report_check.py "$CHECK_NAME" "$DEPENDENCIES"
       - name: Cleanup
         if: always()
         run: |

.github/workflows/master.yml

@@ -947,6 +947,34 @@ jobs:
           docker rm -f "$(docker ps -a -q)" ||:
           sudo rm -fr "$TEMP_PATH" "$CACHES_PATH"
 ############################################################################################
+##################################### Docker images  #######################################
+############################################################################################
+  DockerServerImages:
+    needs:
+      - BuilderDebRelease
+      - BuilderDebAarch64
+    runs-on: [self-hosted, style-checker]
+    steps:
+      - name: Clear repository
+        run: |
+          sudo rm -fr "$GITHUB_WORKSPACE" && mkdir "$GITHUB_WORKSPACE"
+      - name: Check out repository code
+        uses: actions/checkout@v2
+        with:
+          fetch-depth: 0  # otherwise we will have no version info
+      - name: Check docker clickhouse/clickhouse-server building
+        run: |
+          cd "$GITHUB_WORKSPACE/tests/ci"
+          python3 docker_server.py --release-type head
+          python3 docker_server.py --release-type head --no-ubuntu \
+            --image-repo clickhouse/clickhouse-keeper --image-path docker/keeper
+      - name: Cleanup
+        if: always()
+        run: |
+          docker kill "$(docker ps -q)" ||:
+          docker rm -f "$(docker ps -a -q)" ||:
+          sudo rm -fr "$TEMP_PATH"
+############################################################################################
 ##################################### BUILD REPORTER #######################################
 ############################################################################################
   BuilderReport:
@@ -964,10 +992,16 @@ jobs:
     steps:
       - name: Set envs
         run: |
+          DEPENDENCIES=$(cat << 'EOF' | jq '. | length'
+          ${{ toJSON(needs) }}
+          EOF
+          )
+          echo "DEPENDENCIES=$DEPENDENCIES" >> "$GITHUB_ENV"
           cat >> "$GITHUB_ENV" << 'EOF'
-          TEMP_PATH=${{runner.temp}}/report_check
-          REPORTS_PATH=${{runner.temp}}/reports_dir
           CHECK_NAME=ClickHouse build check (actions)
+          REPORTS_PATH=${{runner.temp}}/reports_dir
+          REPORTS_PATH=${{runner.temp}}/reports_dir
+          TEMP_PATH=${{runner.temp}}/report_check
           EOF
       - name: Download json reports
         uses: actions/download-artifact@v2
@@ -983,7 +1017,7 @@ jobs:
           sudo rm -fr "$TEMP_PATH"
           mkdir -p "$TEMP_PATH"
           cd "$GITHUB_WORKSPACE/tests/ci"
-          python3 build_report_check.py "$CHECK_NAME"
+          python3 build_report_check.py "$CHECK_NAME" "$DEPENDENCIES"
       - name: Cleanup
         if: always()
         run: |

.github/workflows/nightly.yml

@@ -72,3 +72,52 @@ jobs:
         with:
           name: changed_images
           path: ${{ runner.temp }}/changed_images.json
+  BuilderCoverity:
+    needs: DockerHubPush
+    runs-on: [self-hosted, builder]
+    steps:
+      - name: Set envs
+        run: |
+          cat >> "$GITHUB_ENV" << 'EOF'
+          TEMP_PATH=${{runner.temp}}/build_check
+          IMAGES_PATH=${{runner.temp}}/images_path
+          REPO_COPY=${{runner.temp}}/build_check/ClickHouse
+          CACHES_PATH=${{runner.temp}}/../ccaches
+          CHECK_NAME=ClickHouse build check (actions)
+          BUILD_NAME=coverity
+          EOF
+      - name: Download changed images
+        uses: actions/download-artifact@v2
+        with:
+          name: changed_images
+          path: ${{ env.IMAGES_PATH }}
+      - name: Clear repository
+        run: |
+          sudo rm -fr "$GITHUB_WORKSPACE" && mkdir "$GITHUB_WORKSPACE"
+      - name: Check out repository code
+        id: coverity-checkout
+        uses: actions/checkout@v2
+        with:
+          submodules: 'true'
+          fetch-depth: 0 # otherwise we will have no info about contributors
+      - name: Build
+        run: |
+          sudo rm -fr "$TEMP_PATH"
+          mkdir -p "$TEMP_PATH"
+          cp -r "$GITHUB_WORKSPACE" "$TEMP_PATH"
+          cd "$REPO_COPY/tests/ci" && python3 build_check.py "$CHECK_NAME" "$BUILD_NAME" "${{ secrets.COV_TOKEN }}"
+      - name: Upload Coverity Analysis
+        if: ${{ success() || failure() }}
+        run: |
+          curl --form token='${{ secrets.COV_TOKEN }}' \
+            --form email='${{ secrets.ROBOT_CLICKHOUSE_EMAIL }}' \
+            --form file="@$TEMP_PATH/$BUILD_NAME/clickhouse-scan.tgz" \
+            --form version="${GITHUB_REF#refs/heads/}-${GITHUB_SHA::6}" \
+            --form description="Nighly Scan: $(date +'%Y-%m-%dT%H:%M:%S')" \
+            https://scan.coverity.com/builds?project=ClickHouse%2FClickHouse
+      - name: Cleanup
+        if: always()
+        run: |
+          docker kill "$(docker ps -q)" ||:
+          docker rm -f "$(docker ps -a -q)" ||:
+          sudo rm -fr "$TEMP_PATH" "$CACHES_PATH"

.github/workflows/pull_request.yml

@@ -4,7 +4,7 @@ env:
   # Force the stdout and stderr streams to be unbuffered
   PYTHONUNBUFFERED: 1
 
-on: # yamllint disable-line rule:truthy
+on:  # yamllint disable-line rule:truthy
   pull_request:
     types:
       - synchronize
@@ -998,6 +998,34 @@ jobs:
           docker rm -f "$(docker ps -a -q)" ||:
           sudo rm -fr "$TEMP_PATH" "$CACHES_PATH"
 ############################################################################################
+##################################### Docker images  #######################################
+############################################################################################
+  DockerServerImages:
+    needs:
+      - BuilderDebRelease
+      - BuilderDebAarch64
+    runs-on: [self-hosted, style-checker]
+    steps:
+      - name: Clear repository
+        run: |
+          sudo rm -fr "$GITHUB_WORKSPACE" && mkdir "$GITHUB_WORKSPACE"
+      - name: Check out repository code
+        uses: actions/checkout@v2
+        with:
+          fetch-depth: 0  # otherwise we will have no version info
+      - name: Check docker clickhouse/clickhouse-server building
+        run: |
+          cd "$GITHUB_WORKSPACE/tests/ci"
+          python3 docker_server.py --release-type head --no-push
+          python3 docker_server.py --release-type head --no-push --no-ubuntu \
+            --image-repo clickhouse/clickhouse-keeper --image-path docker/keeper
+      - name: Cleanup
+        if: always()
+        run: |
+          docker kill "$(docker ps -q)" ||:
+          docker rm -f "$(docker ps -a -q)" ||:
+          sudo rm -fr "$TEMP_PATH"
+############################################################################################
 ##################################### BUILD REPORTER #######################################
 ############################################################################################
   BuilderReport:
@@ -1016,10 +1044,16 @@ jobs:
     steps:
       - name: Set envs
         run: |
+          DEPENDENCIES=$(cat << 'EOF' | jq '. | length'
+          ${{ toJSON(needs) }}
+          EOF
+          )
+          echo "DEPENDENCIES=$DEPENDENCIES" >> "$GITHUB_ENV"
           cat >> "$GITHUB_ENV" << 'EOF'
-          TEMP_PATH=${{runner.temp}}/report_check
-          REPORTS_PATH=${{runner.temp}}/reports_dir
           CHECK_NAME=ClickHouse build check (actions)
+          REPORTS_PATH=${{runner.temp}}/reports_dir
+          REPORTS_PATH=${{runner.temp}}/reports_dir
+          TEMP_PATH=${{runner.temp}}/report_check
           EOF
       - name: Download json reports
         uses: actions/download-artifact@v2
@@ -1035,7 +1069,7 @@ jobs:
           sudo rm -fr "$TEMP_PATH"
           mkdir -p "$TEMP_PATH"
           cd "$GITHUB_WORKSPACE/tests/ci"
-          python3 build_report_check.py "$CHECK_NAME"
+          python3 build_report_check.py "$CHECK_NAME" "$DEPENDENCIES"
       - name: Cleanup
         if: always()
         run: |
@@ -3138,6 +3172,7 @@ jobs:
     needs:
       - StyleCheck
       - DockerHubPush
+      - DockerServerImages
       - CheckLabels
       - BuilderReport
       - FastTest

.github/workflows/release.yml

@@ -36,3 +36,28 @@ jobs:
         overwrite: true
         tag: ${{ github.ref }}
         file_glob: true
+  ############################################################################################
+  ##################################### Docker images  #######################################
+  ############################################################################################
+  DockerServerImages:
+    runs-on: [self-hosted, style-checker]
+    steps:
+    - name: Clear repository
+      run: |
+        sudo rm -fr "$GITHUB_WORKSPACE" && mkdir "$GITHUB_WORKSPACE"
+    - name: Check out repository code
+      uses: actions/checkout@v2
+      with:
+        fetch-depth: 0  # otherwise we will have no version info
+    - name: Check docker clickhouse/clickhouse-server building
+      run: |
+        cd "$GITHUB_WORKSPACE/tests/ci"
+        python3 docker_server.py --release-type auto
+        python3 docker_server.py --release-type auto --no-ubuntu \
+          --image-repo clickhouse/clickhouse-keeper --image-path docker/keeper
+    - name: Cleanup
+      if: always()
+      run: |
+        docker kill "$(docker ps -q)" ||:
+        docker rm -f "$(docker ps -a -q)" ||:
+        sudo rm -fr "$TEMP_PATH"

.github/workflows/release_branches.yml

@@ -436,10 +436,16 @@ jobs:
     steps:
       - name: Set envs
         run: |
+          DEPENDENCIES=$(cat << 'EOF' | jq '. | length'
+          ${{ toJSON(needs) }}
+          EOF
+          )
+          echo "DEPENDENCIES=$DEPENDENCIES" >> "$GITHUB_ENV"
           cat >> "$GITHUB_ENV" << 'EOF'
-          TEMP_PATH=${{runner.temp}}/report_check
-          REPORTS_PATH=${{runner.temp}}/reports_dir
           CHECK_NAME=ClickHouse build check (actions)
+          REPORTS_PATH=${{runner.temp}}/reports_dir
+          REPORTS_PATH=${{runner.temp}}/reports_dir
+          TEMP_PATH=${{runner.temp}}/report_check
           EOF
       - name: Download json reports
         uses: actions/download-artifact@v2
@@ -455,7 +461,7 @@ jobs:
           sudo rm -fr "$TEMP_PATH"
           mkdir -p "$TEMP_PATH"
           cd "$GITHUB_WORKSPACE/tests/ci"
-          python3 build_report_check.py "$CHECK_NAME"
+          python3 build_report_check.py "$CHECK_NAME" "$DEPENDENCIES"
       - name: Cleanup
         if: always()
         run: |

CMakeLists.txt

@@ -261,8 +261,8 @@ endif ()
 # Add a section with the hash of the compiled machine code for integrity checks.
 # Only for official builds, because adding a section can be time consuming (rewrite of several GB).
 # And cross compiled binaries are not supported (since you cannot execute clickhouse hash-binary)
-if (OBJCOPY_PATH AND CLICKHOUSE_OFFICIAL_BUILD AND (NOT CMAKE_TOOLCHAIN_FILE))
+if (OBJCOPY_PATH AND CLICKHOUSE_OFFICIAL_BUILD AND (NOT CMAKE_TOOLCHAIN_FILE OR CMAKE_TOOLCHAIN_FILE MATCHES "linux/toolchain-x86_64.cmake$"))
-    set (USE_BINARY_HASH 1)
+    set (USE_BINARY_HASH 1 CACHE STRING "Calculate binary hash and store it in the separate section")
 endif ()
 
 # Allows to build stripped binary in a separate directory
@@ -294,14 +294,19 @@ include(cmake/cpu_features.cmake)
 # Enable it explicitly.
 set (COMPILER_FLAGS "${COMPILER_FLAGS} -fasynchronous-unwind-tables")
 
-# Reproducible builds
+# Reproducible builds.
-# If turned `ON`, remap file source paths in debug info, predefined preprocessor macros and __builtin_FILE().
+if (CMAKE_BUILD_TYPE_UC STREQUAL "DEBUG")
-option(ENABLE_BUILD_PATH_MAPPING "Enable remap file source paths in debug info, predefined preprocessor macros and __builtin_FILE(). It's to generate reproducible builds. See https://reproducible-builds.org/docs/build-path" ON)
+    set (ENABLE_BUILD_PATH_MAPPING_DEFAULT OFF)
+else ()
+    set (ENABLE_BUILD_PATH_MAPPING_DEFAULT ON)
+endif ()
+
+option (ENABLE_BUILD_PATH_MAPPING "Enable remapping of file source paths in debug info, predefined preprocessor macros, and __builtin_FILE(). It's used to generate reproducible builds. See https://reproducible-builds.org/docs/build-path" ${ENABLE_BUILD_PATH_MAPPING_DEFAULT})
 
 if (ENABLE_BUILD_PATH_MAPPING)
     set (COMPILER_FLAGS "${COMPILER_FLAGS} -ffile-prefix-map=${CMAKE_SOURCE_DIR}=.")
     set (CMAKE_ASM_FLAGS "${CMAKE_ASM_FLAGS} -ffile-prefix-map=${CMAKE_SOURCE_DIR}=.")
-endif()
+endif ()
 
 if (${CMAKE_VERSION} VERSION_LESS "3.12.4")
     # CMake < 3.12 doesn't support setting 20 as a C++ standard version.

base/base/CMakeLists.txt

@@ -2,6 +2,7 @@ set (SRCS
     argsToConfig.cpp
     coverage.cpp
     demangle.cpp
+    getAvailableMemoryAmount.cpp
     getFQDNOrHostName.cpp
     getMemoryAmount.cpp
     getPageSize.cpp

base/base/getAvailableMemoryAmount.cpp

@@ -0,0 +1,44 @@
+#include <stdexcept>
+#include <fstream>
+#include <base/getAvailableMemoryAmount.h>
+#include <base/getPageSize.h>
+
+#include <unistd.h>
+#include <sys/types.h>
+#include <sys/param.h>
+#if defined(BSD)
+#include <sys/sysctl.h>
+#include <sys/vmmeter.h>
+#endif
+
+
+uint64_t getAvailableMemoryAmountOrZero()
+{
+#if defined(_SC_AVPHYS_PAGES) // linux
+    return getPageSize() * sysconf(_SC_AVPHYS_PAGES);
+#elif defined(__FreeBSD__)
+    struct vmtotal vmt;
+    size_t vmt_size = sizeof(vmt);
+    if (sysctlbyname("vm.vmtotal", &vmt, &vmt_size, NULL, 0) == 0)
+        return getPageSize() * vmt.t_avm;
+    else
+        return 0;
+#else // darwin
+    unsigned int usermem;
+    size_t len = sizeof(usermem);
+    static int mib[2] = { CTL_HW, HW_USERMEM };
+    if (sysctl(mib, 2, &usermem, &len, nullptr, 0) == 0 && len == sizeof(usermem))
+        return usermem;
+    else
+        return 0;
+#endif
+}
+
+
+uint64_t getAvailableMemoryAmount()
+{
+    auto res = getAvailableMemoryAmountOrZero();
+    if (!res)
+        throw std::runtime_error("Cannot determine available memory amount");
+    return res;
+}

base/base/getAvailableMemoryAmount.h

@@ -0,0 +1,12 @@
+#pragma once
+
+#include <cstdint>
+
+/** Returns the size of currently available physical memory (RAM) in bytes.
+  * Returns 0 on unsupported platform or if it cannot determine the size of physical memory.
+  */
+uint64_t getAvailableMemoryAmountOrZero();
+
+/** Throws exception if it cannot determine the size of physical memory.
+  */
+uint64_t getAvailableMemoryAmount();

base/loggers/Loggers.cpp

@@ -197,7 +197,6 @@ void Loggers::buildLoggers(Poco::Util::AbstractConfiguration & config, Poco::Log
 
         Poco::AutoPtr<OwnPatternFormatter> pf = new OwnPatternFormatter(color_enabled);
         Poco::AutoPtr<DB::OwnFormattingChannel> log = new DB::OwnFormattingChannel(pf, new Poco::ConsoleChannel);
-        logger.warning("Logging " + console_log_level_string + " to console");
         log->setLevel(console_log_level);
         split->addChannel(log, "console");
     }

contrib/krb5-cmake/CMakeLists.txt

@@ -1,4 +1,4 @@
-set (ENABLE_KRB5_DEFAULT 1)
+set (ENABLE_KRB5_DEFAULT ${ENABLE_LIBRARIES})
 if (NOT CMAKE_SYSTEM_NAME MATCHES "Linux" AND NOT (CMAKE_SYSTEM_NAME MATCHES "Darwin" AND NOT CMAKE_CROSSCOMPILING))
     message (WARNING "krb5 disabled in non-Linux and non-native-Darwin environments")
     set (ENABLE_KRB5_DEFAULT 0)
@@ -16,6 +16,7 @@ if(NOT AWK_PROGRAM)
 endif()
 
 set(KRB5_SOURCE_DIR "${ClickHouse_SOURCE_DIR}/contrib/krb5/src")
+set(KRB5_ET_BIN_DIR "${CMAKE_CURRENT_BINARY_DIR}/include_private")
 
 set(ALL_SRCS
     "${KRB5_SOURCE_DIR}/util/et/et_name.c"
@@ -90,7 +91,6 @@ set(ALL_SRCS
     "${KRB5_SOURCE_DIR}/lib/gssapi/krb5/get_tkt_flags.c"
     "${KRB5_SOURCE_DIR}/lib/gssapi/krb5/set_allowable_enctypes.c"
     "${KRB5_SOURCE_DIR}/lib/gssapi/krb5/k5sealiov.c"
-    "${KRB5_SOURCE_DIR}/lib/gssapi/krb5/gssapi_err_krb5.c"
     "${KRB5_SOURCE_DIR}/lib/gssapi/krb5/canon_name.c"
     "${KRB5_SOURCE_DIR}/lib/gssapi/krb5/inq_cred.c"
     "${KRB5_SOURCE_DIR}/lib/gssapi/krb5/export_sec_context.c"
@@ -143,11 +143,12 @@ set(ALL_SRCS
     "${KRB5_SOURCE_DIR}/lib/gssapi/generic/util_buffer_set.c"
     "${KRB5_SOURCE_DIR}/lib/gssapi/generic/util_set.c"
     "${KRB5_SOURCE_DIR}/lib/gssapi/generic/util_token.c"
-    "${KRB5_SOURCE_DIR}/lib/gssapi/generic/gssapi_err_generic.c"
     "${KRB5_SOURCE_DIR}/lib/gssapi/generic/disp_major_status.c"
     "${KRB5_SOURCE_DIR}/lib/gssapi/generic/util_seqstate.c"
     "${KRB5_SOURCE_DIR}/lib/gssapi/generic/util_errmap.c"
     "${KRB5_SOURCE_DIR}/lib/gssapi/generic/rel_buffer.c"
+    "${KRB5_ET_BIN_DIR}/lib/gssapi/krb5/gssapi_err_krb5.c"
+    "${KRB5_ET_BIN_DIR}/lib/gssapi/generic/gssapi_err_generic.c"
 
     "${KRB5_SOURCE_DIR}/lib/gssapi/spnego/spnego_mech.c"
     "${KRB5_SOURCE_DIR}/lib/gssapi/spnego/negoex_util.c"
@@ -256,8 +257,8 @@ set(ALL_SRCS
     "${KRB5_SOURCE_DIR}/util/profile/prof_parse.c"
     "${KRB5_SOURCE_DIR}/util/profile/prof_get.c"
     "${KRB5_SOURCE_DIR}/util/profile/prof_set.c"
-    "${KRB5_SOURCE_DIR}/util/profile/prof_err.c"
     "${KRB5_SOURCE_DIR}/util/profile/prof_init.c"
+    "${KRB5_ET_BIN_DIR}/util/profile/prof_err.c"
     "${KRB5_SOURCE_DIR}/lib/krb5/krb/fwd_tgt.c"
     "${KRB5_SOURCE_DIR}/lib/krb5/krb/conv_creds.c"
     "${KRB5_SOURCE_DIR}/lib/krb5/krb/fast.c"
@@ -450,13 +451,12 @@ set(ALL_SRCS
 
 
 
-    "${KRB5_SOURCE_DIR}/lib/krb5/error_tables/k5e1_err.c"
+    "${KRB5_ET_BIN_DIR}/lib/krb5/error_tables/k5e1_err.c"
-    "${KRB5_SOURCE_DIR}/lib/krb5/error_tables/kdb5_err.c"
+    "${KRB5_ET_BIN_DIR}/lib/krb5/error_tables/kdb5_err.c"
-    "${KRB5_SOURCE_DIR}/lib/krb5/error_tables/asn1_err.c"
+    "${KRB5_ET_BIN_DIR}/lib/krb5/error_tables/asn1_err.c"
-    "${KRB5_SOURCE_DIR}/lib/krb5/error_tables/krb5_err.c"
+    "${KRB5_ET_BIN_DIR}/lib/krb5/error_tables/krb5_err.c"
-    "${KRB5_SOURCE_DIR}/lib/krb5/error_tables/krb524_err.c"
+    "${KRB5_ET_BIN_DIR}/lib/krb5/error_tables/krb524_err.c"
-    "${KRB5_SOURCE_DIR}/lib/krb5/error_tables/kv5m_err.c"
+    "${KRB5_ET_BIN_DIR}/lib/krb5/error_tables/kv5m_err.c"
-
 
 
     "${KRB5_SOURCE_DIR}/lib/krb5/rcache/rc_base.c"
@@ -473,7 +473,7 @@ set(ALL_SRCS
 )
 
 add_custom_command(
-    OUTPUT "${KRB5_SOURCE_DIR}/util/et/compile_et"
+    OUTPUT "${CMAKE_CURRENT_BINARY_DIR}/compile_et"
     COMMAND /bin/sh
     ./config_script
     ./compile_et.sh
@@ -481,50 +481,17 @@ add_custom_command(
     ${AWK_PROGRAM}
     sed
     >
-    compile_et
+    ${CMAKE_CURRENT_BINARY_DIR}/compile_et
     DEPENDS "${KRB5_SOURCE_DIR}/util/et/compile_et.sh" "${KRB5_SOURCE_DIR}/util/et/config_script"
     WORKING_DIRECTORY "${KRB5_SOURCE_DIR}/util/et"
 )
 
-file(GLOB_RECURSE ET_FILES
-    "${KRB5_SOURCE_DIR}/*.et"
-)
-
-function(preprocess_et out_var)
-  set(result)
-  foreach(in_f ${ARGN})
-    string(REPLACE
-        .et
-        .c
-        F_C
-        ${in_f}
-    )
-    string(REPLACE
-        .et
-        .h
-        F_H
-        ${in_f}
-    )
-
-    get_filename_component(ET_PATH ${in_f} DIRECTORY)
-
-    add_custom_command(OUTPUT ${F_C} ${F_H}
-      COMMAND perl "${KRB5_SOURCE_DIR}/util/et/compile_et" -d "${KRB5_SOURCE_DIR}/util/et" ${in_f}
-      DEPENDS ${in_f} "${KRB5_SOURCE_DIR}/util/et/compile_et"
-      WORKING_DIRECTORY ${ET_PATH}
-      VERBATIM
-      )
-    list(APPEND result ${F_C})
-  endforeach()
-  set(${out_var} "${result}" PARENT_SCOPE)
-endfunction()
-
 add_custom_command(
-    OUTPUT "${KRB5_SOURCE_DIR}/lib/gssapi/krb5/error_map.h"
+    OUTPUT "${KRB5_ET_BIN_DIR}/error_map.h"
     COMMAND perl
     -I../../../util
     ../../../util/gen-map.pl
-    -oerror_map.h
+    -o${KRB5_ET_BIN_DIR}/error_map.h
     NAME=gsserrmap
     KEY=OM_uint32
     VALUE=char*
@@ -536,22 +503,21 @@ add_custom_command(
 
 add_custom_target(
     ERROR_MAP_H
-    DEPENDS "${KRB5_SOURCE_DIR}/lib/gssapi/krb5/error_map.h"
+    DEPENDS "${KRB5_ET_BIN_DIR}/error_map.h"
     VERBATIM
 )
 
 add_custom_command(
-    OUTPUT "${KRB5_SOURCE_DIR}/lib/gssapi/generic/errmap.h"
+    OUTPUT "${KRB5_ET_BIN_DIR}/errmap.h"
-    COMMAND perl -w -I../../../util  ../../../util/gen.pl bimap errmap.h NAME=mecherrmap LEFT=OM_uint32 RIGHT=struct\ mecherror LEFTPRINT=print_OM_uint32 RIGHTPRINT=mecherror_print LEFTCMP=cmp_OM_uint32 RIGHTCMP=mecherror_cmp
+    COMMAND perl -w -I../../../util  ../../../util/gen.pl bimap ${KRB5_ET_BIN_DIR}/errmap.h NAME=mecherrmap LEFT=OM_uint32 RIGHT=struct\ mecherror LEFTPRINT=print_OM_uint32 RIGHTPRINT=mecherror_print LEFTCMP=cmp_OM_uint32 RIGHTCMP=mecherror_cmp
     WORKING_DIRECTORY "${KRB5_SOURCE_DIR}/lib/gssapi/generic"
 )
 
 add_custom_target(
     ERRMAP_H
-    DEPENDS "${KRB5_SOURCE_DIR}/lib/gssapi/generic/errmap.h"
+    DEPENDS "${KRB5_ET_BIN_DIR}/errmap.h"
     VERBATIM
 )
-
 add_custom_target(
     KRB_5_H
     DEPENDS "${CMAKE_CURRENT_BINARY_DIR}/include/krb5/krb5.h"
@@ -567,7 +533,40 @@ add_dependencies(
     KRB_5_H
 )
 
-preprocess_et(processed_et_files ${ET_FILES})
+#
+# Generate error tables
+#
+function(preprocess_et et_path)
+    string(REPLACE .et .c F_C ${et_path})
+    string(REPLACE .et .h F_H ${et_path})
+    get_filename_component(et_dir ${et_path} DIRECTORY)
+    get_filename_component(et_name ${et_path} NAME_WLE)
+
+    add_custom_command(OUTPUT ${F_C} ${F_H} ${KRB5_ET_BIN_DIR}/${et_name}.h
+        COMMAND perl "${CMAKE_CURRENT_BINARY_DIR}/compile_et" -d "${KRB5_SOURCE_DIR}/util/et" ${et_path}
+        # for #include w/o path (via -iquote)
+        COMMAND ${CMAKE_COMMAND} -E create_symlink ${F_H} ${KRB5_ET_BIN_DIR}/${et_name}.h
+        DEPENDS ${et_path} "${CMAKE_CURRENT_BINARY_DIR}/compile_et"
+        WORKING_DIRECTORY ${et_dir}
+        VERBATIM
+    )
+endfunction()
+
+function(generate_error_tables)
+    file(GLOB_RECURSE ET_FILES "${KRB5_SOURCE_DIR}/*.et")
+    foreach(et_path ${ET_FILES})
+        string(REPLACE ${KRB5_SOURCE_DIR} ${KRB5_ET_BIN_DIR} et_bin_path ${et_path})
+        string(REPLACE / _ et_target_name ${et_path})
+        get_filename_component(et_bin_dir ${et_bin_path} DIRECTORY)
+        add_custom_command(OUTPUT ${et_bin_path}
+            COMMAND ${CMAKE_COMMAND} -E make_directory ${et_bin_dir}
+            COMMAND ${CMAKE_COMMAND} -E copy_if_different ${et_path} ${et_bin_path}
+            VERBATIM
+        )
+        preprocess_et(${et_bin_path})
+    endforeach()
+endfunction()
+generate_error_tables()
 
 if(CMAKE_SYSTEM_NAME MATCHES "Darwin")
     add_custom_command(
@@ -634,12 +633,12 @@ file(MAKE_DIRECTORY
 
 SET(KRBHDEP
     "${KRB5_SOURCE_DIR}/include/krb5/krb5.hin"
-    "${KRB5_SOURCE_DIR}/lib/krb5/error_tables/krb5_err.h"
+    "${KRB5_ET_BIN_DIR}/lib/krb5/error_tables/krb5_err.h"
-    "${KRB5_SOURCE_DIR}/lib/krb5/error_tables/k5e1_err.h"
+    "${KRB5_ET_BIN_DIR}/lib/krb5/error_tables/k5e1_err.h"
-    "${KRB5_SOURCE_DIR}/lib/krb5/error_tables/kdb5_err.h"
+    "${KRB5_ET_BIN_DIR}/lib/krb5/error_tables/kdb5_err.h"
-    "${KRB5_SOURCE_DIR}/lib/krb5/error_tables/kv5m_err.h"
+    "${KRB5_ET_BIN_DIR}/lib/krb5/error_tables/kv5m_err.h"
-    "${KRB5_SOURCE_DIR}/lib/krb5/error_tables/krb524_err.h"
+    "${KRB5_ET_BIN_DIR}/lib/krb5/error_tables/krb524_err.h"
-    "${KRB5_SOURCE_DIR}/lib/krb5/error_tables/asn1_err.h"
+    "${KRB5_ET_BIN_DIR}/lib/krb5/error_tables/asn1_err.h"
 )
 
 # cmake < 3.18 does not have 'cat' command
@@ -656,6 +655,11 @@ target_include_directories(_krb5 SYSTEM BEFORE PUBLIC
     "${CMAKE_CURRENT_BINARY_DIR}/include"
 )
 
+target_compile_options(_krb5 PRIVATE
+    # For '#include "file.h"'
+    -iquote "${CMAKE_CURRENT_BINARY_DIR}/include_private"
+)
+
 target_include_directories(_krb5 PRIVATE
     "${CMAKE_CURRENT_BINARY_DIR}/include_private" # For autoconf.h and other generated headers.
     ${KRB5_SOURCE_DIR}

contrib/llvm-cmake/CMakeLists.txt

@@ -1,12 +1,9 @@
-# During cross-compilation in our CI we have to use llvm-tblgen and other building tools
+if (APPLE OR NOT ARCH_AMD64 OR SANITIZE STREQUAL "undefined")
-# tools to be build for host architecture and everything else for target architecture (e.g. AArch64)
+   set (ENABLE_EMBEDDED_COMPILER_DEFAULT OFF)
-# Possible workaround is to use llvm-tblgen from some package...
-# But lets just enable LLVM for native builds
-if (CMAKE_CROSSCOMPILING OR SANITIZE STREQUAL "undefined")
-    set (ENABLE_EMBEDDED_COMPILER_DEFAULT OFF)
 else()
-    set (ENABLE_EMBEDDED_COMPILER_DEFAULT ON)
+   set (ENABLE_EMBEDDED_COMPILER_DEFAULT ON)
 endif()
+
 option (ENABLE_EMBEDDED_COMPILER "Enable support for 'compile_expressions' option for query execution" ${ENABLE_EMBEDDED_COMPILER_DEFAULT})
 
 if (NOT ENABLE_EMBEDDED_COMPILER)

contrib/poco

@@ -1 +1 @@
-Subproject commit 520a90e02e3e5cb90afeae1846d161dbc508a6f1
+Subproject commit 008b16469471d55b176db181756c94e3f14dd2dc

contrib/unixodbc

@@ -1 +1 @@
-Subproject commit b0ad30f7f6289c12b76f04bfb9d466374bb32168
+Subproject commit a2cd5395e8c7f7390025ec93af5bfebef3fb5fcd

debian/clickhouse-server.service

@@ -20,7 +20,7 @@ ExecStart=/usr/bin/clickhouse-server --config=/etc/clickhouse-server/config.xml
 EnvironmentFile=-/etc/default/clickhouse
 LimitCORE=infinity
 LimitNOFILE=500000
-CapabilityBoundingSet=CAP_NET_ADMIN CAP_IPC_LOCK CAP_SYS_NICE
+CapabilityBoundingSet=CAP_NET_ADMIN CAP_IPC_LOCK CAP_SYS_NICE CAP_NET_BIND_SERVICE
 
 [Install]
 # ClickHouse should not start from the rescue shell (rescue.target).

docker/keeper/Dockerfile

@@ -0,0 +1,74 @@
+FROM ubuntu:20.04 AS glibc-donor
+
+ARG TARGETARCH
+RUN arch=${TARGETARCH:-amd64} \
+    && case $arch in \
+        amd64) rarch=x86_64 ;; \
+        arm64) rarch=aarch64 ;; \
+    esac \
+    && ln -s "${rarch}-linux-gnu" /lib/linux-gnu
+
+
+FROM alpine
+
+ENV LANG=en_US.UTF-8 \
+    LANGUAGE=en_US:en \
+    LC_ALL=en_US.UTF-8 \
+    TZ=UTC \
+    CLICKHOUSE_CONFIG=/etc/clickhouse-server/config.xml
+
+COPY --from=glibc-donor /lib/linux-gnu/libc.so.6 /lib/linux-gnu/libdl.so.2 /lib/linux-gnu/libm.so.6 /lib/linux-gnu/libpthread.so.0 /lib/linux-gnu/librt.so.1 /lib/linux-gnu/libnss_dns.so.2 /lib/linux-gnu/libnss_files.so.2 /lib/linux-gnu/libresolv.so.2 /lib/linux-gnu/ld-2.31.so /lib/
+COPY --from=glibc-donor /etc/nsswitch.conf /etc/
+COPY entrypoint.sh /entrypoint.sh
+
+ARG TARGETARCH
+RUN arch=${TARGETARCH:-amd64} \
+    && case $arch in \
+        amd64) mkdir -p /lib64 && ln -sf /lib/ld-2.31.so /lib64/ld-linux-x86-64.so.2 ;; \
+        arm64) ln -sf /lib/ld-2.31.so /lib/ld-linux-aarch64.so.1 ;; \
+    esac
+
+ARG REPOSITORY="https://s3.amazonaws.com/clickhouse-builds/22.4/31c367d3cd3aefd316778601ff6565119fe36682/package_release"
+ARG VERSION="22.4.1.917"
+ARG PACKAGES="clickhouse-keeper"
+
+# user/group precreated explicitly with fixed uid/gid on purpose.
+# It is especially important for rootless containers: in that case entrypoint
+# can't do chown and owners of mounted volumes should be configured externally.
+# We do that in advance at the begining of Dockerfile before any packages will be
+# installed to prevent picking those uid / gid by some unrelated software.
+# The same uid / gid (101) is used both for alpine and ubuntu.
+
+
+ARG TARGETARCH
+RUN arch=${TARGETARCH:-amd64} \
+    && for package in ${PACKAGES}; do \
+        { \
+            { echo "Get ${REPOSITORY}/${package}-${VERSION}-${arch}.tgz" \
+                && wget -c -q "${REPOSITORY}/${package}-${VERSION}-${arch}.tgz" -O "/tmp/${package}-${VERSION}-${arch}.tgz" \
+                && tar xvzf "/tmp/${package}-${VERSION}-${arch}.tgz" --strip-components=1 -C / ; \
+            } || \
+            { echo "Fallback to ${REPOSITORY}/${package}-${VERSION}.tgz" \
+                && wget -c -q "${REPOSITORY}/${package}-${VERSION}.tgz" -O "/tmp/${package}-${VERSION}.tgz" \
+                && tar xvzf "/tmp/${package}-${VERSION}.tgz" --strip-components=2 -C / ; \
+            } ; \
+        } || exit 1 \
+    ; done \
+    && rm /tmp/*.tgz /install -r \
+    && addgroup -S -g 101 clickhouse \
+    && adduser -S -h /var/lib/clickhouse -s /bin/bash -G clickhouse -g "ClickHouse keeper" -u 101 clickhouse \
+    && mkdir -p /var/lib/clickhouse /var/log/clickhouse-keeper /etc/clickhouse-keeper \
+    && chown clickhouse:clickhouse /var/lib/clickhouse \
+    && chown root:clickhouse /var/log/clickhouse-keeper \
+    && chmod +x /entrypoint.sh \
+    && apk add --no-cache su-exec bash tzdata \
+    && cp /usr/share/zoneinfo/UTC /etc/localtime \
+    && echo "UTC" > /etc/timezone \
+    && chmod ugo+Xrw -R /var/lib/clickhouse /var/log/clickhouse-keeper /etc/clickhouse-keeper
+
+
+EXPOSE 2181 10181 44444
+
+VOLUME /var/lib/clickhouse /var/log/clickhouse-keeper /etc/clickhouse-keeper
+
+ENTRYPOINT ["/entrypoint.sh"]

docker/keeper/Dockerfile.alpine

@@ -0,0 +1 @@
+Dockerfile

docker/keeper/entrypoint.sh

@@ -0,0 +1,93 @@
+#!/bin/bash
+
+set +x
+set -eo pipefail
+shopt -s nullglob
+
+DO_CHOWN=1
+if [ "${CLICKHOUSE_DO_NOT_CHOWN:-0}" = "1" ]; then
+    DO_CHOWN=0
+fi
+
+CLICKHOUSE_UID="${CLICKHOUSE_UID:-"$(id -u clickhouse)"}"
+CLICKHOUSE_GID="${CLICKHOUSE_GID:-"$(id -g clickhouse)"}"
+
+# support --user
+if [ "$(id -u)" = "0" ]; then
+    USER=$CLICKHOUSE_UID
+    GROUP=$CLICKHOUSE_GID
+    if command -v gosu &> /dev/null; then
+        gosu="gosu $USER:$GROUP"
+    elif command -v su-exec &> /dev/null; then
+        gosu="su-exec $USER:$GROUP"
+    else
+        echo "No gosu/su-exec detected!"
+        exit 1
+    fi
+else
+    USER="$(id -u)"
+    GROUP="$(id -g)"
+    gosu=""
+    DO_CHOWN=0
+fi
+
+KEEPER_CONFIG="${KEEPER_CONFIG:-/etc/clickhouse-keeper/config.yaml}"
+
+if [ -f "$KEEPER_CONFIG" ] && ! $gosu test -f "$KEEPER_CONFIG" -a -r "$KEEPER_CONFIG"; then
+    echo "Configuration file '$KEEPER_CONFIG' isn't readable by user with id '$USER'"
+    exit 1
+fi
+
+DATA_DIR="${CLICKHOUSE_DATA_DIR:-/var/lib/clickhouse}"
+LOG_DIR="${LOG_DIR:-/var/log/clickhouse-keeper}"
+LOG_PATH="${LOG_DIR}/clickhouse-keeper.log"
+ERROR_LOG_PATH="${LOG_DIR}/clickhouse-keeper.err.log"
+COORDINATION_LOG_DIR="${DATA_DIR}/coordination/log"
+COORDINATION_SNAPSHOT_DIR="${DATA_DIR}/coordination/snapshots"
+CLICKHOUSE_WATCHDOG_ENABLE=${CLICKHOUSE_WATCHDOG_ENABLE:-0}
+
+for dir in "$DATA_DIR" \
+  "$LOG_DIR" \
+  "$TMP_DIR" \
+  "$COORDINATION_LOG_DIR" \
+  "$COORDINATION_SNAPSHOT_DIR"
+do
+    # check if variable not empty
+    [ -z "$dir" ] && continue
+    # ensure directories exist
+    if ! mkdir -p "$dir"; then
+        echo "Couldn't create necessary directory: $dir"
+        exit 1
+    fi
+
+    if [ "$DO_CHOWN" = "1" ]; then
+        # ensure proper directories permissions
+        # but skip it for if directory already has proper premissions, cause recursive chown may be slow
+        if [ "$(stat -c %u "$dir")" != "$USER" ] || [ "$(stat -c %g "$dir")" != "$GROUP" ]; then
+            chown -R "$USER:$GROUP" "$dir"
+        fi
+    elif ! $gosu test -d "$dir" -a -w "$dir" -a -r "$dir"; then
+        echo "Necessary directory '$dir' isn't accessible by user with id '$USER'"
+        exit 1
+    fi
+done
+
+# if no args passed to `docker run` or first argument start with `--`, then the user is passing clickhouse-server arguments
+if [[ $# -lt 1 ]] || [[ "$1" == "--"* ]]; then
+    # Watchdog is launched by default, but does not send SIGINT to the main process,
+    # so the container can't be finished by ctrl+c
+    export CLICKHOUSE_WATCHDOG_ENABLE
+
+    cd /var/lib/clickhouse
+
+    # There is a config file. It is already tested with gosu (if it is readably by keeper user)
+    if [ -f "$KEEPER_CONFIG" ]; then
+        exec $gosu /usr/bin/clickhouse-keeper --config-file="$KEEPER_CONFIG" --log-file="$LOG_PATH" --errorlog-file="$ERROR_LOG_PATH" "$@"
+    fi
+
+    # There is no config file. Will use embedded one
+    exec $gosu /usr/bin/clickhouse-keeper --log-file="$LOG_PATH" --errorlog-file="$ERROR_LOG_PATH" "$@"
+fi
+
+# Otherwise, we assume the user want to run his own process, for example a `bash` shell to explore this image
+exec "$@"

docker/packager/binary/build.sh

@@ -25,13 +25,21 @@ read -ra CMAKE_FLAGS <<< "${CMAKE_FLAGS:-}"
 env
 cmake --debug-trycompile --verbose=1 -DCMAKE_VERBOSE_MAKEFILE=1 -LA "-DCMAKE_BUILD_TYPE=$BUILD_TYPE" "-DSANITIZE=$SANITIZER" -DENABLE_CHECK_HEAVY_BUILDS=1 "${CMAKE_FLAGS[@]}" ..
 
+if [ "coverity" == "$COMBINED_OUTPUT" ]
+then
+    wget --post-data "token=$COV_TOKEN&project=ClickHouse%2FClickHouse" -qO- https://scan.coverity.com/download/linux64 | tar xz -C /opt/cov-analysis --strip-components 1
+    export PATH=$PATH:/opt/cov-analysis/bin
+    cov-configure --config ./coverity.config --template --comptype clangcc --compiler "$CC"
+    SCAN_WRAPPER="cov-build --config ./coverity.config --dir cov-int"
+fi
+
 cache_status
 # clear cache stats
 ccache --zero-stats ||:
 
 # No quotes because I want it to expand to nothing if empty.
-# shellcheck disable=SC2086
+# shellcheck disable=SC2086 # No quotes because I want it to expand to nothing if empty.
-ninja $NINJA_FLAGS clickhouse-bundle
+$SCAN_WRAPPER ninja $NINJA_FLAGS clickhouse-bundle
 
 cache_status
 
@@ -91,6 +99,12 @@ then
     mv "$COMBINED_OUTPUT.tgz" /output
 fi
 
+if [ "coverity" == "$COMBINED_OUTPUT" ]
+then
+    tar -cv -I pigz -f "coverity-scan.tgz" cov-int
+    mv "coverity-scan.tgz" /output
+fi
+
 # Also build fuzzers if any sanitizer specified
 # if [ -n "$SANITIZER" ]
 # then

docker/packager/packager

@@ -86,6 +86,7 @@ def parse_env_variables(
     additional_pkgs,
     with_coverage,
     with_binaries,
+    coverity_scan,
 ):
     DARWIN_SUFFIX = "-darwin"
     DARWIN_ARM_SUFFIX = "-darwin-aarch64"
@@ -176,6 +177,9 @@ def parse_env_variables(
     if package_type == "performance":
         result.append("COMBINED_OUTPUT=performance")
         cmake_flags.append("-DENABLE_TESTS=0")
+    elif package_type == "coverity":
+        result.append("COMBINED_OUTPUT=coverity")
+        result.append("COV_TOKEN={}".format(cov_token))
     elif split_binary:
         result.append("COMBINED_OUTPUT=shared_build")
 
@@ -262,9 +266,8 @@ if __name__ == "__main__":
     # and configs to be used for performance test.
     parser.add_argument(
         "--package-type",
-        choices=("deb", "binary", "performance"),
+        choices=["deb", "binary", "performance", "coverity"],
         required=True,
-        help="a build type",
     )
     parser.add_argument(
         "--clickhouse-repo-path",
@@ -325,12 +328,13 @@ if __name__ == "__main__":
     parser.add_argument(
         "--docker-image-version", default="latest", help="docker image tag to use"
     )
+    parser.add_argument("--cov_token", default="")
 
     args = parser.parse_args()
     if not os.path.isabs(args.output_dir):
         args.output_dir = os.path.abspath(os.path.join(os.getcwd(), args.output_dir))
 
-    image_type = "binary" if args.package_type == "performance" else args.package_type
+    image_type = "binary" if args.package_type in ("performance", "coverity") else args.package_type
     image_name = "clickhouse/binary-builder"
 
     if not os.path.isabs(args.clickhouse_repo_path):
@@ -372,6 +376,7 @@ if __name__ == "__main__":
         args.additional_pkgs,
         args.with_coverage,
         args.with_binaries,
+        args.cov_token,
     )
 
     run_docker_image_with_env(

docker/server/.gitignore

@@ -1,2 +0,0 @@
-alpine-root/*
-tgz-packages/*

docker/server/Dockerfile

@@ -1,122 +0,0 @@
-FROM ubuntu:20.04
-
-# ARG for quick switch to a given ubuntu mirror
-ARG apt_archive="http://archive.ubuntu.com"
-RUN sed -i "s|http://archive.ubuntu.com|$apt_archive|g" /etc/apt/sources.list
-
-ARG repository="deb https://packages.clickhouse.com/deb stable main"
-ARG version=22.1.1.*
-
-# set non-empty deb_location_url url to create a docker image
-# from debs created by CI build, for example:
-# docker build . --network host --build-arg version="21.4.1.6282" --build-arg deb_location_url="https://clickhouse-builds.s3.yandex.net/21852/069cfbff388b3d478d1a16dc7060b48073f5d522/clickhouse_build_check/clang-11_relwithdebuginfo_none_bundled_unsplitted_disable_False_deb/" -t filimonovq/clickhouse-server:pr21852
-ARG deb_location_url=""
-
-# set non-empty single_binary_location_url to create docker image
-# from a single binary url (useful for non-standard builds - with sanitizers, for arm64).
-# for example (run on aarch64 server):
-# docker build . --network host --build-arg single_binary_location_url="https://builds.clickhouse.com/master/aarch64/clickhouse" -t altinity/clickhouse-server:master-testing-arm
-# note: clickhouse-odbc-bridge is not supported there.
-ARG single_binary_location_url=""
-
-# see https://github.com/moby/moby/issues/4032#issuecomment-192327844
-ARG DEBIAN_FRONTEND=noninteractive
-
-# user/group precreated explicitly with fixed uid/gid on purpose.
-# It is especially important for rootless containers: in that case entrypoint
-# can't do chown and owners of mounted volumes should be configured externally.
-# We do that in advance at the begining of Dockerfile before any packages will be
-# installed to prevent picking those uid / gid by some unrelated software.
-# The same uid / gid (101) is used both for alpine and ubuntu.
-
-# To drop privileges, we need 'su' command, that simply changes uid and gid.
-# In fact, the 'su' command from Linux is not so simple, due to inherent vulnerability in Linux:
-# https://ruderich.org/simon/notes/su-sudo-from-root-tty-hijacking
-# It has to mitigate this drawback of Linux, and to do this, 'su' command is creating it's own pseudo-terminal
-# and forwarding commands. Due to some ridiculous curcumstances, it does not work in Docker (or it does)
-# and for these reasons people are using alternatives to the 'su' command in Docker,
-# that don't mess with the terminal, don't care about closing the opened files, etc...
-# but can only be safe to drop privileges inside Docker.
-# The question - what implementation of 'su' command to use.
-# It should be a simple script doing about just two syscalls.
-# Some people tend to use 'gosu' tool that is written in Go.
-# It is not used for several reasons:
-# 1. Dependency on some foreign code in yet another programming language - does not sound alright.
-# 2. Anselmo D. Adams suggested not to use it due to false positive alarms in some undisclosed security scanners.
-
-COPY su-exec.c /su-exec.c
-
-RUN groupadd -r clickhouse --gid=101 \
-    && useradd -r -g clickhouse --uid=101 --home-dir=/var/lib/clickhouse --shell=/bin/bash clickhouse \
-    && apt-get update \
-    && apt-get install --yes --no-install-recommends \
-        apt-transport-https \
-        ca-certificates \
-        dirmngr \
-        gnupg \
-        locales \
-        wget \
-        tzdata \
-    && mkdir -p /etc/apt/sources.list.d \
-    && apt-key adv --keyserver keyserver.ubuntu.com --recv 8919F6BD2B48D754 \
-    && echo $repository > /etc/apt/sources.list.d/clickhouse.list \
-    && if [ -n "$deb_location_url" ]; then \
-            echo "installing from custom url with deb packages: $deb_location_url" \
-            rm -rf /tmp/clickhouse_debs \
-            && mkdir -p /tmp/clickhouse_debs \
-            && wget --progress=bar:force:noscroll "${deb_location_url}/clickhouse-common-static_${version}_amd64.deb" -P /tmp/clickhouse_debs \
-            && wget --progress=bar:force:noscroll "${deb_location_url}/clickhouse-client_${version}_all.deb" -P /tmp/clickhouse_debs \
-            && wget --progress=bar:force:noscroll "${deb_location_url}/clickhouse-server_${version}_all.deb" -P /tmp/clickhouse_debs \
-            && dpkg -i /tmp/clickhouse_debs/*.deb ; \
-       elif [ -n "$single_binary_location_url" ]; then \
-            echo "installing from single binary url: $single_binary_location_url" \
-            && rm -rf /tmp/clickhouse_binary \
-            && mkdir -p /tmp/clickhouse_binary \
-            && wget --progress=bar:force:noscroll "$single_binary_location_url" -O /tmp/clickhouse_binary/clickhouse \
-            && chmod +x /tmp/clickhouse_binary/clickhouse \
-            && /tmp/clickhouse_binary/clickhouse install --user "clickhouse" --group "clickhouse" ; \
-       else \
-           echo "installing from repository: $repository" \
-           && apt-get update \
-           && apt-get --yes -o "Dpkg::Options::=--force-confdef" -o "Dpkg::Options::=--force-confold" upgrade \
-           && apt-get install --allow-unauthenticated --yes --no-install-recommends \
-                clickhouse-common-static=$version \
-                clickhouse-client=$version \
-                clickhouse-server=$version ; \
-       fi \
-    && apt-get install -y --no-install-recommends tcc libc-dev && \
-        tcc /su-exec.c -o /bin/su-exec && \
-        chown root:root /bin/su-exec && \
-        chmod 0755 /bin/su-exec && \
-        rm /su-exec.c && \
-        apt-get purge -y --auto-remove tcc libc-dev libc-dev-bin libc6-dev linux-libc-dev \
-    && clickhouse-local -q 'SELECT * FROM system.build_options' \
-    && rm -rf \
-        /var/lib/apt/lists/* \
-        /var/cache/debconf \
-        /tmp/* \
-    && apt-get clean \
-    && mkdir -p /var/lib/clickhouse /var/log/clickhouse-server /etc/clickhouse-server /etc/clickhouse-client \
-    && chmod ugo+Xrw -R /var/lib/clickhouse /var/log/clickhouse-server /etc/clickhouse-server /etc/clickhouse-client
-
-# we need to allow "others" access to clickhouse folder, because docker container
-# can be started with arbitrary uid (openshift usecase)
-
-RUN locale-gen en_US.UTF-8
-ENV LANG en_US.UTF-8
-ENV LANGUAGE en_US:en
-ENV LC_ALL en_US.UTF-8
-ENV TZ UTC
-
-RUN mkdir /docker-entrypoint-initdb.d
-
-COPY docker_related_config.xml /etc/clickhouse-server/config.d/
-COPY entrypoint.sh /entrypoint.sh
-RUN chmod +x /entrypoint.sh
-
-EXPOSE 9000 8123 9009
-VOLUME /var/lib/clickhouse
-
-ENV CLICKHOUSE_CONFIG /etc/clickhouse-server/config.xml
-
-ENTRYPOINT ["/entrypoint.sh"]

docker/server/Dockerfile

@@ -0,0 +1 @@
+Dockerfile.ubuntu

docker/server/Dockerfile.alpine

@@ -1,3 +1,14 @@
+FROM ubuntu:20.04 AS glibc-donor
+ARG TARGETARCH
+
+RUN arch=${TARGETARCH:-amd64} \
+    && case $arch in \
+        amd64) rarch=x86_64 ;; \
+        arm64) rarch=aarch64 ;; \
+    esac \
+    && ln -s "${rarch}-linux-gnu" /lib/linux-gnu
+
+
 FROM alpine
 
 ENV LANG=en_US.UTF-8 \
@@ -6,7 +17,24 @@ ENV LANG=en_US.UTF-8 \
     TZ=UTC \
     CLICKHOUSE_CONFIG=/etc/clickhouse-server/config.xml
 
-COPY alpine-root/ /
+COPY --from=glibc-donor /lib/linux-gnu/libc.so.6 /lib/linux-gnu/libdl.so.2 /lib/linux-gnu/libm.so.6 /lib/linux-gnu/libpthread.so.0 /lib/linux-gnu/librt.so.1 /lib/linux-gnu/libnss_dns.so.2 /lib/linux-gnu/libnss_files.so.2 /lib/linux-gnu/libresolv.so.2 /lib/linux-gnu/ld-2.31.so /lib/
+COPY --from=glibc-donor /etc/nsswitch.conf /etc/
+COPY docker_related_config.xml /etc/clickhouse-server/config.d/
+COPY entrypoint.sh /entrypoint.sh
+
+ARG TARGETARCH
+
+RUN arch=${TARGETARCH:-amd64} \
+    && case $arch in \
+        amd64) mkdir -p /lib64 && ln -sf /lib/ld-2.31.so /lib64/ld-linux-x86-64.so.2 ;; \
+        arm64) ln -sf /lib/ld-2.31.so /lib/ld-linux-aarch64.so.1 ;; \
+    esac
+
+# lts / testing / prestable / etc
+ARG REPO_CHANNEL="stable"
+ARG REPOSITORY="https://packages.clickhouse.com/tgz/${REPO_CHANNEL}"
+ARG VERSION="20.9.3.45"
+ARG PACKAGES="clickhouse-client clickhouse-server clickhouse-common-static"
 
 # user/group precreated explicitly with fixed uid/gid on purpose.
 # It is especially important for rootless containers: in that case entrypoint
@@ -15,9 +43,23 @@ COPY alpine-root/ /
 # installed to prevent picking those uid / gid by some unrelated software.
 # The same uid / gid (101) is used both for alpine and ubuntu.
 
-RUN addgroup -S -g 101 clickhouse \
+RUN arch=${TARGETARCH:-amd64} \
+    && for package in ${PACKAGES}; do \
+        { \
+            { echo "Get ${REPOSITORY}/${package}-${VERSION}-${arch}.tgz" \
+                && wget -c -q "${REPOSITORY}/${package}-${VERSION}-${arch}.tgz" -O "/tmp/${package}-${VERSION}-${arch}.tgz" \
+                && tar xvzf "/tmp/${package}-${VERSION}-${arch}.tgz" --strip-components=1 -C / ; \
+            } || \
+            { echo "Fallback to ${REPOSITORY}/${package}-${VERSION}.tgz" \
+                && wget -c -q "${REPOSITORY}/${package}-${VERSION}.tgz" -O "/tmp/${package}-${VERSION}.tgz" \
+                && tar xvzf "/tmp/${package}-${VERSION}.tgz" --strip-components=2 -C / ; \
+            } ; \
+        } || exit 1 \
+    ; done \
+    && rm /tmp/*.tgz /install -r \
+    && addgroup -S -g 101 clickhouse \
     && adduser -S -h /var/lib/clickhouse -s /bin/bash -G clickhouse -g "ClickHouse server" -u 101 clickhouse \
-    && mkdir -p /var/lib/clickhouse /var/log/clickhouse-server /etc/clickhouse-server /etc/clickhouse-client \
+    && mkdir -p /var/lib/clickhouse /var/log/clickhouse-server /etc/clickhouse-server/config.d /etc/clickhouse-server/users.d /etc/clickhouse-client /docker-entrypoint-initdb.d \
     && chown clickhouse:clickhouse /var/lib/clickhouse \
     && chown root:clickhouse /var/log/clickhouse-server \
     && chmod +x /entrypoint.sh \

docker/server/Dockerfile.ubuntu

@@ -0,0 +1,129 @@
+FROM ubuntu:20.04
+
+# see https://github.com/moby/moby/issues/4032#issuecomment-192327844
+ARG DEBIAN_FRONTEND=noninteractive
+
+COPY su-exec.c /su-exec.c
+
+# ARG for quick switch to a given ubuntu mirror
+ARG apt_archive="http://archive.ubuntu.com"
+RUN sed -i "s|http://archive.ubuntu.com|${apt_archive}|g" /etc/apt/sources.list \
+    && groupadd -r clickhouse --gid=101 \
+    && useradd -r -g clickhouse --uid=101 --home-dir=/var/lib/clickhouse --shell=/bin/bash clickhouse \
+    && apt-get update \
+    && apt-get install --yes --no-install-recommends \
+        apt-transport-https \
+        ca-certificates \
+        dirmngr \
+        gnupg \
+        locales \
+        wget \
+        tzdata \
+    && apt-get install -y --no-install-recommends tcc libc-dev && \
+        tcc /su-exec.c -o /bin/su-exec && \
+        chown root:root /bin/su-exec && \
+        chmod 0755 /bin/su-exec && \
+        rm /su-exec.c && \
+        apt-get purge -y --auto-remove tcc libc-dev libc-dev-bin libc6-dev linux-libc-dev \
+    && apt-get clean
+
+ARG REPO_CHANNEL="stable"
+ARG REPOSITORY="deb https://packages.clickhouse.com/deb ${REPO_CHANNEL} main"
+ARG VERSION=22.1.1.*
+ARG PACKAGES="clickhouse-client clickhouse-server clickhouse-common-static"
+
+# set non-empty deb_location_url url to create a docker image
+# from debs created by CI build, for example:
+# docker build . --network host --build-arg version="21.4.1.6282" --build-arg deb_location_url="https://clickhouse-builds.s3.yandex.net/21852/069cfbff388b3d478d1a16dc7060b48073f5d522/clickhouse_build_check/clang-11_relwithdebuginfo_none_bundled_unsplitted_disable_False_deb/" -t filimonovq/clickhouse-server:pr21852
+ARG deb_location_url=""
+
+# set non-empty single_binary_location_url to create docker image
+# from a single binary url (useful for non-standard builds - with sanitizers, for arm64).
+# for example (run on aarch64 server):
+# docker build . --network host --build-arg single_binary_location_url="https://builds.clickhouse.com/master/aarch64/clickhouse" -t altinity/clickhouse-server:master-testing-arm
+# note: clickhouse-odbc-bridge is not supported there.
+ARG single_binary_location_url=""
+
+# user/group precreated explicitly with fixed uid/gid on purpose.
+# It is especially important for rootless containers: in that case entrypoint
+# can't do chown and owners of mounted volumes should be configured externally.
+# We do that in advance at the begining of Dockerfile before any packages will be
+# installed to prevent picking those uid / gid by some unrelated software.
+# The same uid / gid (101) is used both for alpine and ubuntu.
+
+# To drop privileges, we need 'su' command, that simply changes uid and gid.
+# In fact, the 'su' command from Linux is not so simple, due to inherent vulnerability in Linux:
+# https://ruderich.org/simon/notes/su-sudo-from-root-tty-hijacking
+# It has to mitigate this drawback of Linux, and to do this, 'su' command is creating it's own pseudo-terminal
+# and forwarding commands. Due to some ridiculous curcumstances, it does not work in Docker (or it does)
+# and for these reasons people are using alternatives to the 'su' command in Docker,
+# that don't mess with the terminal, don't care about closing the opened files, etc...
+# but can only be safe to drop privileges inside Docker.
+# The question - what implementation of 'su' command to use.
+# It should be a simple script doing about just two syscalls.
+# Some people tend to use 'gosu' tool that is written in Go.
+# It is not used for several reasons:
+# 1. Dependency on some foreign code in yet another programming language - does not sound alright.
+# 2. Anselmo D. Adams suggested not to use it due to false positive alarms in some undisclosed security scanners.
+
+ARG TARGETARCH
+
+RUN arch=${TARGETARCH:-amd64} \
+    && if [ -n "${deb_location_url}" ]; then \
+        echo "installing from custom url with deb packages: ${deb_location_url}" \
+        rm -rf /tmp/clickhouse_debs \
+        && mkdir -p /tmp/clickhouse_debs \
+        && for package in ${PACKAGES}; do \
+            { wget --progress=bar:force:noscroll "${deb_location_url}/${package}_${VERSION}_${arch}.deb" -P /tmp/clickhouse_debs || \
+                wget --progress=bar:force:noscroll "${deb_location_url}/${package}_${VERSION}_all.deb" -P /tmp/clickhouse_debs ; } \
+            || exit 1 \
+        ; done \
+        && dpkg -i /tmp/clickhouse_debs/*.deb ; \
+    elif [ -n "${single_binary_location_url}" ]; then \
+        echo "installing from single binary url: ${single_binary_location_url}" \
+        && rm -rf /tmp/clickhouse_binary \
+        && mkdir -p /tmp/clickhouse_binary \
+        && wget --progress=bar:force:noscroll "${single_binary_location_url}" -O /tmp/clickhouse_binary/clickhouse \
+        && chmod +x /tmp/clickhouse_binary/clickhouse \
+        && /tmp/clickhouse_binary/clickhouse install --user "clickhouse" --group "clickhouse" ; \
+    else \
+        mkdir -p /etc/apt/sources.list.d \
+        && apt-key adv --keyserver keyserver.ubuntu.com --recv 8919F6BD2B48D754 \
+        && echo ${REPOSITORY} > /etc/apt/sources.list.d/clickhouse.list \
+        && echo "installing from repository: ${REPOSITORY}" \
+        && apt-get update \
+        && apt-get --yes -o "Dpkg::Options::=--force-confdef" -o "Dpkg::Options::=--force-confold" upgrade \
+        && for package in ${PACKAGES}; do \
+            packages="${packages} ${package}=${VERSION}" \
+        ; done \
+        && apt-get install --allow-unauthenticated --yes --no-install-recommends ${packages} || exit 1 \
+    ; fi \
+    && clickhouse-local -q 'SELECT * FROM system.build_options' \
+    && rm -rf \
+        /var/lib/apt/lists/* \
+        /var/cache/debconf \
+        /tmp/* \
+    && mkdir -p /var/lib/clickhouse /var/log/clickhouse-server /etc/clickhouse-server /etc/clickhouse-client \
+    && chmod ugo+Xrw -R /var/lib/clickhouse /var/log/clickhouse-server /etc/clickhouse-server /etc/clickhouse-client
+
+# we need to allow "others" access to clickhouse folder, because docker container
+# can be started with arbitrary uid (openshift usecase)
+
+RUN locale-gen en_US.UTF-8
+ENV LANG en_US.UTF-8
+ENV LANGUAGE en_US:en
+ENV LC_ALL en_US.UTF-8
+ENV TZ UTC
+
+RUN mkdir /docker-entrypoint-initdb.d
+
+COPY docker_related_config.xml /etc/clickhouse-server/config.d/
+COPY entrypoint.sh /entrypoint.sh
+RUN chmod +x /entrypoint.sh
+
+EXPOSE 9000 8123 9009
+VOLUME /var/lib/clickhouse
+
+ENV CLICKHOUSE_CONFIG /etc/clickhouse-server/config.xml
+
+ENTRYPOINT ["/entrypoint.sh"]

docker/server/alpine-build.sh

@@ -1,63 +0,0 @@
-#!/bin/bash
-set -x
-
-REPO_CHANNEL="${REPO_CHANNEL:-stable}" # lts / testing / prestable / etc
-REPO_URL="${REPO_URL:-"https://repo.yandex.ru/clickhouse/tgz/${REPO_CHANNEL}"}"
-VERSION="${VERSION:-20.9.3.45}"
-DOCKER_IMAGE="${DOCKER_IMAGE:-clickhouse/clickhouse-server}"
-
-# where original files live
-DOCKER_BUILD_FOLDER="${BASH_SOURCE%/*}"
-
-# we will create root for our image here
-CONTAINER_ROOT_FOLDER="${DOCKER_BUILD_FOLDER}/alpine-root"
-
-# clean up the root from old runs, it's reconstructed each time
-rm -rf "$CONTAINER_ROOT_FOLDER"
-mkdir -p "$CONTAINER_ROOT_FOLDER"
-
-# where to put downloaded tgz
-TGZ_PACKAGES_FOLDER="${DOCKER_BUILD_FOLDER}/tgz-packages"
-mkdir -p "$TGZ_PACKAGES_FOLDER"
-
-PACKAGES=( "clickhouse-client" "clickhouse-server" "clickhouse-common-static" )
-
-# download tars from the repo
-for package in "${PACKAGES[@]}"
-do
-    wget -c -q --show-progress "${REPO_URL}/${package}-${VERSION}.tgz" -O "${TGZ_PACKAGES_FOLDER}/${package}-${VERSION}.tgz"
-done
-
-# unpack tars
-for package in "${PACKAGES[@]}"
-do
-    tar xvzf "${TGZ_PACKAGES_FOLDER}/${package}-${VERSION}.tgz" --strip-components=2 -C "$CONTAINER_ROOT_FOLDER"
-done
-
-# prepare few more folders
-mkdir -p "${CONTAINER_ROOT_FOLDER}/etc/clickhouse-server/users.d" \
-         "${CONTAINER_ROOT_FOLDER}/etc/clickhouse-server/config.d" \
-         "${CONTAINER_ROOT_FOLDER}/var/log/clickhouse-server" \
-         "${CONTAINER_ROOT_FOLDER}/var/lib/clickhouse" \
-         "${CONTAINER_ROOT_FOLDER}/docker-entrypoint-initdb.d" \
-         "${CONTAINER_ROOT_FOLDER}/lib64"
-
-cp "${DOCKER_BUILD_FOLDER}/docker_related_config.xml" "${CONTAINER_ROOT_FOLDER}/etc/clickhouse-server/config.d/"
-cp "${DOCKER_BUILD_FOLDER}/entrypoint.sh"             "${CONTAINER_ROOT_FOLDER}/entrypoint.sh"
-
-## get glibc components from ubuntu 20.04 and put them to expected place
-docker pull ubuntu:20.04
-ubuntu20image=$(docker create --rm ubuntu:20.04)
-docker cp -L "${ubuntu20image}":/lib/x86_64-linux-gnu/libc.so.6       "${CONTAINER_ROOT_FOLDER}/lib"
-docker cp -L "${ubuntu20image}":/lib/x86_64-linux-gnu/libdl.so.2      "${CONTAINER_ROOT_FOLDER}/lib"
-docker cp -L "${ubuntu20image}":/lib/x86_64-linux-gnu/libm.so.6       "${CONTAINER_ROOT_FOLDER}/lib"
-docker cp -L "${ubuntu20image}":/lib/x86_64-linux-gnu/libpthread.so.0 "${CONTAINER_ROOT_FOLDER}/lib"
-docker cp -L "${ubuntu20image}":/lib/x86_64-linux-gnu/librt.so.1      "${CONTAINER_ROOT_FOLDER}/lib"
-docker cp -L "${ubuntu20image}":/lib/x86_64-linux-gnu/libnss_dns.so.2 "${CONTAINER_ROOT_FOLDER}/lib"
-docker cp -L "${ubuntu20image}":/lib/x86_64-linux-gnu/libnss_files.so.2 "${CONTAINER_ROOT_FOLDER}/lib"
-docker cp -L "${ubuntu20image}":/lib/x86_64-linux-gnu/libresolv.so.2  "${CONTAINER_ROOT_FOLDER}/lib"
-docker cp -L "${ubuntu20image}":/lib64/ld-linux-x86-64.so.2           "${CONTAINER_ROOT_FOLDER}/lib64"
-docker cp -L "${ubuntu20image}":/etc/nsswitch.conf                    "${CONTAINER_ROOT_FOLDER}/etc"
-
-docker build "$DOCKER_BUILD_FOLDER" -f Dockerfile.alpine -t "${DOCKER_IMAGE}:${VERSION}-alpine" --pull
-rm -rf "$CONTAINER_ROOT_FOLDER"

docker/server/local.Dockerfile

@@ -1,47 +0,0 @@
-# Since right now we can't set volumes to the docker during build, we split building container in stages:
-# 1. build base container
-# 2. run base conatiner with mounted volumes
-# 3. commit container as image
-# 4. build final container atop that image
-# Middle steps are performed by the bash script.
-
-FROM ubuntu:18.04 as clickhouse-server-base
-ARG gosu_ver=1.14
-
-VOLUME /packages/
-
-# update to allow installing dependencies of clickhouse automatically
-RUN apt update; \
-    DEBIAN_FRONTEND=noninteractive \
-    apt install -y locales;
-
-ADD https://github.com/tianon/gosu/releases/download/${gosu_ver}/gosu-amd64 /bin/gosu
-
-RUN locale-gen en_US.UTF-8
-ENV LANG en_US.UTF-8
-ENV LANGUAGE en_US:en
-ENV LC_ALL en_US.UTF-8
-
-# installing via apt to simulate real-world scenario, where user installs deb package and all it's dependecies automatically.
-CMD DEBIAN_FRONTEND=noninteractive \
-    apt install -y \
-        /packages/clickhouse-common-static_*.deb \
-        /packages/clickhouse-server_*.deb ;
-
-FROM clickhouse-server-base:postinstall as clickhouse-server
-
-RUN mkdir /docker-entrypoint-initdb.d
-
-COPY docker_related_config.xml /etc/clickhouse-server/config.d/
-COPY entrypoint.sh /entrypoint.sh
-
-RUN chmod +x \
-    /entrypoint.sh \
-    /bin/gosu
-
-EXPOSE 9000 8123 9009
-VOLUME /var/lib/clickhouse
-
-ENV CLICKHOUSE_CONFIG /etc/clickhouse-server/config.xml
-
-ENTRYPOINT ["/entrypoint.sh"]

docker/test/fuzzer/run-fuzzer.sh

@@ -226,7 +226,6 @@ quit
         --receive_data_timeout_ms=10000 \
         --stacktrace \
         --query-fuzzer-runs=1000 \
-        --testmode \
         --queries-file $(ls -1 ch/tests/queries/0_stateless/*.sql | sort -R) \
         $NEW_TESTS_OPT \
         > >(tail -n 100000 > fuzzer.log) \

docker/test/integration/mysql_js_client/Dockerfile

@@ -1,8 +1,10 @@
 # docker build -t clickhouse/mysql-js-client .
 # MySQL JavaScript client docker container
 
-FROM node:8
+FROM node:16.14.2
+
+WORKDIR /usr/app
 
 RUN npm install mysql
 
-COPY ./test.js test.js
+COPY ./test.js ./test.js

docker/test/stateless/run.sh

@@ -131,8 +131,23 @@ clickhouse-client -q "system flush logs" ||:
 
 grep -Fa "Fatal" /var/log/clickhouse-server/clickhouse-server.log ||:
 pigz < /var/log/clickhouse-server/clickhouse-server.log > /test_output/clickhouse-server.log.gz &
-clickhouse-client -q "select * from system.query_log format TSVWithNamesAndTypes" | pigz > /test_output/query-log.tsv.gz &
+
-clickhouse-client -q "select * from system.query_thread_log format TSVWithNamesAndTypes" | pigz > /test_output/query-thread-log.tsv.gz &
+# Compress tables.
+#
+# NOTE:
+# - that due to tests with s3 storage we cannot use /var/lib/clickhouse/data
+#   directly
+# - even though ci auto-compress some files (but not *.tsv) it does this only
+#   for files >64MB, we want this files to be compressed explicitly
+for table in query_log zookeeper_log trace_log
+do
+    clickhouse-client -q "select * from system.$table format TSVWithNamesAndTypes" | pigz > /test_output/$table.tsv.gz &
+    if [[ -n "$USE_DATABASE_REPLICATED" ]] && [[ "$USE_DATABASE_REPLICATED" -eq 1 ]]; then
+        clickhouse-client --port 19000 -q "select * from system.$table format TSVWithNamesAndTypes" | pigz > /test_output/$table.1.tsv.gz &
+        clickhouse-client --port 29000 -q "select * from system.$table format TSVWithNamesAndTypes" | pigz > /test_output/$table.2.tsv.gz &
+    fi
+done
+wait ||:
 
 # Also export trace log in flamegraph-friendly format.
 for trace_type in CPU Memory Real
@@ -161,14 +176,6 @@ fi
 
 tar -chf /test_output/coordination.tar /var/lib/clickhouse/coordination ||:
 
-# Replace the engine with Ordinary to avoid extra symlinks stuff in artifacts.
-# (so that clickhouse-local --path can read it w/o extra care).
-sed -i -e "s/ATTACH DATABASE _ UUID '[^']*'/ATTACH DATABASE system/" -e "s/Atomic/Ordinary/" /var/lib/clickhouse/metadata/system.sql
-for table in text_log query_log zookeeper_log trace_log; do
-    sed -i "s/ATTACH TABLE _ UUID '[^']*'/ATTACH TABLE $table/" /var/lib/clickhouse/metadata/system/${table}.sql
-    tar -chf /test_output/${table}_dump.tar /var/lib/clickhouse/metadata/system.sql /var/lib/clickhouse/metadata/system/${table}.sql /var/lib/clickhouse/data/system/${table} ||:
-done
-
 if [[ -n "$USE_DATABASE_REPLICATED" ]] && [[ "$USE_DATABASE_REPLICATED" -eq 1 ]]; then
     grep -Fa "Fatal" /var/log/clickhouse-server/clickhouse-server1.log ||:
     grep -Fa "Fatal" /var/log/clickhouse-server/clickhouse-server2.log ||:
@@ -179,8 +186,6 @@ if [[ -n "$USE_DATABASE_REPLICATED" ]] && [[ "$USE_DATABASE_REPLICATED" -eq 1 ]]
     rm /var/log/clickhouse-server/clickhouse-server2.log
     mv /var/log/clickhouse-server/stderr1.log /test_output/ ||:
     mv /var/log/clickhouse-server/stderr2.log /test_output/ ||:
-    tar -chf /test_output/zookeeper_log_dump1.tar /var/lib/clickhouse1/data/system/zookeeper_log ||:
-    tar -chf /test_output/zookeeper_log_dump2.tar /var/lib/clickhouse2/data/system/zookeeper_log ||:
     tar -chf /test_output/coordination1.tar /var/lib/clickhouse1/coordination ||:
     tar -chf /test_output/coordination2.tar /var/lib/clickhouse2/coordination ||:
 fi

docker/test/stress/run.sh

@@ -348,13 +348,13 @@ then
     rm -f /test_output/tmp
 
     # OOM
-    zgrep -Fa " <Fatal> Application: Child process was terminated by signal 9" /var/log/clickhouse-server/clickhouse-server.log > /dev/null \
+    zgrep -Fa " <Fatal> Application: Child process was terminated by signal 9" /var/log/clickhouse-server/clickhouse-server.log* > /dev/null \
         && echo -e 'Backward compatibility check: OOM killer (or signal 9) in clickhouse-server.log\tFAIL' >> /test_output/test_results.tsv \
         || echo -e 'Backward compatibility check: No OOM messages in clickhouse-server.log\tOK' >> /test_output/test_results.tsv
 
     # Logical errors
     echo "Check for Logical errors in server log:"
-    zgrep -Fa -A20 "Code: 49, e.displayText() = DB::Exception:" /var/log/clickhouse-server/clickhouse-server.log > /test_output/bc_check_logical_errors.txt \
+    zgrep -Fa -A20 "Code: 49, e.displayText() = DB::Exception:" /var/log/clickhouse-server/clickhouse-server.log* > /test_output/bc_check_logical_errors.txt \
         && echo -e 'Backward compatibility check: Logical error thrown (see clickhouse-server.log or bc_check_logical_errors.txt)\tFAIL' >> /test_output/test_results.tsv \
         || echo -e 'Backward compatibility check: No logical errors\tOK' >> /test_output/test_results.tsv
 
@@ -362,13 +362,13 @@ then
     [ -s /test_output/bc_check_logical_errors.txt ] || rm /test_output/bc_check_logical_errors.txt
 
     # Crash
-    zgrep -Fa "########################################" /var/log/clickhouse-server/clickhouse-server.log > /dev/null \
+    zgrep -Fa "########################################" /var/log/clickhouse-server/clickhouse-server.log* > /dev/null \
         && echo -e 'Backward compatibility check: Killed by signal (in clickhouse-server.log)\tFAIL' >> /test_output/test_results.tsv \
         || echo -e 'Backward compatibility check: Not crashed\tOK' >> /test_output/test_results.tsv
 
     # It also checks for crash without stacktrace (printed by watchdog)
     echo "Check for Fatal message in server log:"
-    zgrep -Fa " <Fatal> " /var/log/clickhouse-server/clickhouse-server.log > /test_output/bc_check_fatal_messages.txt \
+    zgrep -Fa " <Fatal> " /var/log/clickhouse-server/clickhouse-server.log* > /test_output/bc_check_fatal_messages.txt \
         && echo -e 'Backward compatibility check: Fatal message in clickhouse-server.log (see bc_check_fatal_messages.txt)\tFAIL' >> /test_output/test_results.tsv \
         || echo -e 'Backward compatibility check: No fatal messages in clickhouse-server.log\tOK' >> /test_output/test_results.tsv
 

docker/test/test_runner.sh

@@ -1,86 +0,0 @@
-#!/bin/sh
-
-set -e -x
-
-# Not sure why shellcheck complains that rc is not assigned before it is referenced.
-# shellcheck disable=SC2154
-trap 'rc=$?; echo EXITED WITH: $rc; exit $rc' EXIT
-
-# CLI option to prevent rebuilding images, just re-run tests with images leftover from previuos time
-readonly NO_REBUILD_FLAG="--no-rebuild"
-
-readonly CLICKHOUSE_DOCKER_DIR="$(realpath "${1}")"
-readonly CLICKHOUSE_PACKAGES_ARG="${2}"
-CLICKHOUSE_SERVER_IMAGE="${3}"
-
-if [ "${CLICKHOUSE_PACKAGES_ARG}" != "${NO_REBUILD_FLAG}" ]; then
-    readonly CLICKHOUSE_PACKAGES_DIR="$(realpath "${2}")" # or --no-rebuild
-fi
-
-
-# In order to allow packages directory to be anywhere, and to reduce amount of context sent to the docker daemon,
-# all images are built in multiple stages:
-# 1. build base image, install dependencies
-# 2. run image with volume mounted, install what needed from those volumes
-# 3. tag container as image
-# 4. [optional] build another image atop of tagged.
-
-# TODO: optionally mount most recent clickhouse-test and queries directory from local machine
-
-if [ "${CLICKHOUSE_PACKAGES_ARG}" != "${NO_REBUILD_FLAG}" ]; then
-    docker build --network=host \
-        -f "${CLICKHOUSE_DOCKER_DIR}/test/stateless/clickhouse-statelest-test-runner.Dockerfile" \
-        --target clickhouse-test-runner-base \
-        -t clickhouse-test-runner-base:preinstall \
-        "${CLICKHOUSE_DOCKER_DIR}/test/stateless"
-
-    docker rm -f clickhouse-test-runner-installing-packages || true
-    docker run  --network=host \
-        -v "${CLICKHOUSE_PACKAGES_DIR}:/packages" \
-        --name clickhouse-test-runner-installing-packages \
-        clickhouse-test-runner-base:preinstall
-    docker commit clickhouse-test-runner-installing-packages clickhouse-statelest-test-runner:local
-    docker rm -f clickhouse-test-runner-installing-packages || true
-fi
-
-# # Create a bind-volume to the clickhouse-test script file
-# docker volume create --driver local --opt type=none --opt device=/home/enmk/proj/ClickHouse_master/tests/clickhouse-test --opt o=bind clickhouse-test-script-volume
-# docker volume create --driver local --opt type=none --opt device=/home/enmk/proj/ClickHouse_master/tests/queries --opt o=bind clickhouse-test-queries-dir-volume
-
-# Build server image (optional) from local packages
-if [ -z "${CLICKHOUSE_SERVER_IMAGE}" ]; then
-    CLICKHOUSE_SERVER_IMAGE="clickhouse/server:local"
-
-    if [ "${CLICKHOUSE_PACKAGES_ARG}" != "${NO_REBUILD_FLAG}" ]; then
-        docker build --network=host \
-            -f "${CLICKHOUSE_DOCKER_DIR}/server/local.Dockerfile" \
-            --target clickhouse-server-base \
-            -t clickhouse-server-base:preinstall \
-            "${CLICKHOUSE_DOCKER_DIR}/server"
-
-        docker rm -f clickhouse_server_base_installing_server || true
-        docker run  --network=host -v "${CLICKHOUSE_PACKAGES_DIR}:/packages" \
-            --name clickhouse_server_base_installing_server \
-            clickhouse-server-base:preinstall
-        docker commit clickhouse_server_base_installing_server clickhouse-server-base:postinstall
-
-        docker build --network=host \
-            -f "${CLICKHOUSE_DOCKER_DIR}/server/local.Dockerfile" \
-            --target clickhouse-server \
-            -t "${CLICKHOUSE_SERVER_IMAGE}" \
-            "${CLICKHOUSE_DOCKER_DIR}/server"
-    fi
-fi
-
-docker rm -f test-runner || true
-docker-compose down
-CLICKHOUSE_SERVER_IMAGE="${CLICKHOUSE_SERVER_IMAGE}" \
-    docker-compose -f "${CLICKHOUSE_DOCKER_DIR}/test/test_runner_docker_compose.yaml" \
-    create \
-    --build --force-recreate
-
-CLICKHOUSE_SERVER_IMAGE="${CLICKHOUSE_SERVER_IMAGE}" \
-    docker-compose -f "${CLICKHOUSE_DOCKER_DIR}/test/test_runner_docker_compose.yaml" \
-    run \
-    --name test-runner \
-    test-runner

docker/test/test_runner_docker_compose.yaml

@@ -1,34 +0,0 @@
-version: "2"
-
-services:
-  clickhouse-server:
-    image: ${CLICKHOUSE_SERVER_IMAGE}
-    expose:
-      - "8123" # HTTP
-      - "9000" # TCP
-      - "9009" # HTTP-interserver
-    restart: "no"
-
-  test-runner:
-    image: clickhouse-statelest-test-runner:local
-
-    restart: "no"
-    depends_on:
-      - clickhouse-server
-    environment:
-      # these are used by clickhouse-test to point clickhouse-client to the right server
-      - CLICKHOUSE_HOST=clickhouse-server
-      - CLICKHOUSE_PORT=9009
-      - CLICKHOUSE_TEST_HOST_EXPOSED_PORT=51234
-    expose:
-      # port for any test to serve data to clickhouse-server on rare occasion (like URL-engine tables in 00646),
-      # should match value of CLICKHOUSE_TEST_HOST_EXPOSED_PORT above
-      - "51234"
-
-    # NOTE: Dev-mode: mount newest versions of the queries and clickhouse-test script into container.
-    # volumes:
-    #   - /home/enmk/proj/ClickHouse_master/tests/queries:/usr/share/clickhouse-test/queries:ro
-    #   - /home/enmk/proj/ClickHouse_master/tests/clickhouse-test:/usr/bin/clickhouse-test:ro
-
-    # String-form instead of list-form to allow multiple arguments in "${CLICKHOUSE_TEST_ARGS}"
-    entrypoint: "clickhouse-test ${CLICKHOUSE_TEST_ARGS}"

docs/en/engines/table-engines/mergetree-family/mergetree.md

@@ -688,7 +688,7 @@ Tags:
 -   `volume_name_N` — Volume name. Volume names must be unique.
 -   `disk` — a disk within a volume.
 -   `max_data_part_size_bytes` — the maximum size of a part that can be stored on any of the volume’s disks. If the a size of a merged part estimated to be bigger than `max_data_part_size_bytes` then this part will be written to a next volume. Basically this feature allows to keep new/small parts on a hot (SSD) volume and move them to a cold (HDD) volume when they reach large size. Do not use this setting if your policy has only one volume.
--   `move_factor` — when the amount of available space gets lower than this factor, data automatically start to move on the next volume if any (by default, 0.1).
+-   `move_factor` — when the amount of available space gets lower than this factor, data automatically starts to move on the next volume if any (by default, 0.1). ClickHouse sorts existing parts by size from largest to smallest (in descending order) and selects parts with the total size that is sufficient to meet the `move_factor` condition. If the total size of all parts is insufficient, all parts will be moved. 
 -   `prefer_not_to_merge` — Disables merging of data parts on this volume. When this setting is enabled, merging data on this volume is not allowed. This allows controlling how ClickHouse works with slow disks.
 
 Cofiguration examples:

docs/en/introduction/adopters.md

@@ -43,7 +43,7 @@ toc_title: Adopters
 | <a href="https://city-mobil.ru" class="favicon">Citymobil</a> | Taxi | Analytics | — | — | [Blog Post in Russian, March 2020](https://habr.com/en/company/citymobil/blog/490660/) |
 | <a href="https://cloudflare.com" class="favicon">Cloudflare</a> | CDN | Traffic analysis | 36 servers | — | [Blog post, May 2017](https://blog.cloudflare.com/how-cloudflare-analyzes-1m-dns-queries-per-second/), [Blog post, March 2018](https://blog.cloudflare.com/http-analytics-for-6m-requests-per-second-using-clickhouse/) |
 | <a href="https://corporate.comcast.com/" class="favicon">Comcast</a> | Media | CDN Traffic Analysis | — | — | [ApacheCon 2019 Talk](https://www.youtube.com/watch?v=e9TZ6gFDjNg) |
-| <a href="https://contentsquare.com" class="favicon">ContentSquare</a> | Web analytics | Main product | — | — | [Blog post in French, November 2018](http://souslecapot.net/2018/11/21/patrick-chatain-vp-engineering-chez-contentsquare-penser-davantage-amelioration-continue-que-revolution-constante/) |
+| <a href="https://contentsquare.com" class="favicon">Contentsquare</a> | Web analytics | Main product | — | — | [Blog post in French, November 2018](http://souslecapot.net/2018/11/21/patrick-chatain-vp-engineering-chez-contentsquare-penser-davantage-amelioration-continue-que-revolution-constante/) |
 | <a href="https://coru.net/" class="favicon">Corunet</a> | Analytics | Main product | — | — | [Slides in English, April 2019](https://github.com/ClickHouse/clickhouse-presentations/blob/master/meetup21/predictive_models.pdf) |
 | <a href="https://www.creditx.com" class="favicon">CraiditX 氪信</a> | Finance AI | Analysis | — | — | [Slides in English, November 2019](https://github.com/ClickHouse/clickhouse-presentations/blob/master/meetup33/udf.pptx) |
 | <a href="https://crazypanda.ru/en/" class="favicon">Crazypanda</a> | Games | | — | — | Live session on ClickHouse meetup |

docs/en/operations/named-collections.md

@@ -36,6 +36,7 @@ Example of configuration:
             <access_key_id>AKIAIOSFODNN7EXAMPLE</access_key_id>
             <secret_access_key> wJalrXUtnFEMI/K7MDENG/bPxRfiCYEXAMPLEKEY</secret_access_key>
             <format>CSV</format>
+            <url>https://s3.us-east-1.amazonaws.com/yourbucket/mydata/</url>
         </s3_mydata>
     </named_collections>
 </clickhouse>
@@ -44,12 +45,12 @@ Example of configuration:
 ### Example of using named connections with the s3 function
 
 ```sql
-INSERT INTO FUNCTION s3(s3_mydata, url = 'https://s3.us-east-1.amazonaws.com/yourbucket/mydata/test_file.tsv.gz',
+INSERT INTO FUNCTION s3(s3_mydata, filename = 'test_file.tsv.gz',
    format = 'TSV', structure = 'number UInt64', compression_method = 'gzip')
 SELECT * FROM numbers(10000);
 
 SELECT count()
-FROM s3(s3_mydata, url = 'https://s3.us-east-1.amazonaws.com/yourbucket/mydata/test_file.tsv.gz')
+FROM s3(s3_mydata, filename = 'test_file.tsv.gz')
 
 ┌─count()─┐
 │   10000 │

docs/en/sql-reference/functions/date-time-functions.md

@@ -393,6 +393,13 @@ This is a generalization of other functions named `toStartOf*`. For example,
 `toStartOfInterval(t, INTERVAL 1 day)` returns the same as `toStartOfDay(t)`,
 `toStartOfInterval(t, INTERVAL 15 minute)` returns the same as `toStartOfFifteenMinutes(t)` etc.
 
+## toLastDayOfMonth {#toLastDayOfMonth}
+
+Rounds up a date or date with time to the last day of the month.
+Returns the date.
+
+Alias: `LAST_DAY`.
+
 ## toTime {#totime}
 
 Converts a date with time to a certain fixed date, while preserving the time.

docs/en/sql-reference/functions/index.md

@@ -77,7 +77,7 @@ A function configuration contains the following settings:
 -   `argument` - argument description with the `type`, and optional `name` of an argument. Each argument is described in a separate setting. Specifying name is necessary if argument names are part of serialization for user defined function format like [Native](../../interfaces/formats.md#native) or [JSONEachRow](../../interfaces/formats.md#jsoneachrow). Default argument name value is `c` + argument_number.
 -   `format` - a [format](../../interfaces/formats.md) in which arguments are passed to the command.
 -   `return_type` - the type of a returned value.
--   `return_name` - name of retuned value. Specifying return name is necessary if return name is part of serialization for user defined function format like [Native](../../interfaces/formats.md#native) or [JSONEachRow](../../interfaces/formats.md#jsoneachrow). Optional. Default value is `result`.
+-   `return_name` - name of returned value. Specifying return name is necessary if return name is part of serialization for user defined function format like [Native](../../interfaces/formats.md#native) or [JSONEachRow](../../interfaces/formats.md#jsoneachrow). Optional. Default value is `result`.
 -   `type` - an executable type. If `type` is set to `executable` then single command is started. If it is set to `executable_pool` then a pool of commands is created.
 -   `max_command_execution_time` - maximum execution time in seconds for processing block of data. This setting is valid for `executable_pool` commands only. Optional. Default value is `10`.
 -   `command_termination_timeout` - time in seconds during which a command should finish after its pipe is closed. After that time `SIGTERM` is sent to the process executing the command. Optional. Default value is `10`.

docs/en/sql-reference/functions/other-functions.md

@@ -2499,3 +2499,41 @@ Result:
 │                      286 │
 └──────────────────────────┘
 ```
+
+## getTypeSerializationStreams {#getTypeSerializationStreams}
+
+return the serialization streams of data type.
+
+**Syntax**
+``` sql
+getTypeSerializationStreams(type_name)
+
+getTypeSerializationStreams(column)
+```
+
+**Arguments**
+- `type_name` - Name of data type to get its serialization paths. [String](../../sql-reference/data-types/string.md#string).
+- `column`       - any column which has a data type
+
+**Returned value**
+- List of serialization streams;
+
+Type: [Array](../../sql-reference/data-types/array.md)([String](../../sql-reference/data-types/string.md)).
+
+
+
+**Example**
+
+Query:
+
+``` sql
+SELECT getTypeSerializationStreams('Array(Array(Int8))')
+```
+
+Result:
+
+``` text
+┌───────────────────────getTypeSerializationStreams('Array(Array(Int8))')─────────────────────────────┐
+│ ['{ArraySizes}','{ArrayElements, ArraySizes}','{ArrayElements, ArrayElements, Regular}']            │
+└─────────────────────────────────────────────────────────────────────────────────────────────────────┘
+```

docs/en/sql-reference/statements/create/table.md

@@ -114,9 +114,9 @@ In addition, this column is not substituted when using an asterisk in a SELECT q
 
 ### EPHEMERAL {#ephemeral}
 
-`EPHEMERAL expr`
+`EPHEMERAL [expr]`
 
-Ephemeral column. Such a column isn't stored in the table and cannot be SELECTed, but can be referenced in the defaults of CREATE statement.
+Ephemeral column. Such a column isn't stored in the table and cannot be SELECTed, but can be referenced in the defaults of CREATE statement. If `expr` is omitted type for column is required.
 INSERT without list of columns will skip such column, so SELECT/INSERT invariant is preserved -  the dump obtained using `SELECT *` can be inserted back into the table using INSERT without specifying the list of columns.
 
 ### ALIAS {#alias}

docs/ja/development/developer-instruction.md

@@ -273,7 +273,7 @@ GitHubのUIでforkリポジトリに移動します。 ブランチで開発し
 
 プル要求は、作業がまだ完了していない場合でも作成できます。 この場合、単語を入れてください “WIP” （進行中の作業）タイトルの先頭に、それは後で変更することができます。 これは、変更の協調的なレビューと議論、および利用可能なすべてのテストの実行に役立ちます。 変更の簡単な説明を提供することが重要です。
 
-Yandexの従業員がタグであなたのPRにラベルを付けるとすぐにテストが開始されます “can be tested”. The results of some first checks (e.g. code style) will come in within several minutes. Build check results will arrive within half an hour. And the main set of tests will report itself within an hour.
+ClickHouseの従業員がタグであなたのPRにラベルを付けるとすぐにテストが開始されます “can be tested”. The results of some first checks (e.g. code style) will come in within several minutes. Build check results will arrive within half an hour. And the main set of tests will report itself within an hour.
 
 システムは、プル要求用にClickHouseバイナリビルドを個別に準備します。 これらのビルドを取得するには “Details” 次のリンク “ClickHouse build check” 小切手のリストのエントリ。 そこには、ビルドへの直接リンクがあります。ClickHouseのdebパッケージは、本番サーバーにも展開できます（恐れがない場合）。
 

docs/ja/introduction/adopters.md

@@ -27,7 +27,7 @@ toc_title: "\u30A2\u30C0\u30D7\u30BF\u30FC"
 | <a href="http://cisco.com/" class="favicon">Cisco</a>                                          | ネットワーク                 | トラフィック分析     | —                                                 | —                                                                        | [ライトニングトーク2019](https://youtu.be/-hI1vDR2oPY?t=5057)                                                                                                                                                     |
 | <a href="https://www.citadelsecurities.com/" class="favicon">Citadel Securities</a>            | 金融                         | —                    | —                                                 | —                                                                        | [2019年の貢献](https://github.com/ClickHouse/ClickHouse/pull/4774)                                                                                                                                                |
 | <a href="https://city-mobil.ru" class="favicon">シティモービル</a>                             | タクシー                     | 分析                 | —                                                 | —                                                                        | [ロシア語でのブログ投稿,月2020](https://habr.com/en/company/citymobil/blog/490660/)                                                                                                                               |
-| <a href="https://contentsquare.com" class="favicon">ContentSquare</a>                          | ウェブ分析                   | 主な製品             | —                                                 | —                                                                        | [フランス語でのブログ投稿,November2018](http://souslecapot.net/2018/11/21/patrick-chatain-vp-engineering-chez-contentsquare-penser-davantage-amelioration-continue-que-revolution-constante/)                     |
+| <a href="https://contentsquare.com" class="favicon">Contentsquare</a>                          | ウェブ分析                   | 主な製品             | —                                                 | —                                                                        | [フランス語でのブログ投稿,November2018](http://souslecapot.net/2018/11/21/patrick-chatain-vp-engineering-chez-contentsquare-penser-davantage-amelioration-continue-que-revolution-constante/)                     |
 | <a href="https://cloudflare.com" class="favicon">Cloudflare</a>                                | CDN                          | トラフィック分析     | 36台のサーバー                                    | —                                                                        | [ブログ投稿,月2017](https://blog.cloudflare.com/how-cloudflare-analyzes-1m-dns-queries-per-second/), [ブログ投稿,月2018](https://blog.cloudflare.com/http-analytics-for-6m-requests-per-second-using-clickhouse/) |
 | <a href="https://coru.net/" class="favicon">コルネット</a>                                     | 分析                         | 主な製品             | —                                                 | —                                                                        | [2019年英語スライド](https://github.com/ClickHouse/clickhouse-presentations/blob/master/meetup21/predictive_models.pdf)                                                                                           |
 | <a href="https://www.creditx.com" class="favicon">CraiditX 氪信</a>                            | ファイナンスAI               | 分析                 | —                                                 | —                                                                        | [2019年のスライド](https://github.com/ClickHouse/clickhouse-presentations/blob/master/meetup33/udf.pptx)                                                                                                          |

docs/ru/development/developer-instruction.md

@@ -72,11 +72,11 @@ ClickHouse не работает и не собирается на 32-битны
 
 Этот вариант не подходит для отправки изменений на сервер. Вы можете временно его использовать, а затем добавить ssh ключи и заменить адрес репозитория с помощью команды `git remote`.
 
-Вы можете также добавить для своего локального репозитория адрес оригинального репозитория Яндекса, чтобы притягивать оттуда обновления:
+Вы можете также добавить для своего локального репозитория адрес оригинального репозитория, чтобы притягивать оттуда обновления:
 
     git remote add upstream git@github.com:ClickHouse/ClickHouse.git
 
-После этого, вы сможете добавлять в свой репозиторий обновления из репозитория Яндекса с помощью команды `git pull upstream master`.
+После этого, вы сможете добавлять в свой репозиторий обновления из репозитория ClickHouse с помощью команды `git pull upstream master`.
 
 ### Работа с сабмодулями Git {#rabota-s-sabmoduliami-git}
 
@@ -288,7 +288,7 @@ sudo ./llvm.sh 12
 
 Pull request можно создать, даже если работа над задачей ещё не завершена. В этом случае, добавьте в его название слово «WIP» (work in progress). Название можно будет изменить позже. Это полезно для совместного просмотра и обсуждения изменений, а также для запуска всех имеющихся тестов. Введите краткое описание изменений - впоследствии, оно будет использовано для релизных changelog.
 
-Тесты будут запущены, как только сотрудники Яндекса поставят для pull request тег «Can be tested». Результаты первых проверок (стиль кода) появятся уже через несколько минут. Результаты сборки появятся примерно через пол часа. Результаты основного набора тестов будут доступны в пределах часа.
+Тесты будут запущены, как только сотрудники ClickHouse поставят для pull request тег «Can be tested». Результаты первых проверок (стиль кода) появятся уже через несколько минут. Результаты сборки появятся примерно через пол часа. Результаты основного набора тестов будут доступны в пределах часа.
 
 Система подготовит сборки ClickHouse специально для вашего pull request. Для их получения, нажмите на ссылку «Details» у проверки «Clickhouse build check». Там вы сможете найти прямые ссылки на собранные .deb пакеты ClickHouse, которые, при желании, вы даже сможете установить на свои продакшен серверы (если не страшно).
 

docs/ru/engines/table-engines/mergetree-family/mergetree.md

@@ -678,7 +678,7 @@ TTL d + INTERVAL 1 MONTH GROUP BY k1, k2 SET x = max(x), y = min(y);
 -   `volume_name_N` — название тома. Названия томов должны быть уникальны.
 -   `disk` — диск, находящийся внутри тома.
 -   `max_data_part_size_bytes` — максимальный размер куска данных, который может находится на любом из дисков этого тома. Если в результате слияния размер куска ожидается больше, чем max_data_part_size_bytes, то этот кусок будет записан в следующий том. В основном эта функция позволяет хранить новые / мелкие куски на горячем (SSD) томе и перемещать их на холодный (HDD) том, когда они достигают большого размера. Не используйте этот параметр, если политика имеет только один том. 
--   `move_factor` — доля доступного свободного места на томе, если места становится меньше, то данные начнут перемещение на следующий том, если он есть (по умолчанию 0.1).
+-   `move_factor` — доля доступного свободного места на томе, если места становится меньше, то данные начнут перемещение на следующий том, если он есть (по умолчанию 0.1). Для перемещения куски сортируются по размеру от большего к меньшему (по убыванию) и выбираются куски, совокупный размер которых достаточен для соблюдения условия `move_factor`, если совокупный размер всех партов недостаточен, будут перемещены все парты.
 -   `prefer_not_to_merge` — Отключает слияние кусков данных, хранящихся на данном томе. Если данная настройка включена, то слияние данных, хранящихся на данном томе, не допускается. Это позволяет контролировать работу ClickHouse с медленными дисками.
 
 Примеры конфигураций:

docs/ru/sql-reference/statements/create/table.md

@@ -110,9 +110,9 @@ SELECT x, toTypeName(x) FROM t1;
 
 ### EPHEMERAL {#ephemeral}
 
-`EPHEMERAL expr`
+`EPHEMERAL [expr]`
 
-Эфемерное выражение. Такой столбец не хранится в таблице и не может быть получен в запросе SELECT, но на него можно ссылаться в выражениях по умолчанию запроса CREATE.
+Эфемерное выражение. Такой столбец не хранится в таблице и не может быть получен в запросе SELECT, но на него можно ссылаться в выражениях по умолчанию запроса CREATE. Если значение по умолчанию `expr` не указано, то тип колонки должен быть специфицирован.
 INSERT без списка столбцов игнорирует этот столбец, таким образом сохраняется инвариант - т.е. дамп, полученный путём `SELECT *`, можно вставить обратно в таблицу INSERT-ом без указания списка столбцов.
 
 ### ALIAS {#alias}

docs/ru/sql-reference/table-functions/postgresql.md

@@ -126,7 +126,7 @@ CREATE TABLE pg_table_schema_with_dots (a UInt32)
 
 **См. также**
 
--   [Движок таблиц PostgreSQL](../../sql-reference/table-functions/postgresql.md)
+-   [Движок таблиц PostgreSQL](../../engines/table-engines/integrations/postgresql.md)
 -   [Использование PostgreSQL как источника данных для внешнего словаря](../../sql-reference/dictionaries/external-dictionaries/external-dicts-dict-sources.md#dicts-external_dicts_dict_sources-postgresql)
 
 [Оригинальная статья](https://clickhouse.com/docs/ru/sql-reference/table-functions/postgresql/) <!--hide-->

docs/tools/requirements.txt

@@ -16,7 +16,7 @@ jsmin==3.0.0
 livereload==2.6.3
 Markdown==3.3.2
 MarkupSafe==2.1.0
-mkdocs==1.1.2
+mkdocs==1.3.0
 mkdocs-htmlproofer-plugin==0.0.3
 mkdocs-macros-plugin==0.4.20
 nltk==3.7

docs/zh/development/developer-instruction.md

@@ -259,7 +259,7 @@ ClickHouse的架构描述可以在此处查看：https://clickhouse.com/docs/en/
 
 即使工作尚未完成，也可以创建拉取请求。在这种情况下，请在标题的开头加上«WIP»（正在进行中），以便后续更改。这对于协同审查和讨论更改以及运行所有可用测试用例很有用。提供有关变更的简短描述很重要，这将在后续用于生成重新发布变更日志。
 
-Yandex成员一旦在您的拉取请求上贴上«可以测试»标签，就会开始测试。一些初始检查项（例如，代码类型）的结果会在几分钟内反馈。构建的检查结果将在半小时内完成。而主要的测试用例集结果将在一小时内报告给您。
+ClickHouse成员一旦在您的拉取请求上贴上«可以测试»标签，就会开始测试。一些初始检查项（例如，代码类型）的结果会在几分钟内反馈。构建的检查结果将在半小时内完成。而主要的测试用例集结果将在一小时内报告给您。
 
 系统将分别为您的拉取请求准备ClickHouse二进制版本。若要检索这些构建信息，请在检查列表中单击« ClickHouse构建检查»旁边的«详细信息»链接。在这里，您会找到指向ClickHouse的.deb软件包的直接链接，此外，甚至可以将其部署在生产服务器上（如果您不担心）。
 

packages/clickhouse-server.service

@@ -20,7 +20,7 @@ ExecStart=/usr/bin/clickhouse-server --config=/etc/clickhouse-server/config.xml
 EnvironmentFile=-/etc/default/clickhouse
 LimitCORE=infinity
 LimitNOFILE=500000
-CapabilityBoundingSet=CAP_NET_ADMIN CAP_IPC_LOCK CAP_SYS_NICE
+CapabilityBoundingSet=CAP_NET_ADMIN CAP_IPC_LOCK CAP_SYS_NICE CAP_NET_BIND_SERVICE
 
 [Install]
 # ClickHouse should not start from the rescue shell (rescue.target).

programs/client/Client.cpp

@@ -163,10 +163,24 @@ void Client::initialize(Poco::Util::Application & self)
 
     configReadClient(config(), home_path);
 
+    /** getenv is thread-safe in Linux glibc and in all sane libc implementations.
+      * But the standard does not guarantee that subsequent calls will not rewrite the value by returned pointer.
+      *
+      * man getenv:
+      *
+      * As typically implemented, getenv() returns a pointer to a string within the environment list.
+      * The caller must take care not to modify this string, since that would change the environment of
+      * the process.
+      *
+      * The implementation of getenv() is not required to be reentrant. The string pointed to by the return value of getenv()
+      * may be statically allocated, and can be modified by a subsequent call to getenv(), putenv(3), setenv(3), or unsetenv(3).
+      */
+
     const char * env_user = getenv("CLICKHOUSE_USER");
-    const char * env_password = getenv("CLICKHOUSE_PASSWORD");
     if (env_user)
         config().setString("user", env_user);
+
+    const char * env_password = getenv("CLICKHOUSE_PASSWORD");
     if (env_password)
         config().setString("password", env_password);
 
@@ -810,7 +824,7 @@ void Client::addOptions(OptionsDescription & options_description)
         ("quota_key", po::value<std::string>(), "A string to differentiate quotas when the user have keyed quotas configured on server")
 
         ("max_client_network_bandwidth", po::value<int>(), "the maximum speed of data exchange over the network for the client in bytes per second.")
-        ("compression", po::value<bool>(), "enable or disable compression")
+        ("compression", po::value<bool>(), "enable or disable compression (enabled by default for remote communication and disabled for localhost communication).")
 
         ("query-fuzzer-runs", po::value<int>()->default_value(0), "After executing every SELECT query, do random mutations in it and run again specified number of times. This is used for testing to discover unexpected corner cases.")
         ("interleave-queries-file", po::value<std::vector<std::string>>()->multitoken(),
@@ -1005,6 +1019,7 @@ void Client::processConfig()
             global_context->setCurrentQueryId(query_id);
     }
     print_stack_trace = config().getBool("stacktrace", false);
+    logging_initialized = true;
 
     if (config().has("multiquery"))
         is_multiquery = true;

programs/local/LocalServer.cpp

@@ -434,6 +434,14 @@ catch (...)
     return getCurrentExceptionCode();
 }
 
+void LocalServer::updateLoggerLevel(const String & logs_level)
+{
+    if (!logging_initialized)
+        return;
+
+    config().setString("logger.level", logs_level);
+    updateLevels(config(), logger());
+}
 
 void LocalServer::processConfig()
 {
@@ -460,30 +468,31 @@ void LocalServer::processConfig()
     auto logging = (config().has("logger.console")
                     || config().has("logger.level")
                     || config().has("log-level")
+                    || config().has("send_logs_level")
                     || config().has("logger.log"));
 
-    auto file_logging = config().has("server_logs_file");
+    auto level = config().getString("log-level", "trace");
-    if (is_interactive && logging && !file_logging)
-        throw Exception("For interactive mode logging is allowed only with --server_logs_file option",
-                        ErrorCodes::BAD_ARGUMENTS);
 
-    if (file_logging)
+    if (config().has("server_logs_file"))
     {
-        auto level = Poco::Logger::parseLevel(config().getString("log-level", "trace"));
+        auto poco_logs_level = Poco::Logger::parseLevel(level);
-        Poco::Logger::root().setLevel(level);
+        Poco::Logger::root().setLevel(poco_logs_level);
         Poco::Logger::root().setChannel(Poco::AutoPtr<Poco::SimpleFileChannel>(new Poco::SimpleFileChannel(server_logs_file)));
+        logging_initialized = true;
     }
-    else if (logging)
+    else if (logging || is_interactive)
     {
-        // force enable logging
         config().setString("logger", "logger");
-        // sensitive data rules are not used here
+        auto log_level_default = is_interactive && !logging ? "none" : level;
+        config().setString("logger.level", config().getString("log-level", config().getString("send_logs_level", log_level_default)));
         buildLoggers(config(), logger(), "clickhouse-local");
+        logging_initialized = true;
     }
     else
     {
         Poco::Logger::root().setLevel("none");
         Poco::Logger::root().setChannel(Poco::AutoPtr<Poco::NullChannel>(new Poco::NullChannel()));
+        logging_initialized = false;
     }
 
     shared_context = Context::createShared();
@@ -713,6 +722,8 @@ void LocalServer::processOptions(const OptionsDescription &, const CommandLineOp
         config().setString("logger.log", options["logger.log"].as<std::string>());
     if (options.count("logger.level"))
         config().setString("logger.level", options["logger.level"].as<std::string>());
+    if (options.count("send_logs_level"))
+        config().setString("send_logs_level", options["send_logs_level"].as<std::string>());
 }
 
 }

programs/local/LocalServer.h

@@ -46,6 +46,8 @@ protected:
 
     void processConfig() override;
 
+    void updateLoggerLevel(const String & logs_level) override;
+
 private:
     /** Composes CREATE subquery based on passed arguments (--structure --file --table and --input-format)
       * This query will be executed first, before queries passed through --query argument

programs/server/Server.cpp

@@ -20,6 +20,7 @@
 #include <base/phdr_cache.h>
 #include <base/ErrorHandlers.h>
 #include <base/getMemoryAmount.h>
+#include <base/getAvailableMemoryAmount.h>
 #include <base/errnoToString.h>
 #include <base/coverage.h>
 #include <base/getFQDNOrHostName.h>
@@ -45,6 +46,7 @@
 #include <Core/ServerUUID.h>
 #include <IO/HTTPCommon.h>
 #include <IO/ReadHelpers.h>
+#include <IO/ReadBufferFromFile.h>
 #include <IO/IOThreadPool.h>
 #include <IO/UseSSL.h>
 #include <Interpreters/AsynchronousMetrics.h>
@@ -80,6 +82,7 @@
 #include <Common/SensitiveDataMasker.h>
 #include <Common/ThreadFuzzer.h>
 #include <Common/getHashOfLoadedBinary.h>
+#include <Common/filesystemHelpers.h>
 #include <Common/Elf.h>
 #include <Server/MySQLHandlerFactory.h>
 #include <Server/PostgreSQLHandlerFactory.h>
@@ -505,6 +508,101 @@ void checkForUsersNotInMainConfig(
     }
 }
 
+/// Unused in other builds
+#if defined(OS_LINUX)
+static String readString(const String & path)
+{
+    ReadBufferFromFile in(path);
+    String contents;
+    readStringUntilEOF(contents, in);
+    return contents;
+}
+
+static int readNumber(const String & path)
+{
+    ReadBufferFromFile in(path);
+    int result;
+    readText(result, in);
+    return result;
+}
+
+#endif
+
+static void sanityChecks(Server * server)
+{
+    std::string data_path = getCanonicalPath(server->config().getString("path", DBMS_DEFAULT_PATH));
+    std::string logs_path = server->config().getString("logger.log", "");
+
+#if defined(OS_LINUX)
+    try
+    {
+        if (readString("/sys/devices/system/clocksource/clocksource0/current_clocksource").find("tsc") == std::string::npos)
+            server->context()->addWarningMessage("Linux is not using fast TSC clock source. Performance can be degraded.");
+    }
+    catch (...)
+    {
+    }
+
+    try
+    {
+        if (readNumber("/proc/sys/vm/overcommit_memory") == 2)
+            server->context()->addWarningMessage("Linux memory overcommit is disabled.");
+    }
+    catch (...)
+    {
+    }
+
+    try
+    {
+        if (readString("/sys/kernel/mm/transparent_hugepage/enabled").find("[always]") != std::string::npos)
+            server->context()->addWarningMessage("Linux transparent hugepage are set to \"always\".");
+    }
+    catch (...)
+    {
+    }
+
+    try
+    {
+        if (readNumber("/proc/sys/kernel/pid_max") < 30000)
+            server->context()->addWarningMessage("Linux max PID is too low.");
+    }
+    catch (...)
+    {
+    }
+
+    try
+    {
+        if (readNumber("/proc/sys/kernel/threads-max") < 30000)
+            server->context()->addWarningMessage("Linux threads max count is too low.");
+    }
+    catch (...)
+    {
+    }
+
+    std::string dev_id = getBlockDeviceId(data_path);
+    if (getBlockDeviceType(dev_id) == BlockDeviceType::ROT && getBlockDeviceReadAheadBytes(dev_id) == 0)
+        server->context()->addWarningMessage("Rotational disk with disabled readahead is in use. Performance can be degraded.");
+#endif
+
+    try
+    {
+        if (getAvailableMemoryAmount() < (2l << 30))
+            server->context()->addWarningMessage("Available memory at server startup is too low (2GiB).");
+
+        if (!enoughSpaceInDirectory(data_path, 1ull << 30))
+            server->context()->addWarningMessage("Available disk space at server startup is too low (1GiB).");
+
+        if (!logs_path.empty())
+        {
+            if (!enoughSpaceInDirectory(fs::path(logs_path).parent_path(), 1ull << 30))
+                server->context()->addWarningMessage("Available disk space at server startup is too low (1GiB).");
+        }
+    }
+    catch (...)
+    {
+    }
+}
+
 int Server::main(const std::vector<std::string> & /*args*/)
 {
     Poco::Logger * log = &logger();
@@ -538,13 +636,14 @@ int Server::main(const std::vector<std::string> & /*args*/)
     global_context->addWarningMessage("Server was built in debug mode. It will work slowly.");
 #endif
 
-if (ThreadFuzzer::instance().isEffective())
+    if (ThreadFuzzer::instance().isEffective())
-    global_context->addWarningMessage("ThreadFuzzer is enabled. Application will run slowly and unstable.");
+        global_context->addWarningMessage("ThreadFuzzer is enabled. Application will run slowly and unstable.");
 
 #if defined(SANITIZER)
     global_context->addWarningMessage("Server was built with sanitizer. It will work slowly.");
 #endif
 
+    sanityChecks(this);
 
     // Initialize global thread pool. Do it before we fetch configs from zookeeper
     // nodes (`from_zk`), because ZooKeeper interface uses the pool. We will
@@ -766,6 +865,38 @@ if (ThreadFuzzer::instance().isEffective())
         }
     }
 
+    /// Try to increase limit on number of threads.
+    {
+        rlimit rlim;
+        if (getrlimit(RLIMIT_NPROC, &rlim))
+            throw Poco::Exception("Cannot getrlimit");
+
+        if (rlim.rlim_cur == rlim.rlim_max)
+        {
+            LOG_DEBUG(log, "rlimit on number of threads is {}", rlim.rlim_cur);
+        }
+        else
+        {
+            rlim_t old = rlim.rlim_cur;
+            rlim.rlim_cur = rlim.rlim_max;
+            int rc = setrlimit(RLIMIT_NPROC, &rlim);
+            if (rc != 0)
+            {
+                LOG_WARNING(log, "Cannot set max number of threads to {}. error: {}", rlim.rlim_cur, strerror(errno));
+                rlim.rlim_cur = old;
+            }
+            else
+            {
+                LOG_DEBUG(log, "Set max number of threads to {} (was {}).", rlim.rlim_cur, old);
+            }
+        }
+
+        if (rlim.rlim_cur < 30000)
+        {
+            global_context->addWarningMessage("Maximum number of threads is lower than 30000. There could be problems with handling a lot of simultaneous queries.");
+        }
+    }
+
     static ServerErrorHandler error_handler;
     Poco::ErrorHandler::set(&error_handler);
 

programs/server/config.xml

@@ -148,13 +148,13 @@
     <!-- <interserver_https_port>9010</interserver_https_port> -->
 
     <!-- Hostname that is used by other replicas to request this server.
-         If not specified, than it is determined analogous to 'hostname -f' command.
+         If not specified, then it is determined analogous to 'hostname -f' command.
          This setting could be used to switch replication to another network interface
          (the server may be connected to multiple networks via multiple addresses)
       -->
 
     <!--
-    <interserver_http_host>example.yandex.ru</interserver_http_host>
+    <interserver_http_host>example.clickhouse.com</interserver_http_host>
     -->
 
     <!-- You can specify credentials for authenthication between replicas.
@@ -765,14 +765,14 @@
     -->
     <!--<remote_url_allow_hosts>-->
         <!-- Host should be specified exactly as in URL. The name is checked before DNS resolution.
-            Example: "yandex.ru", "yandex.ru." and "www.yandex.ru" are different hosts.
+            Example: "clickhouse.com", "clickhouse.com." and "www.clickhouse.com" are different hosts.
                     If port is explicitly specified in URL, the host:port is checked as a whole.
                     If host specified here without port, any port with this host allowed.
-                    "yandex.ru" -> "yandex.ru:443", "yandex.ru:80" etc. is allowed, but "yandex.ru:80" -> only "yandex.ru:80" is allowed.
+                    "clickhouse.com" -> "clickhouse.com:443", "clickhouse.com:80" etc. is allowed, but "clickhouse.com:80" -> only "clickhouse.com:80" is allowed.
             If the host is specified as IP address, it is checked as specified in URL. Example: "[2a02:6b8:a::a]".
             If there are redirects and support for redirects is enabled, every redirect (the Location field) is checked.
             Host should be specified using the host xml tag:
-                    <host>yandex.ru</host>
+                    <host>clickhouse.com</host>
         -->
 
         <!-- Regular expression can be specified. RE2 engine is used for regexps.
@@ -1030,25 +1030,17 @@
         <flush_interval_milliseconds>1000</flush_interval_milliseconds>
     </crash_log>
 
-    <!-- Session log. Stores user log in (successful or not) and log out events. -->
+    <!-- Session log. Stores user log in (successful or not) and log out events.
-    <session_log>
+
+        Note: session log has known security issues and should not be used in production.
+    -->
+    <!-- <session_log>
         <database>system</database>
         <table>session_log</table>
 
         <partition_by>toYYYYMM(event_date)</partition_by>
         <flush_interval_milliseconds>7500</flush_interval_milliseconds>
-    </session_log>
+    </session_log> -->
-
-    <!-- Parameters for embedded dictionaries, used in Yandex.Metrica.
-         See https://clickhouse.com/docs/en/dicts/internal_dicts/
-    -->
-
-    <!-- Path to file with region hierarchy. -->
-    <!-- <path_to_regions_hierarchy_file>/opt/geo/regions_hierarchy.txt</path_to_regions_hierarchy_file> -->
-
-    <!-- Path to directory with files containing names of regions -->
-    <!-- <path_to_regions_names_files>/opt/geo/</path_to_regions_names_files> -->
-
 
     <!-- <top_level_domains_path>/var/lib/clickhouse/top_level_domains/</top_level_domains_path> -->
     <!-- Custom TLD lists.

programs/server/config.yaml.example

@@ -103,7 +103,7 @@ interserver_http_port: 9009
 # If not specified, than it is determined analogous to 'hostname -f' command.
 # This setting could be used to switch replication to another network interface
 # (the server may be connected to multiple networks via multiple addresses)
-# interserver_http_host: example.yandex.ru
+# interserver_http_host: example.clickhouse.com
 
 # You can specify credentials for authenthication between replicas.
 # This is required when interserver_https_port is accessible from untrusted networks,
@@ -592,10 +592,10 @@ remote_servers:
 # remote_url_allow_hosts:
 
 # Host should be specified exactly as in URL. The name is checked before DNS resolution.
-# Example: "yandex.ru", "yandex.ru." and "www.yandex.ru" are different hosts.
+# Example: "clickhouse.com", "clickhouse.com." and "www.clickhouse.com" are different hosts.
 # If port is explicitly specified in URL, the host:port is checked as a whole.
 # If host specified here without port, any port with this host allowed.
-# "yandex.ru" -> "yandex.ru:443", "yandex.ru:80" etc. is allowed, but "yandex.ru:80" -> only "yandex.ru:80" is allowed.
+# "clickhouse.com" -> "clickhouse.com:443", "clickhouse.com:80" etc. is allowed, but "clickhouse.com:80" -> only "clickhouse.com:80" is allowed.
 # If the host is specified as IP address, it is checked as specified in URL. Example: "[2a02:6b8:a::a]".
 # If there are redirects and support for redirects is enabled, every redirect (the Location field) is checked.
 
@@ -803,16 +803,6 @@ crash_log:
     partition_by: ''
     flush_interval_milliseconds: 1000
 
-# Parameters for embedded dictionaries, used in Yandex.Metrica.
-# See https://clickhouse.com/docs/en/dicts/internal_dicts/
-
-# Path to file with region hierarchy.
-# path_to_regions_hierarchy_file: /opt/geo/regions_hierarchy.txt
-
-# Path to directory with files containing names of regions
-# path_to_regions_names_files: /opt/geo/
-
-
 # top_level_domains_path: /var/lib/clickhouse/top_level_domains/
 # Custom TLD lists.
 # Format: name: /path/to/file

programs/server/play.html

@@ -266,12 +266,25 @@
             color: var(--null-color);
         }
 
+        @keyframes hourglass-animation {
+            0% {
+                transform: rotate(-180deg);
+            }
+            50% {
+                transform: rotate(-180deg);
+            }
+            100% {
+                transform: none;
+            }
+        }
+
         #hourglass
         {
             display: none;
-            padding-left: 1rem;
+            margin-left: 1rem;
             font-size: 110%;
             color: #888;
+            animation: hourglass-animation 1s linear infinite;
         }
 
         #check-mark
@@ -457,7 +470,7 @@
         }
 
         document.getElementById('check-mark').style.display = 'none';
-        document.getElementById('hourglass').style.display = 'inline';
+        document.getElementById('hourglass').style.display = 'inline-block';
 
         xhr.send(query);
     }

programs/server/users.xml

@@ -79,9 +79,9 @@
                  Each element of list has one of the following forms:
                  <ip> IP-address or network mask. Examples: 213.180.204.3 or 10.0.0.1/8 or 10.0.0.1/255.255.255.0
                      2a02:6b8::3 or 2a02:6b8::3/64 or 2a02:6b8::3/ffff:ffff:ffff:ffff::.
-                 <host> Hostname. Example: server01.yandex.ru.
+                 <host> Hostname. Example: server01.clickhouse.com.
                      To check access, DNS query is performed, and all received addresses compared to peer address.
-                 <host_regexp> Regular expression for host names. Example, ^server\d\d-\d\d-\d\.yandex\.ru$
+                 <host_regexp> Regular expression for host names. Example, ^server\d\d-\d\d-\d\.clickhouse\.com$
                      To check access, DNS PTR query is performed for peer address and then regexp is applied.
                      Then, for result of PTR query, another DNS query is performed and all received addresses compared to peer address.
                      Strongly recommended that regexp is ends with $

programs/server/users.yaml.example

@@ -70,9 +70,9 @@ users:
         # Each element of list has one of the following forms:
         # ip: IP-address or network mask. Examples: 213.180.204.3 or 10.0.0.1/8 or 10.0.0.1/255.255.255.0
         # 2a02:6b8::3 or 2a02:6b8::3/64 or 2a02:6b8::3/ffff:ffff:ffff:ffff::.
-        # host: Hostname. Example: server01.yandex.ru.
+        # host: Hostname. Example: server01.clickhouse.com.
         # To check access, DNS query is performed, and all received addresses compared to peer address.
-        # host_regexp: Regular expression for host names. Example, ^server\d\d-\d\d-\d\.yandex\.ru$
+        # host_regexp: Regular expression for host names. Example, ^server\d\d-\d\d-\d\.clickhouse\.com$
         # To check access, DNS PTR query is performed for peer address and then regexp is applied.
         # Then, for result of PTR query, another DNS query is performed and all received addresses compared to peer address.
         # Strongly recommended that regexp is ends with $ and take all expression in ''

src/Access/Common/AccessType.h

@@ -182,6 +182,7 @@ enum class AccessType
     M(JDBC, "", GLOBAL, SOURCES) \
     M(HDFS, "", GLOBAL, SOURCES) \
     M(S3, "", GLOBAL, SOURCES) \
+    M(HIVE, "", GLOBAL, SOURCES) \
     M(SOURCES, "", GROUP, ALL) \
     \
     M(ALL, "ALL PRIVILEGES", GROUP, NONE) /* full access */ \

src/Access/Common/QuotaDefs.cpp

@@ -107,6 +107,11 @@ const QuotaTypeInfo & QuotaTypeInfo::get(QuotaType type)
             static const auto info = make_info("EXECUTION_TIME", 1000000000 /* execution_time is stored in nanoseconds */);
             return info;
         }
+        case QuotaType::WRITTEN_BYTES:
+        {
+            static const auto info = make_info("WRITTEN_BYTES", 1);
+            return info;
+        }
         case QuotaType::MAX: break;
     }
     throw Exception("Unexpected quota type: " + std::to_string(static_cast<int>(type)), ErrorCodes::LOGICAL_ERROR);

src/Access/Common/QuotaDefs.h

@@ -20,6 +20,7 @@ enum class QuotaType
     READ_ROWS,      /// Number of rows read from tables.
     READ_BYTES,     /// Number of bytes read from tables.
     EXECUTION_TIME, /// Total amount of query execution time in nanoseconds.
+    WRITTEN_BYTES,  /// Number of bytes written to tables.
 
     MAX
 };

src/Access/EnabledQuota.cpp

@@ -13,7 +13,7 @@ namespace DB
 {
 namespace ErrorCodes
 {
-    extern const int QUOTA_EXPIRED;
+    extern const int QUOTA_EXCEEDED;
 }
 
 
@@ -33,7 +33,7 @@ struct EnabledQuota::Impl
             "Quota for user " + backQuote(user_name) + " for " + to_string(duration) + " has been exceeded: "
                 + type_info.valueToStringWithName(used) + "/" + type_info.valueToString(max) + ". "
                 + "Interval will end at " + to_string(end_of_interval) + ". " + "Name of quota template: " + backQuote(quota_name),
-            ErrorCodes::QUOTA_EXPIRED);
+            ErrorCodes::QUOTA_EXCEEDED);
     }
 
 

src/CMakeLists.txt

@@ -49,6 +49,18 @@ if (COMPILER_GCC)
     add_definitions ("-fno-tree-loop-distribute-patterns")
 endif ()
 
+# ClickHouse developers may use platform-dependent code under some macro (e.g. `#ifdef ENABLE_MULTITARGET`).
+# If turned ON, this option defines such macro.
+# See `src/Common/TargetSpecific.h`
+option(ENABLE_MULTITARGET_CODE "Enable platform-dependent code" ON)
+
+if (ENABLE_MULTITARGET_CODE)
+    add_definitions(-DENABLE_MULTITARGET_CODE=1)
+else()
+    add_definitions(-DENABLE_MULTITARGET_CODE=0)
+endif()
+
+
 add_subdirectory (Access)
 add_subdirectory (Backups)
 add_subdirectory (Columns)

src/Client/ClientBase.cpp

@@ -220,7 +220,7 @@ static void incrementProfileEventsBlock(Block & dst, const Block & src)
 }
 
 
-std::atomic_flag exit_on_signal = ATOMIC_FLAG_INIT;
+std::atomic_flag exit_on_signal;
 
 class QueryInterruptHandler : private boost::noncopyable
 {
@@ -1298,6 +1298,13 @@ void ClientBase::processParsedSingleQuery(const String & full_query, const Strin
         }
     }
 
+    if (const auto * set_query = parsed_query->as<ASTSetQuery>())
+    {
+        const auto * logs_level_field = set_query->changes.tryGet(std::string_view{"send_logs_level"});
+        if (logs_level_field)
+            updateLoggerLevel(logs_level_field->safeGet<String>());
+    }
+
     processed_rows = 0;
     written_first_block = false;
     progress_indication.resetProgress();
@@ -1494,24 +1501,19 @@ MultiQueryProcessingStage ClientBase::analyzeMultiQueryText(
 
 bool ClientBase::executeMultiQuery(const String & all_queries_text)
 {
-    // It makes sense not to base any control flow on this, so that it is
-    // the same in tests and in normal usage. The only difference is that in
-    // normal mode we ignore the test hints.
-    const bool test_mode = config().has("testmode");
-    if (test_mode)
-    {
-        /// disable logs if expects errors
-        TestHint test_hint(test_mode, all_queries_text);
-        if (test_hint.clientError() || test_hint.serverError())
-            processTextAsSingleQuery("SET send_logs_level = 'fatal'");
-    }
-
     bool echo_query = echo_queries;
 
     /// Test tags are started with "--" so they are interpreted as comments anyway.
     /// But if the echo is enabled we have to remove the test tags from `all_queries_text`
     /// because we don't want test tags to be echoed.
-    size_t test_tags_length = test_mode ? getTestTagsLength(all_queries_text) : 0;
+    {
+        /// disable logs if expects errors
+        TestHint test_hint(all_queries_text);
+        if (test_hint.clientError() || test_hint.serverError())
+            processTextAsSingleQuery("SET send_logs_level = 'fatal'");
+    }
+
+    size_t test_tags_length = getTestTagsLength(all_queries_text);
 
     /// Several queries separated by ';'.
     /// INSERT data is ended by the end of line, not ';'.
@@ -1548,7 +1550,7 @@ bool ClientBase::executeMultiQuery(const String & all_queries_text)
                 // Try to find test hint for syntax error. We don't know where
                 // the query ends because we failed to parse it, so we consume
                 // the entire line.
-                TestHint hint(test_mode, String(this_query_begin, this_query_end - this_query_begin));
+                TestHint hint(String(this_query_begin, this_query_end - this_query_begin));
                 if (hint.serverError())
                 {
                     // Syntax errors are considered as client errors
@@ -1586,7 +1588,7 @@ bool ClientBase::executeMultiQuery(const String & all_queries_text)
                 // Look for the hint in the text of query + insert data + trailing
                 // comments, e.g. insert into t format CSV 'a' -- { serverError 123 }.
                 // Use the updated query boundaries we just calculated.
-                TestHint test_hint(test_mode, full_query);
+                TestHint test_hint(full_query);
 
                 // Echo all queries if asked; makes for a more readable reference file.
                 echo_query = test_hint.echoQueries().value_or(echo_query);
@@ -2187,8 +2189,6 @@ void ClientBase::init(int argc, char ** argv)
         ("suggestion_limit", po::value<int>()->default_value(10000),
             "Suggestion limit for how many databases, tables and columns to fetch.")
 
-        ("testmode,T", "enable test hints in comments")
-
         ("format,f", po::value<std::string>(), "default output format")
         ("vertical,E", "vertical output format, same as --format=Vertical or FORMAT Vertical or \\G at end of command")
         ("highlight", po::value<bool>()->default_value(true), "enable or disable basic syntax highlight in interactive command line")
@@ -2294,8 +2294,6 @@ void ClientBase::init(int argc, char ** argv)
         config().setBool("interactive", true);
     if (options.count("pager"))
         config().setString("pager", options["pager"].as<std::string>());
-    if (options.count("testmode"))
-        config().setBool("testmode", true);
 
     if (options.count("log-level"))
         Poco::Logger::root().setLevel(options["log-level"].as<std::string>());

src/Client/ClientBase.h

@@ -95,6 +95,7 @@ protected:
         std::optional<ProgramOptionsDescription> hosts_and_ports_description;
     };
 
+    virtual void updateLoggerLevel(const String &) {}
     virtual void printHelpMessage(const OptionsDescription & options_description) = 0;
     virtual void addOptions(OptionsDescription & options_description) = 0;
     virtual void processOptions(const OptionsDescription & options_description,
@@ -265,6 +266,8 @@ protected:
     bool allow_repeated_settings = false;
 
     bool cancelled = false;
+
+    bool logging_initialized = false;
 };
 
 }

src/Client/TestHint.cpp

@@ -32,12 +32,9 @@ int parseErrorCode(DB::ReadBufferFromString & in)
 namespace DB
 {
 
-TestHint::TestHint(bool enabled_, const String & query_)
+TestHint::TestHint(const String & query_)
     : query(query_)
 {
-    if (!enabled_)
-        return;
-
     // Don't parse error hints in leading comments, because it feels weird.
     // Leading 'echo' hint is OK.
     bool is_leading_hint = true;

src/Client/TestHint.h

@@ -7,7 +7,7 @@
 namespace DB
 {
 
-/// Checks expected server and client error codes in --testmode.
+/// Checks expected server and client error codes.
 ///
 /// The following comment hints are supported:
 ///
@@ -25,12 +25,12 @@ namespace DB
 ///
 /// Examples:
 ///
-/// - echo 'select / -- { clientError 62 }' | clickhouse-client --testmode -nm
+/// - echo 'select / -- { clientError 62 }' | clickhouse-client -nm
 ///
 //    Here the client parses the query but it is incorrect, so it expects
 ///   SYNTAX_ERROR (62).
 ///
-/// - echo 'select foo -- { serverError 47 }' | clickhouse-client --testmode -nm
+/// - echo 'select foo -- { serverError 47 }' | clickhouse-client -nm
 ///
 ///   But here the query is correct, but there is no such column "foo", so it
 ///   is UNKNOWN_IDENTIFIER server error.
@@ -43,7 +43,7 @@ namespace DB
 class TestHint
 {
 public:
-    TestHint(bool enabled_, const String & query_);
+    TestHint(const String & query_);
 
     int serverError() const { return server_error; }
     int clientError() const { return client_error; }

src/Common/ColumnsHashingImpl.h

@@ -125,7 +125,7 @@ class FindResultImpl : public FindResultImplBase, public FindResultImplOffsetBas
 
 public:
     FindResultImpl()
-        : FindResultImplBase(false), FindResultImplOffsetBase<need_offset>(0)
+        : FindResultImplBase(false), FindResultImplOffsetBase<need_offset>(0) // NOLINT(clang-analyzer-optin.cplusplus.UninitializedObject)  intentionally allow uninitialized value here
     {}
 
     FindResultImpl(Mapped * value_, bool found_, size_t off)

src/Common/CompactArray.h

@@ -214,6 +214,9 @@ private:
 
         /// offset in bits to the next to the rightmost bit at that byte; or zero if the rightmost bit is the rightmost bit in that byte.
         offset_r = (l + content_width) % 8;
+
+        content_l = nullptr;
+        content_r = nullptr;
     }
 
     UInt8 ALWAYS_INLINE read(UInt8 value_l) const

src/Common/CurrentMetrics.cpp

@@ -81,6 +81,14 @@
     M(ActiveSyncDrainedConnections, "Number of active connections drained synchronously.") \
     M(AsynchronousReadWait, "Number of threads waiting for asynchronous read.") \
     M(PendingAsyncInsert, "Number of asynchronous inserts that are waiting for flush.") \
+    M(KafkaConsumers, "Number of active Kafka consumers") \
+    M(KafkaConsumersWithAssignment, "Number of active Kafka consumers which have some partitions assigned.") \
+    M(KafkaProducers, "Number of active Kafka producer created") \
+    M(KafkaLibrdkafkaThreads, "Number of active librdkafka threads") \
+    M(KafkaBackgroundReads, "Number of background reads currently working (populating materialized views from Kafka)") \
+    M(KafkaConsumersInUse, "Number of consumers which are currently used by direct or background reads") \
+    M(KafkaWrites, "Number of currently running inserts to Kafka") \
+    M(KafkaAssignedPartitions, "Number of partitions Kafka tables currently assigned to") \
 
 namespace CurrentMetrics
 {

src/Common/DateLUTImpl.h

@@ -360,6 +360,27 @@ public:
             return toDayNum(LUTIndex(i - (lut[i].day_of_month - 1)));
     }
 
+    /// Round up to last day of month.
+    template <typename DateOrTime>
+    inline Time toLastDayOfMonth(DateOrTime v) const
+    {
+        const LUTIndex i = toLUTIndex(v);
+        if constexpr (std::is_unsigned_v<DateOrTime> || std::is_same_v<DateOrTime, DayNum>)
+            return lut_saturated[i - lut[i].day_of_month + lut[i].days_in_month].date;
+        else
+            return lut[i - lut[i].day_of_month + lut[i].days_in_month].date;
+    }
+
+    template <typename DateOrTime>
+    inline auto toLastDayNumOfMonth(DateOrTime v) const
+    {
+        const LUTIndex i = toLUTIndex(v);
+        if constexpr (std::is_unsigned_v<DateOrTime> || std::is_same_v<DateOrTime, DayNum>)
+            return toDayNum(LUTIndexWithSaturation(i - lut[i].day_of_month + lut[i].days_in_month));
+        else
+            return toDayNum(LUTIndex(i - lut[i].day_of_month + lut[i].days_in_month));
+    }
+
     /// Round down to start of quarter.
     template <typename DateOrTime>
     inline auto toFirstDayNumOfQuarter(DateOrTime v) const

src/Common/ErrorCodes.cpp

@@ -208,7 +208,7 @@
     M(198, DNS_ERROR) \
     M(199, UNKNOWN_QUOTA) \
     M(200, QUOTA_DOESNT_ALLOW_KEYS) \
-    M(201, QUOTA_EXPIRED) \
+    M(201, QUOTA_EXCEEDED) \
     M(202, TOO_MANY_SIMULTANEOUS_QUERIES) \
     M(203, NO_FREE_CONNECTION) \
     M(204, CANNOT_FSYNC) \

src/Common/FileCache.cpp

@@ -241,6 +241,10 @@ FileSegmentsHolder LRUFileCache::getOrSet(const Key & key, size_t offset, size_t
 
     std::lock_guard cache_lock(mutex);
 
+#ifndef NDEBUG
+    assertCacheCorrectness(key, cache_lock);
+#endif
+
     /// Get all segments which intersect with the given range.
     auto file_segments = getImpl(key, range, cache_lock);
 
@@ -315,7 +319,7 @@ FileSegmentsHolder LRUFileCache::getOrSet(const Key & key, size_t offset, size_t
 
 LRUFileCache::FileSegmentCell * LRUFileCache::addCell(
     const Key & key, size_t offset, size_t size, FileSegment::State state,
-    std::lock_guard<std::mutex> & /* cache_lock */)
+    std::lock_guard<std::mutex> & cache_lock)
 {
     /// Create a file segment cell and put it in `files` map by [key][offset].
 
@@ -323,8 +327,10 @@ LRUFileCache::FileSegmentCell * LRUFileCache::addCell(
         return nullptr; /// Empty files are not cached.
 
     if (files[key].contains(offset))
-        throw Exception(ErrorCodes::REMOTE_FS_OBJECT_CACHE_ERROR,
+        throw Exception(
-            "Cache already exists for key: `{}`, offset: {}, size: {}", keyToStr(key), offset, size);
+            ErrorCodes::REMOTE_FS_OBJECT_CACHE_ERROR,
+            "Cache already exists for key: `{}`, offset: {}, size: {}.\nCurrent cache structure: {}",
+            keyToStr(key), offset, size, dumpStructureImpl(key, cache_lock));
 
     auto file_segment = std::make_shared<FileSegment>(offset, size, key, this, state);
     FileSegmentCell cell(std::move(file_segment), queue);
@@ -340,8 +346,10 @@ LRUFileCache::FileSegmentCell * LRUFileCache::addCell(
 
     auto [it, inserted] = offsets.insert({offset, std::move(cell)});
     if (!inserted)
-        throw Exception(ErrorCodes::LOGICAL_ERROR,
+        throw Exception(
-            "Failed to insert into cache key: `{}`, offset: {}, size: {}", keyToStr(key), offset, size);
+            ErrorCodes::LOGICAL_ERROR,
+            "Failed to insert into cache key: `{}`, offset: {}, size: {}",
+            keyToStr(key), offset, size);
 
     return &(it->second);
 }
@@ -523,8 +531,8 @@ void LRUFileCache::loadCacheInfoIntoMemory()
     std::lock_guard cache_lock(mutex);
 
     Key key;
-    UInt64 offset;
+    UInt64 offset = 0;
-    size_t size;
+    size_t size = 0;
     std::vector<FileSegmentCell *> cells;
 
     /// cache_base_path / key_prefix / key / offset
@@ -687,22 +695,32 @@ LRUFileCache::FileSegmentCell::FileSegmentCell(FileSegmentPtr file_segment_, LRU
     }
 }
 
-String LRUFileCache::dumpStructure(const Key & key_)
+String LRUFileCache::dumpStructure(const Key & key)
 {
     std::lock_guard cache_lock(mutex);
+    return dumpStructureImpl(key, cache_lock);
+}
 
+String LRUFileCache::dumpStructureImpl(const Key & key, std::lock_guard<std::mutex> & /* cache_lock */)
+{
     WriteBufferFromOwnString result;
-    for (auto it = queue.begin(); it != queue.end(); ++it)
+    const auto & cells_by_offset = files[key];
-    {
+
-        auto [key, offset] = *it;
+    for (const auto & [offset, cell] : cells_by_offset)
-        if (key == key_)
+        result << cell.file_segment->getInfoForLog() << "\n";
-        {
+
-            auto * cell = getCell(key, offset, cache_lock);
-            result << (it != queue.begin() ? ", " : "") << cell->file_segment->range().toString();
-            result << "(state: " << cell->file_segment->download_state << ")";
-        }
-    }
     return result.str();
 }
 
+void LRUFileCache::assertCacheCorrectness(const Key & key, std::lock_guard<std::mutex> & /* cache_lock */)
+{
+    const auto & cells_by_offset = files[key];
+
+    for (const auto & [_, cell] : cells_by_offset)
+    {
+        const auto & file_segment = cell.file_segment;
+        file_segment->assertCorrectness();
+    }
+}
+
 }

src/Common/FileCache.h

@@ -25,6 +25,7 @@ namespace DB
 class IFileCache : private boost::noncopyable
 {
 friend class FileSegment;
+friend struct FileSegmentsHolder;
 
 public:
     using Key = UInt128;
@@ -196,6 +197,8 @@ private:
     FileSegments splitRangeIntoEmptyCells(
         const Key & key, size_t offset, size_t size, std::lock_guard<std::mutex> & cache_lock);
 
+    String dumpStructureImpl(const Key & key_, std::lock_guard<std::mutex> & cache_lock);
+
 public:
     struct Stat
     {
@@ -208,6 +211,7 @@ public:
     Stat getStat();
 
     String dumpStructure(const Key & key_) override;
+    void assertCacheCorrectness(const Key & key, std::lock_guard<std::mutex> & cache_lock);
 };
 
 }

src/Common/FileSegment.cpp

@@ -159,7 +159,18 @@ void FileSegment::setRemoteFileReader(RemoteFileReaderPtr remote_file_reader_)
     remote_file_reader = remote_file_reader_;
 }
 
-void FileSegment::write(const char * from, size_t size)
+void FileSegment::resetRemoteFileReader()
+{
+    if (!isDownloader())
+        throw Exception(ErrorCodes::REMOTE_FS_OBJECT_CACHE_ERROR, "Only downloader can use remote filesystem file reader");
+
+    if (!remote_file_reader)
+        throw Exception(ErrorCodes::LOGICAL_ERROR, "Remote file reader does not exist");
+
+    remote_file_reader.reset();
+}
+
+void FileSegment::write(const char * from, size_t size, size_t offset_)
 {
     if (!size)
         throw Exception(ErrorCodes::REMOTE_FS_OBJECT_CACHE_ERROR, "Writing zero size is not allowed");
@@ -174,8 +185,24 @@ void FileSegment::write(const char * from, size_t size)
                         "Only downloader can do the downloading. (CallerId: {}, DownloaderId: {})",
                         getCallerId(), downloader_id);
 
+    if (downloaded_size == range().size())
+        throw Exception(ErrorCodes::REMOTE_FS_OBJECT_CACHE_ERROR,
+                        "Attempt to write {} bytes to offset: {}, but current file segment is already fully downloaded",
+                        size, offset_);
+
+    auto download_offset = range().left + downloaded_size;
+    if (offset_ != download_offset)
+        throw Exception(ErrorCodes::REMOTE_FS_OBJECT_CACHE_ERROR,
+                        "Attempt to write {} bytes to offset: {}, but current download offset is {}",
+                        size, offset_, download_offset);
+
     if (!cache_writer)
     {
+        if (downloaded_size > 0)
+            throw Exception(ErrorCodes::LOGICAL_ERROR,
+                            "Cache writer was finalized (downloaded size: {}, state: {})",
+                            downloaded_size, stateToString(download_state));
+
         auto download_path = cache->getPathInLocalCache(key(), offset());
         cache_writer = std::make_unique<WriteBufferFromFile>(download_path);
     }
@@ -190,19 +217,26 @@ void FileSegment::write(const char * from, size_t size)
 
         downloaded_size += size;
     }
-    catch (...)
+    catch (Exception & e)
     {
         std::lock_guard segment_lock(mutex);
 
-        LOG_ERROR(log, "Failed to write to cache. File segment info: {}", getInfoForLogImpl(segment_lock));
+        auto info = getInfoForLogImpl(segment_lock);
+        e.addMessage("while writing into cache, info: " + info);
+
+        LOG_ERROR(log, "Failed to write to cache. File segment info: {}", info);
 
         download_state = State::PARTIALLY_DOWNLOADED_NO_CONTINUATION;
 
         cache_writer->finalize();
         cache_writer.reset();
 
+        cv.notify_all();
+
         throw;
     }
+
+    assert(getDownloadOffset() == offset_ + size);
 }
 
 FileSegment::State FileSegment::wait()
@@ -270,7 +304,6 @@ void FileSegment::setDownloaded(std::lock_guard<std::mutex> & /* segment_lock */
     download_state = State::DOWNLOADED;
     is_downloaded = true;
 
-    assert(cache_writer);
     if (cache_writer)
     {
         cache_writer->finalize();
@@ -299,107 +332,125 @@ void FileSegment::completeBatchAndResetDownloader()
 
 void FileSegment::complete(State state)
 {
-    {
-        std::lock_guard segment_lock(mutex);
-
-        bool is_downloader = downloader_id == getCallerId();
-        if (!is_downloader)
-        {
-            cv.notify_all();
-            throw Exception(ErrorCodes::REMOTE_FS_OBJECT_CACHE_ERROR,
-                            "File segment can be completed only by downloader or downloader's FileSegmentsHodler");
-        }
-
-        if (state != State::DOWNLOADED
-            && state != State::PARTIALLY_DOWNLOADED
-            && state != State::PARTIALLY_DOWNLOADED_NO_CONTINUATION)
-        {
-            cv.notify_all();
-            throw Exception(ErrorCodes::REMOTE_FS_OBJECT_CACHE_ERROR,
-                            "Cannot complete file segment with state: {}", stateToString(state));
-        }
-
-        download_state = state;
-    }
-
-    completeImpl();
-    cv.notify_all();
-}
-
-void FileSegment::complete()
-{
-    {
-        std::lock_guard segment_lock(mutex);
-
-        if (download_state == State::SKIP_CACHE || detached)
-            return;
-
-        if (download_state != State::DOWNLOADED && getDownloadedSize(segment_lock) == range().size())
-            setDownloaded(segment_lock);
-
-        if (download_state == State::DOWNLOADING || download_state == State::EMPTY)
-            download_state = State::PARTIALLY_DOWNLOADED;
-    }
-
-    completeImpl(true);
-    cv.notify_all();
-}
-
-void FileSegment::completeImpl(bool allow_non_strict_checking)
-{
-    /// cache lock is always taken before segment lock.
     std::lock_guard cache_lock(cache->mutex);
     std::lock_guard segment_lock(mutex);
 
-    bool download_can_continue = false;
+    bool is_downloader = downloader_id == getCallerId();
-
+    if (!is_downloader)
-    if (download_state == State::PARTIALLY_DOWNLOADED
-                || download_state == State::PARTIALLY_DOWNLOADED_NO_CONTINUATION)
     {
-        bool is_last_holder = cache->isLastFileSegmentHolder(key(), offset(), cache_lock, segment_lock);
+        cv.notify_all();
-        download_can_continue = !is_last_holder && download_state == State::PARTIALLY_DOWNLOADED;
+        throw Exception(ErrorCodes::REMOTE_FS_OBJECT_CACHE_ERROR,
+                        "File segment can be completed only by downloader or downloader's FileSegmentsHodler");
+    }
 
-        if (!download_can_continue)
+    if (state != State::DOWNLOADED
+        && state != State::PARTIALLY_DOWNLOADED
+        && state != State::PARTIALLY_DOWNLOADED_NO_CONTINUATION)
+    {
+        cv.notify_all();
+        throw Exception(ErrorCodes::REMOTE_FS_OBJECT_CACHE_ERROR,
+                        "Cannot complete file segment with state: {}", stateToString(state));
+    }
+
+    download_state = state;
+
+    try
+    {
+        completeImpl(cache_lock, segment_lock);
+    }
+    catch (...)
+    {
+        if (!downloader_id.empty() && downloader_id == getCallerIdImpl(true))
+            downloader_id.clear();
+
+        cv.notify_all();
+        throw;
+    }
+
+    cv.notify_all();
+}
+
+void FileSegment::complete(std::lock_guard<std::mutex> & cache_lock)
+{
+    std::lock_guard segment_lock(mutex);
+
+    if (download_state == State::SKIP_CACHE || detached)
+        return;
+
+    if (download_state != State::DOWNLOADED && getDownloadedSize(segment_lock) == range().size())
+        setDownloaded(segment_lock);
+
+    if (download_state == State::DOWNLOADING || download_state == State::EMPTY)
+    {
+        /// Segment state can be changed from DOWNLOADING or EMPTY only if the caller is the
+        /// downloader or the only owner of the segment.
+
+        bool can_update_segment_state = downloader_id == getCallerIdImpl(true)
+            || cache->isLastFileSegmentHolder(key(), offset(), cache_lock, segment_lock);
+
+        if (can_update_segment_state)
+            download_state = State::PARTIALLY_DOWNLOADED;
+    }
+
+    try
+    {
+        completeImpl(cache_lock, segment_lock, /* allow_non_strict_checking */true);
+    }
+    catch (...)
+    {
+        if (!downloader_id.empty() && downloader_id == getCallerIdImpl(true))
+            downloader_id.clear();
+
+        cv.notify_all();
+        throw;
+    }
+
+    cv.notify_all();
+}
+
+void FileSegment::completeImpl(std::lock_guard<std::mutex> & cache_lock, std::lock_guard<std::mutex> & segment_lock, bool allow_non_strict_checking)
+{
+    bool is_last_holder = cache->isLastFileSegmentHolder(key(), offset(), cache_lock, segment_lock);
+
+    if (is_last_holder
+        && (download_state == State::PARTIALLY_DOWNLOADED || download_state == State::PARTIALLY_DOWNLOADED_NO_CONTINUATION))
+    {
+        size_t current_downloaded_size = getDownloadedSize(segment_lock);
+        if (current_downloaded_size == 0)
         {
-            size_t current_downloaded_size = getDownloadedSize(segment_lock);
+            download_state = State::SKIP_CACHE;
-            if (current_downloaded_size == 0)
+            LOG_TEST(log, "Remove cell {} (nothing downloaded)", range().toString());
-            {
+            cache->remove(key(), offset(), cache_lock, segment_lock);
-                download_state = State::SKIP_CACHE;
+        }
-                LOG_TEST(log, "Remove cell {} (nothing downloaded)", range().toString());
+        else
-                cache->remove(key(), offset(), cache_lock, segment_lock);
+        {
+            /**
+            * Only last holder of current file segment can resize the cell,
+            * because there is an invariant that file segments returned to users
+            * in FileSegmentsHolder represent a contiguous range, so we can resize
+            * it only when nobody needs it.
+            */
+            LOG_TEST(log, "Resize cell {} to downloaded: {}", range().toString(), current_downloaded_size);
+            cache->reduceSizeToDownloaded(key(), offset(), cache_lock, segment_lock);
+        }
 
-                detached = true;
+        detached = true;
-            }
-            else if (is_last_holder)
-            {
-                /**
-                * Only last holder of current file segment can resize the cell,
-                * because there is an invariant that file segments returned to users
-                * in FileSegmentsHolder represent a contiguous range, so we can resize
-                * it only when nobody needs it.
-                */
-                LOG_TEST(log, "Resize cell {} to downloaded: {}", range().toString(), current_downloaded_size);
-                cache->reduceSizeToDownloaded(key(), offset(), cache_lock, segment_lock);
 
-                detached = true;
+        if (cache_writer)
-            }
+        {
+            cache_writer->finalize();
+            cache_writer.reset();
+            remote_file_reader.reset();
         }
     }
 
-    if (!downloader_id.empty() && downloader_id == getCallerIdImpl(allow_non_strict_checking))
+    if (!downloader_id.empty() && (downloader_id == getCallerIdImpl(allow_non_strict_checking) || is_last_holder))
     {
         LOG_TEST(log, "Clearing downloader id: {}, current state: {}", downloader_id, stateToString(download_state));
         downloader_id.clear();
     }
 
-    if (!download_can_continue && cache_writer)
+    assertCorrectnessImpl(segment_lock);
-    {
-        cache_writer->finalize();
-        cache_writer.reset();
-        remote_file_reader.reset();
-    }
-
-    assert(download_state != FileSegment::State::DOWNLOADED || std::filesystem::file_size(cache->getPathInLocalCache(key(), offset())) > 0);
 }
 
 String FileSegment::getInfoForLog() const
@@ -440,6 +491,53 @@ String FileSegment::stateToString(FileSegment::State state)
     __builtin_unreachable();
 }
 
+void FileSegment::assertCorrectness() const
+{
+    std::lock_guard segment_lock(mutex);
+    assertCorrectnessImpl(segment_lock);
+}
+
+void FileSegment::assertCorrectnessImpl(std::lock_guard<std::mutex> & /* segment_lock */) const
+{
+    assert(downloader_id.empty() == (download_state != FileSegment::State::DOWNLOADING));
+    assert(!downloader_id.empty() == (download_state == FileSegment::State::DOWNLOADING));
+    assert(download_state != FileSegment::State::DOWNLOADED || std::filesystem::file_size(cache->getPathInLocalCache(key(), offset())) > 0);
+}
+
+FileSegmentsHolder::~FileSegmentsHolder()
+{
+    /// In CacheableReadBufferFromRemoteFS file segment's downloader removes file segments from
+    /// FileSegmentsHolder right after calling file_segment->complete(), so on destruction here
+    /// remain only uncompleted file segments.
+
+    IFileCache * cache = nullptr;
+
+    for (auto file_segment_it = file_segments.begin(); file_segment_it != file_segments.end();)
+    {
+        auto current_file_segment_it = file_segment_it;
+        auto & file_segment = *current_file_segment_it;
+
+        if (!cache)
+            cache = file_segment->cache;
+
+        try
+        {
+            /// File segment pointer must be reset right after calling complete() and
+            /// under the same mutex, because complete() checks for segment pointers.
+            std::lock_guard cache_lock(cache->mutex);
+
+            file_segment->complete(cache_lock);
+
+            file_segment_it = file_segments.erase(current_file_segment_it);
+        }
+        catch (...)
+        {
+            tryLogCurrentException(__PRETTY_FUNCTION__);
+            assert(false);
+        }
+    }
+}
+
 String FileSegmentsHolder::toString()
 {
     String ranges;

src/Common/FileSegment.h

@@ -95,12 +95,14 @@ public:
 
     bool reserve(size_t size);
 
-    void write(const char * from, size_t size);
+    void write(const char * from, size_t size, size_t offset_);
 
     RemoteFileReaderPtr getRemoteFileReader();
 
     void setRemoteFileReader(RemoteFileReaderPtr remote_file_reader_);
 
+    void resetRemoteFileReader();
+
     String getOrSetDownloader();
 
     String getDownloader() const;
@@ -121,16 +123,32 @@ public:
 
     String getInfoForLog() const;
 
+    void assertCorrectness() const;
+
 private:
     size_t availableSize() const { return reserved_size - downloaded_size; }
-    bool lastFileSegmentHolder() const;
+
-    void complete();
-    void completeImpl(bool allow_non_strict_checking = false);
-    void setDownloaded(std::lock_guard<std::mutex> & segment_lock);
-    static String getCallerIdImpl(bool allow_non_strict_checking = false);
-    void resetDownloaderImpl(std::lock_guard<std::mutex> & segment_lock);
     size_t getDownloadedSize(std::lock_guard<std::mutex> & segment_lock) const;
     String getInfoForLogImpl(std::lock_guard<std::mutex> & segment_lock) const;
+    void assertCorrectnessImpl(std::lock_guard<std::mutex> & segment_lock) const;
+
+    void setDownloaded(std::lock_guard<std::mutex> & segment_lock);
+
+    bool lastFileSegmentHolder() const;
+
+    /// complete() without any completion state is called from destructor of
+    /// FileSegmentsHolder. complete() might check if the caller of the method
+    /// is the last alive holder of the segment. Therefore, complete() and destruction
+    /// of the file segment pointer must be done under the same cache mutex.
+    void complete(std::lock_guard<std::mutex> & cache_lock);
+
+    void completeImpl(
+        std::lock_guard<std::mutex> & cache_lock,
+        std::lock_guard<std::mutex> & segment_lock, bool allow_non_strict_checking = false);
+
+    static String getCallerIdImpl(bool allow_non_strict_checking = false);
+
+    void resetDownloaderImpl(std::lock_guard<std::mutex> & segment_lock);
 
     const Range segment_range;
 
@@ -169,28 +187,7 @@ struct FileSegmentsHolder : private boost::noncopyable
     explicit FileSegmentsHolder(FileSegments && file_segments_) : file_segments(std::move(file_segments_)) {}
     FileSegmentsHolder(FileSegmentsHolder && other) : file_segments(std::move(other.file_segments)) {}
 
-    ~FileSegmentsHolder()
+    ~FileSegmentsHolder();
-    {
-        /// In CacheableReadBufferFromRemoteFS file segment's downloader removes file segments from
-        /// FileSegmentsHolder right after calling file_segment->complete(), so on destruction here
-        /// remain only uncompleted file segments.
-
-        for (auto & segment : file_segments)
-        {
-            try
-            {
-                segment->complete();
-            }
-            catch (...)
-            {
-#ifndef NDEBUG
-                throw;
-#else
-                tryLogCurrentException(__PRETTY_FUNCTION__);
-#endif
-            }
-        }
-    }
 
     FileSegments file_segments{};
 

src/Common/JSONBuilder.h

@@ -61,7 +61,7 @@ private:
 class JSONBool : public IItem
 {
 public:
-    explicit JSONBool(bool value_) : value(std::move(value_)) {}
+    explicit JSONBool(bool value_) : value(value_) {}
     void format(const FormatSettings & settings, FormatContext & context) override;
 
 private:
@@ -74,7 +74,7 @@ public:
     void add(ItemPtr value) { values.push_back(std::move(value)); }
     void add(std::string value) { add(std::make_unique<JSONString>(std::move(value))); }
     void add(const char * value) { add(std::make_unique<JSONString>(value)); }
-    void add(bool value) { add(std::make_unique<JSONBool>(std::move(value))); }
+    void add(bool value) { add(std::make_unique<JSONBool>(value)); }
 
     template <typename T>
     requires std::is_arithmetic_v<T>
@@ -99,7 +99,7 @@ public:
     void add(std::string key, std::string value) { add(std::move(key), std::make_unique<JSONString>(std::move(value))); }
     void add(std::string key, const char * value) { add(std::move(key), std::make_unique<JSONString>(value)); }
     void add(std::string key, std::string_view value) { add(std::move(key), std::make_unique<JSONString>(value)); }
-    void add(std::string key, bool value) { add(std::move(key), std::make_unique<JSONBool>(std::move(value))); }
+    void add(std::string key, bool value) { add(std::move(key), std::make_unique<JSONBool>(value)); }
 
     template <typename T>
     requires std::is_arithmetic_v<T>

src/Common/NamePrompter.cpp

@@ -0,0 +1,15 @@
+#include <IO/WriteHelpers.h>
+#include <Common/NamePrompter.h>
+
+namespace DB::detail
+{
+void appendHintsMessageImpl(String & message, const std::vector<String> & hints)
+{
+    if (hints.empty())
+    {
+        return;
+    }
+
+    message += ". Maybe you meant: " + toString(hints);
+}
+}

src/Common/NamePrompter.h

@@ -90,6 +90,10 @@ private:
     }
 };
 
+namespace detail
+{
+void appendHintsMessageImpl(String & message, const std::vector<String> & hints);
+}
 
 template <size_t MaxNumHints, typename Self>
 class IHints
@@ -102,6 +106,12 @@ public:
         return prompter.getHints(name, getAllRegisteredNames());
     }
 
+    void appendHintsMessage(String & message, const String & name) const
+    {
+        auto hints = getHints(name);
+        detail::appendHintsMessageImpl(message, hints);
+    }
+
     IHints() = default;
 
     IHints(const IHints &) = default;
@@ -114,5 +124,4 @@ public:
 private:
     NamePrompter<MaxNumHints> prompter;
 };
-
 }

src/Common/ProfileEvents.cpp

@@ -112,6 +112,8 @@
     M(CompileExpressionsMicroseconds, "Total time spent for compilation of expressions to LLVM code.") \
     M(CompileExpressionsBytes, "Number of bytes used for expressions compilation.") \
     \
+    M(ExecuteShellCommand, "Number of shell command executions.") \
+    \
     M(ExternalSortWritePart, "") \
     M(ExternalSortMerge, "") \
     M(ExternalAggregationWritePart, "") \
@@ -295,6 +297,25 @@
     M(MergeTreeMetadataCacheHit, "Number of times the read of meta file was done from MergeTree metadata cache") \
     M(MergeTreeMetadataCacheMiss, "Number of times the read of meta file was not done from MergeTree metadata cache") \
     \
+    M(KafkaRebalanceRevocations, "Number of partition revocations (the first stage of consumer group rebalance)") \
+    M(KafkaRebalanceAssignments, "Number of partition assignments (the final stage of consumer group rebalance)") \
+    M(KafkaRebalanceErrors, "Number of failed consumer group rebalances") \
+    M(KafkaMessagesPolled, "Number of Kafka messages polled from librdkafka to ClickHouse") \
+    M(KafkaMessagesRead, "Number of Kafka messages already processed by ClickHouse") \
+    M(KafkaMessagesFailed, "Number of Kafka messages ClickHouse failed to parse") \
+    M(KafkaRowsRead, "Number of rows parsed from Kafka messages") \
+    M(KafkaRowsRejected, "Number of parsed rows which were later rejected (due to rebalances / errors or similar reasons). Those rows will be consumed again after the rebalance.") \
+    M(KafkaDirectReads, "Number of direct selects from Kafka tables since server start") \
+    M(KafkaBackgroundReads, "Number of background reads populating materialized views from Kafka since server start") \
+    M(KafkaCommits, "Number of successful commits of consumed offsets to Kafka (normally should be the same as KafkaBackgroundReads)") \
+    M(KafkaCommitFailures, "Number of failed commits of consumed offsets to Kafka (usually is a sign of some data duplication)") \
+    M(KafkaConsumerErrors, "Number of errors reported by librdkafka during polls") \
+    M(KafkaWrites, "Number of writes (inserts) to Kafka tables ") \
+    M(KafkaRowsWritten, "Number of rows inserted into Kafka tables") \
+    M(KafkaProducerFlushes, "Number of explicit flushes to Kafka producer") \
+    M(KafkaMessagesProduced, "Number of messages produced to Kafka") \
+    M(KafkaProducerErrors, "Number of errors during producing the messages to Kafka") \
+    \
     M(ScalarSubqueriesGlobalCacheHit, "Number of times a read from a scalar subquery was done using the global cache") \
     M(ScalarSubqueriesLocalCacheHit, "Number of times a read from a scalar subquery was done using the local cache") \
     M(ScalarSubqueriesCacheMiss, "Number of times a read from a scalar subquery was not cached and had to be calculated completely")

src/Common/ShellCommand.cpp

@@ -29,6 +29,11 @@ namespace
     };
 }
 
+namespace ProfileEvents
+{
+    extern const Event ExecuteShellCommand;
+}
+
 namespace DB
 {
 
@@ -158,6 +163,7 @@ std::unique_ptr<ShellCommand> ShellCommand::executeImpl(
     const Config & config)
 {
     logCommand(filename, argv);
+    ProfileEvents::increment(ProfileEvents::ExecuteShellCommand);
 
 #if !defined(USE_MUSL)
     /** Here it is written that with a normal call `vfork`, there is a chance of deadlock in multithreaded programs,

src/Common/TargetSpecific.cpp

@@ -1,4 +1,4 @@
-#include <Functions/TargetSpecific.h>
+#include <Common/TargetSpecific.h>
 
 #include <Common/CpuId.h>
 

src/Common/ZooKeeper/ZooKeeper.cpp

@@ -701,24 +701,34 @@ void ZooKeeper::removeChildrenRecursive(const std::string & path, const String &
     }
 }
 
-void ZooKeeper::tryRemoveChildrenRecursive(const std::string & path, const String & keep_child_node)
+bool ZooKeeper::tryRemoveChildrenRecursive(const std::string & path, bool probably_flat, const String & keep_child_node)
 {
     Strings children;
     if (tryGetChildren(path, children) != Coordination::Error::ZOK)
-        return;
+        return false;
+
+    bool removed_as_expected = true;
     while (!children.empty())
     {
         Coordination::Requests ops;
         Strings batch;
+        ops.reserve(MULTI_BATCH_SIZE);
+        batch.reserve(MULTI_BATCH_SIZE);
         for (size_t i = 0; i < MULTI_BATCH_SIZE && !children.empty(); ++i)
         {
             String child_path = fs::path(path) / children.back();
-            tryRemoveChildrenRecursive(child_path);
+
+            /// Will try to avoid recursive getChildren calls if child_path probably has no children.
+            /// It may be extremely slow when path contain a lot of leaf children.
+            if (!probably_flat)
+                tryRemoveChildrenRecursive(child_path);
+
             if (likely(keep_child_node.empty() || keep_child_node != children.back()))
             {
                 batch.push_back(child_path);
                 ops.emplace_back(zkutil::makeRemoveRequest(child_path, -1));
             }
+
             children.pop_back();
         }
 
@@ -726,10 +736,39 @@ void ZooKeeper::tryRemoveChildrenRecursive(const std::string & path, const Strin
         /// this means someone is concurrently removing these children and we will have
         /// to remove them one by one.
         Coordination::Responses responses;
-        if (tryMulti(ops, responses) != Coordination::Error::ZOK)
+        if (tryMulti(ops, responses) == Coordination::Error::ZOK)
-            for (const std::string & child : batch)
+            continue;
-                tryRemove(child);
+
+        removed_as_expected = false;
+
+        std::vector<zkutil::ZooKeeper::FutureRemove> futures;
+        futures.reserve(batch.size());
+        for (const std::string & child : batch)
+            futures.push_back(asyncTryRemoveNoThrow(child, -1));
+
+        for (size_t i = 0; i < batch.size(); ++i)
+        {
+            auto res = futures[i].get();
+            if (res.error == Coordination::Error::ZOK)
+                continue;
+            if (res.error == Coordination::Error::ZNONODE)
+                continue;
+
+            if (res.error == Coordination::Error::ZNOTEMPTY)
+            {
+                if (probably_flat)
+                {
+                    /// It actually has children, let's remove them
+                    tryRemoveChildrenRecursive(batch[i]);
+                    tryRemove(batch[i]);
+                }
+                continue;
+            }
+
+            throw KeeperException(res.error, batch[i]);
+        }
     }
+    return removed_as_expected;
 }
 
 void ZooKeeper::removeRecursive(const std::string & path)

src/Common/ZooKeeper/ZooKeeper.h

@@ -225,7 +225,10 @@ public:
     /// If keep_child_node is not empty, this method will not remove path/keep_child_node (but will remove its subtree).
     /// It can be useful to keep some child node as a flag which indicates that path is currently removing.
     void removeChildrenRecursive(const std::string & path, const String & keep_child_node = {});
-    void tryRemoveChildrenRecursive(const std::string & path, const String & keep_child_node = {});
+    /// If probably_flat is true, this method will optimistically try to remove children non-recursive
+    /// and will fall back to recursive removal if it gets ZNOTEMPTY for some child.
+    /// Returns true if no kind of fallback happened.
+    bool tryRemoveChildrenRecursive(const std::string & path, bool probably_flat = false, const String & keep_child_node = {});
 
     /// Remove all children nodes (non recursive).
     void removeChildren(const std::string & path);

src/Common/filesystemHelpers.cpp

@@ -4,6 +4,8 @@
 #if defined(__linux__)
 #    include <cstdio>
 #    include <mntent.h>
+#    include <sys/stat.h>
+#    include <sys/sysmacros.h>
 #endif
 #include <cerrno>
 #include <Poco/Version.h>
@@ -13,6 +15,9 @@
 #include <unistd.h>
 #include <sys/types.h>
 #include <utime.h>
+#include <IO/ReadBufferFromFile.h>
+#include <IO/Operators.h>
+#include <IO/WriteBufferFromString.h>
 
 namespace fs = std::filesystem;
 
@@ -24,6 +29,7 @@ namespace ErrorCodes
     extern const int LOGICAL_ERROR;
     extern const int SYSTEM_ERROR;
     extern const int NOT_IMPLEMENTED;
+    extern const int CANNOT_STAT;
     extern const int CANNOT_STATVFS;
     extern const int PATH_ACCESS_DENIED;
     extern const int CANNOT_CREATE_FILE;
@@ -57,6 +63,68 @@ std::unique_ptr<TemporaryFile> createTemporaryFile(const std::string & path)
     return std::make_unique<TemporaryFile>(path);
 }
 
+#if !defined(__linux__)
+[[noreturn]]
+#endif
+String getBlockDeviceId([[maybe_unused]] const String & path)
+{
+#if defined(__linux__)
+    struct stat sb;
+    if (lstat(path.c_str(), &sb))
+        throwFromErrnoWithPath("Cannot lstat " + path, path, ErrorCodes::CANNOT_STAT);
+    WriteBufferFromOwnString ss;
+    ss << major(sb.st_dev) << ":" << minor(sb.st_dev);
+    return ss.str();
+#else
+    throw DB::Exception("The function getDeviceId is supported on Linux only", ErrorCodes::NOT_IMPLEMENTED);
+#endif
+}
+
+#if !defined(__linux__)
+[[noreturn]]
+#endif
+BlockDeviceType getBlockDeviceType([[maybe_unused]] const String & device_id)
+{
+#if defined(__linux__)
+    try
+    {
+        ReadBufferFromFile in("/sys/dev/block/" + device_id + "/queue/rotational");
+        int rotational;
+        readText(rotational, in);
+        return rotational ? BlockDeviceType::ROT : BlockDeviceType::NONROT;
+    }
+    catch (...)
+    {
+        return BlockDeviceType::UNKNOWN;
+    }
+#else
+    throw DB::Exception("The function getDeviceType is supported on Linux only", ErrorCodes::NOT_IMPLEMENTED);
+#endif
+}
+
+#if !defined(__linux__)
+[[noreturn]]
+#endif
+UInt64 getBlockDeviceReadAheadBytes([[maybe_unused]] const String & device_id)
+{
+#if defined(__linux__)
+    try
+    {
+        ReadBufferFromFile in("/sys/dev/block/" + device_id + "/queue/read_ahead_kb");
+        int read_ahead_kb;
+        readText(read_ahead_kb, in);
+        return read_ahead_kb * 1024;
+    }
+    catch (...)
+    {
+        return static_cast<UInt64>(-1);
+    }
+#else
+    throw DB::Exception("The function getDeviceType is supported on Linux only", ErrorCodes::NOT_IMPLEMENTED);
+#endif
+}
+
+/// Returns name of filesystem mounted to mount_point
 std::filesystem::path getMountPoint(std::filesystem::path absolute_path)
 {
     if (absolute_path.is_relative())

src/Common/filesystemHelpers.h

@@ -18,6 +18,31 @@ using TemporaryFile = Poco::TemporaryFile;
 bool enoughSpaceInDirectory(const std::string & path, size_t data_size);
 std::unique_ptr<TemporaryFile> createTemporaryFile(const std::string & path);
 
+// Determine what block device is responsible for specified path
+#if !defined(__linux__)
+[[noreturn]]
+#endif
+String getBlockDeviceId([[maybe_unused]] const String & path);
+
+enum class BlockDeviceType
+{
+    UNKNOWN = 0, // we were unable to determine device type
+    NONROT = 1, // not a rotational device (SSD, NVME, etc)
+    ROT = 2 // rotational device (HDD)
+};
+
+// Try to determine block device type
+#if !defined(__linux__)
+[[noreturn]]
+#endif
+BlockDeviceType getBlockDeviceType([[maybe_unused]] const String & device_id);
+
+// Get size of read-ahead in bytes for specified block device
+#if !defined(__linux__)
+[[noreturn]]
+#endif
+UInt64 getBlockDeviceReadAheadBytes([[maybe_unused]] const String & device_id);
+
 /// Returns mount point of filesystem where absolute_path (must exist) is located
 std::filesystem::path getMountPoint(std::filesystem::path absolute_path);
 

src/Common/getNumberOfPhysicalCPUCores.cpp

@@ -38,21 +38,7 @@ unsigned getCGroupLimitedCPUCores(unsigned default_cpu_count)
         quota_count = ceil(static_cast<float>(cgroup_quota) / static_cast<float>(cgroup_period));
     }
 
-    // Share number (typically a number relative to 1024) (2048 typically expresses 2 CPUs worth of processing)
+    return std::min(default_cpu_count, quota_count);
-    // -1 for no share setup
-    int cgroup_share = read_from("/sys/fs/cgroup/cpu/cpu.shares", -1);
-    // Convert 1024 to no shares setup
-    if (cgroup_share == 1024)
-        cgroup_share = -1;
-
-#    define PER_CPU_SHARES 1024
-    unsigned share_count = default_cpu_count;
-    if (cgroup_share > -1)
-    {
-        share_count = ceil(static_cast<float>(cgroup_share) / static_cast<float>(PER_CPU_SHARES));
-    }
-
-    return std::min(default_cpu_count, std::min(share_count, quota_count));
 }
 #endif // OS_LINUX
 
@@ -91,6 +77,7 @@ unsigned getNumberOfPhysicalCPUCores()
             cpu_count = std::thread::hardware_concurrency();
 
 #if defined(OS_LINUX)
+        /// TODO: add a setting for disabling that, similar to UseContainerSupport in java
         cpu_count = getCGroupLimitedCPUCores(cpu_count);
 #endif // OS_LINUX
         return cpu_count;

src/Common/tests/gtest_DateLUTImpl.cpp

@@ -146,6 +146,8 @@ TEST(DateLUTTest, TimeValuesInMiddleOfRange)
     EXPECT_EQ(lut.addYears(time, 10), 1884270011 /*time_t*/);
     EXPECT_EQ(lut.timeToString(time), "2019-09-16 19:20:11" /*std::string*/);
     EXPECT_EQ(lut.dateToString(time), "2019-09-16" /*std::string*/);
+    EXPECT_EQ(lut.toLastDayOfMonth(time), 1569790800 /*time_t*/);
+    EXPECT_EQ(lut.toLastDayNumOfMonth(time), DayNum(18169) /*DayNum*/);
 }
 
 
@@ -206,6 +208,8 @@ TEST(DateLUTTest, TimeValuesAtLeftBoderOfRange)
     EXPECT_EQ(lut.addYears(time, 10), 315532800 /*time_t*/);
     EXPECT_EQ(lut.timeToString(time), "1970-01-01 00:00:00" /*std::string*/);
     EXPECT_EQ(lut.dateToString(time), "1970-01-01" /*std::string*/);
+    EXPECT_EQ(lut.toLastDayOfMonth(time), 2592000 /*time_t*/);
+    EXPECT_EQ(lut.toLastDayNumOfMonth(time), DayNum(30) /*DayNum*/);
 }
 
 TEST(DateLUTTest, TimeValuesAtRightBoderOfRangeOfOldLUT)
@@ -225,7 +229,7 @@ TEST(DateLUTTest, TimeValuesAtRightBoderOfRangeOfOldLUT)
 
     EXPECT_EQ(lut.toFirstDayOfWeek(time), 4293820800 /*time_t*/);
     EXPECT_EQ(lut.toFirstDayNumOfWeek(time), DayNum(49697));
-    EXPECT_EQ(lut.toFirstDayOfMonth(time), 4291747200 /*time_t*/); // 2016-01-01
+    EXPECT_EQ(lut.toFirstDayOfMonth(time), 4291747200 /*time_t*/); // 2106-01-01
     EXPECT_EQ(lut.toFirstDayNumOfMonth(time), DayNum(49673));
     EXPECT_EQ(lut.toFirstDayNumOfQuarter(time), DayNum(49673) /*DayNum*/);
     EXPECT_EQ(lut.toFirstDayOfQuarter(time), 4291747200 /*time_t*/);
@@ -268,6 +272,8 @@ TEST(DateLUTTest, TimeValuesAtRightBoderOfRangeOfOldLUT)
 
     EXPECT_EQ(lut.timeToString(time), "2106-01-31 01:17:53" /*std::string*/);
     EXPECT_EQ(lut.dateToString(time), "2106-01-31" /*std::string*/);
+    EXPECT_EQ(lut.toLastDayOfMonth(time), 4294339200 /*time_t*/); // 2106-01-01
+    EXPECT_EQ(lut.toLastDayNumOfMonth(time), DayNum(49703));
 }
 
 

src/Common/tests/gtest_lru_file_cache.cpp

@@ -67,7 +67,7 @@ void download(DB::FileSegmentPtr file_segment)
         fs::create_directories(subdir);
 
     std::string data(size, '0');
-    file_segment->write(data.data(), size);
+    file_segment->write(data.data(), size, file_segment->getDownloadOffset());
 }
 
 void prepareAndDownload(DB::FileSegmentPtr file_segment)

src/Core/Block.cpp

@@ -46,7 +46,8 @@ static ReturnType checkColumnStructure(const ColumnWithTypeAndName & actual, con
         return onError<ReturnType>("Block structure mismatch in " + std::string(context_description) + " stream: different names of columns:\n"
             + actual.dumpStructure() + "\n" + expected.dumpStructure(), code);
 
-    if (!actual.type->equals(*expected.type))
+    if ((actual.type && !expected.type) || (!actual.type && expected.type)
+        || (actual.type && expected.type && !actual.type->equals(*expected.type)))
         return onError<ReturnType>("Block structure mismatch in " + std::string(context_description) + " stream: different types:\n"
             + actual.dumpStructure() + "\n" + expected.dumpStructure(), code);
 

src/Core/Settings.h

@@ -641,6 +641,8 @@ class IColumn;
     M(DateTimeInputFormat, date_time_input_format, FormatSettings::DateTimeInputFormat::Basic, "Method to read DateTime from text input formats. Possible values: 'basic', 'best_effort' and 'best_effort_us'.", 0) \
     M(DateTimeOutputFormat, date_time_output_format, FormatSettings::DateTimeOutputFormat::Simple, "Method to write DateTime to text output. Possible values: 'simple', 'iso', 'unix_timestamp'.", 0) \
     \
+    M(Bool, input_format_ipv4_default_on_conversion_error, false, "Deserialization of IPv4 will use default values instead of throwing exception on conversion error.", 0) \
+    M(Bool, input_format_ipv6_default_on_conversion_error, false, "Deserialization of IPV6 will use default values instead of throwing exception on conversion error.", 0) \
     M(String, bool_true_representation, "true", "Text to represent bool value in TSV/CSV formats.", 0) \
     M(String, bool_false_representation, "false", "Text to represent bool value in TSV/CSV formats.", 0) \
     \

src/Core/SortCursor.h

@@ -15,10 +15,6 @@
 
 namespace DB
 {
-namespace ErrorCodes
-{
-    extern const int LOGICAL_ERROR;
-}
 
 /** Cursor allows to compare rows in different blocks (and parts).
   * Cursor moves inside single block.
@@ -61,25 +57,21 @@ struct SortCursorImpl
         reset(block, perm);
     }
 
-    SortCursorImpl(const Columns & columns, const SortDescription & desc_, size_t order_ = 0, IColumn::Permutation * perm = nullptr)
+    SortCursorImpl(
+        const Block & header,
+        const Columns & columns,
+        const SortDescription & desc_,
+        size_t order_ = 0,
+        IColumn::Permutation * perm = nullptr)
         : desc(desc_), sort_columns_size(desc.size()), order(order_), need_collation(desc.size())
     {
-        for (auto & column_desc : desc)
+        reset(columns, header, perm);
-        {
-            if (!column_desc.column_name.empty())
-                throw Exception("SortDescription should contain column position if SortCursor was used without header.",
-                        ErrorCodes::LOGICAL_ERROR);
-        }
-        reset(columns, {}, perm);
     }
 
     bool empty() const { return rows == 0; }
 
     /// Set the cursor to the beginning of the new block.
-    void reset(const Block & block, IColumn::Permutation * perm = nullptr)
+    void reset(const Block & block, IColumn::Permutation * perm = nullptr) { reset(block.getColumns(), block, perm); }
-    {
-        reset(block.getColumns(), block, perm);
-    }
 
     /// Set the cursor to the beginning of the new block.
     void reset(const Columns & columns, const Block & block, IColumn::Permutation * perm = nullptr)
@@ -95,9 +87,7 @@ struct SortCursorImpl
         for (size_t j = 0, size = desc.size(); j < size; ++j)
         {
             auto & column_desc = desc[j];
-            size_t column_number = !column_desc.column_name.empty()
+            size_t column_number = block.getPositionByName(column_desc.column_name);
-                                   ? block.getPositionByName(column_desc.column_name)
-                                   : column_desc.column_number;
             sort_columns.push_back(columns[column_number].get());
 
             need_collation[j] = desc[j].collator != nullptr && sort_columns.back()->isCollationSupported();
@@ -367,12 +357,12 @@ private:
 };
 
 template <typename TLeftColumns, typename TRightColumns>
-bool less(const TLeftColumns & lhs, const TRightColumns & rhs, size_t i, size_t j, const SortDescription & descr)
+bool less(const TLeftColumns & lhs, const TRightColumns & rhs, size_t i, size_t j, const SortDescriptionWithPositions & descr)
 {
     for (const auto & elem : descr)
     {
         size_t ind = elem.column_number;
-        int res = elem.direction * lhs[ind]->compareAt(i, j, *rhs[ind], elem.nulls_direction);
+        int res = elem.base.direction * lhs[ind]->compareAt(i, j, *rhs[ind], elem.base.nulls_direction);
         if (res < 0)
             return true;
         else if (res > 0)

src/Core/SortDescription.cpp

@@ -1,12 +1,12 @@
-#include <Core/SortDescription.h>
 #include <Core/Block.h>
+#include <Core/SortDescription.h>
 #include <IO/Operators.h>
 #include <Common/JSONBuilder.h>
 
 namespace DB
 {
 
-void dumpSortDescription(const SortDescription & description, const Block & header, WriteBuffer & out)
+void dumpSortDescription(const SortDescription & description, WriteBuffer & out)
 {
     bool first = true;
 
@@ -16,17 +16,7 @@ void dumpSortDescription(const SortDescription & description, const Block & head
             out << ", ";
         first = false;
 
-        if (!desc.column_name.empty())
+        out << desc.column_name;
-            out << desc.column_name;
-        else
-        {
-            if (desc.column_number < header.columns())
-                out << header.getByPosition(desc.column_number).name;
-            else
-                out << "?";
-
-            out << " (pos " << desc.column_number << ")";
-        }
 
         if (desc.direction > 0)
             out << " ASC";
@@ -38,18 +28,9 @@ void dumpSortDescription(const SortDescription & description, const Block & head
     }
 }
 
-void SortColumnDescription::explain(JSONBuilder::JSONMap & map, const Block & header) const
+void SortColumnDescription::explain(JSONBuilder::JSONMap & map) const
 {
-    if (!column_name.empty())
+    map.add("Column", column_name);
-        map.add("Column", column_name);
-    else
-    {
-        if (column_number < header.columns())
-            map.add("Column", header.getByPosition(column_number).name);
-
-        map.add("Position", column_number);
-    }
-
     map.add("Ascending", direction > 0);
     map.add("With Fill", with_fill);
 }
@@ -57,17 +38,17 @@ void SortColumnDescription::explain(JSONBuilder::JSONMap & map, const Block & he
 std::string dumpSortDescription(const SortDescription & description)
 {
     WriteBufferFromOwnString wb;
-    dumpSortDescription(description, Block{}, wb);
+    dumpSortDescription(description, wb);
     return wb.str();
 }
 
-JSONBuilder::ItemPtr explainSortDescription(const SortDescription & description, const Block & header)
+JSONBuilder::ItemPtr explainSortDescription(const SortDescription & description)
 {
     auto json_array = std::make_unique<JSONBuilder::JSONArray>();
     for (const auto & descr : description)
     {
         auto json_map = std::make_unique<JSONBuilder::JSONMap>();
-        descr.explain(*json_map, header);
+        descr.explain(*json_map);
         json_array->add(std::move(json_map));
     }
 

src/Core/SortDescription.h

@@ -39,7 +39,6 @@ struct FillColumnDescription
 struct SortColumnDescription
 {
     std::string column_name; /// The name of the column.
-    size_t column_number;    /// Column number (used if no name is given).
     int direction;           /// 1 - ascending, -1 - descending.
     int nulls_direction;     /// 1 - NULLs and NaNs are greater, -1 - less.
                              /// To achieve NULLS LAST, set it equal to direction, to achieve NULLS FIRST, set it opposite.
@@ -48,23 +47,24 @@ struct SortColumnDescription
     FillColumnDescription fill_description;
 
     explicit SortColumnDescription(
-            size_t column_number_, int direction_ = 1, int nulls_direction_ = 1,
+        const std::string & column_name_,
-            const std::shared_ptr<Collator> & collator_ = nullptr,
+        int direction_ = 1,
-            bool with_fill_ = false, const FillColumnDescription & fill_description_ = {})
+        int nulls_direction_ = 1,
-            : column_number(column_number_), direction(direction_), nulls_direction(nulls_direction_), collator(collator_)
+        const std::shared_ptr<Collator> & collator_ = nullptr,
-            , with_fill(with_fill_), fill_description(fill_description_) {}
+        bool with_fill_ = false,
-
+        const FillColumnDescription & fill_description_ = {})
-    explicit SortColumnDescription(
+        : column_name(column_name_)
-            const std::string & column_name_, int direction_ = 1, int nulls_direction_ = 1,
+        , direction(direction_)
-            const std::shared_ptr<Collator> & collator_ = nullptr,
+        , nulls_direction(nulls_direction_)
-            bool with_fill_ = false, const FillColumnDescription & fill_description_ = {})
+        , collator(collator_)
-            : column_name(column_name_), column_number(0), direction(direction_), nulls_direction(nulls_direction_)
+        , with_fill(with_fill_)
-            , collator(collator_), with_fill(with_fill_), fill_description(fill_description_) {}
+        , fill_description(fill_description_)
+    {
+    }
 
     bool operator == (const SortColumnDescription & other) const
     {
-        return column_name == other.column_name && column_number == other.column_number
+        return column_name == other.column_name && direction == other.direction && nulls_direction == other.nulls_direction;
-            && direction == other.direction && nulls_direction == other.nulls_direction;
     }
 
     bool operator != (const SortColumnDescription & other) const
@@ -72,22 +72,30 @@ struct SortColumnDescription
         return !(*this == other);
     }
 
-    std::string dump() const
+    std::string dump() const { return fmt::format("{}:dir {}nulls {}", column_name, direction, nulls_direction); }
-    {
-        return fmt::format("{}:{}:dir {}nulls ", column_name, column_number, direction, nulls_direction);
-    }
 
-    void explain(JSONBuilder::JSONMap & map, const Block & header) const;
+    void explain(JSONBuilder::JSONMap & map) const;
+};
+
+struct SortColumnDescriptionWithColumnIndex
+{
+    SortColumnDescription base;
+    size_t column_number;
+
+    SortColumnDescriptionWithColumnIndex(SortColumnDescription description_, size_t column_number_)
+        : base(std::move(description_)), column_number(column_number_)
+    {
+    }
 };
 
 /// Description of the sorting rule for several columns.
 using SortDescription = std::vector<SortColumnDescription>;
+using SortDescriptionWithPositions = std::vector<SortColumnDescriptionWithColumnIndex>;
 
 /// Outputs user-readable description into `out`.
-void dumpSortDescription(const SortDescription & description, const Block & header, WriteBuffer & out);
+void dumpSortDescription(const SortDescription & description, WriteBuffer & out);
 
 std::string dumpSortDescription(const SortDescription & description);
 
-JSONBuilder::ItemPtr explainSortDescription(const SortDescription & description, const Block & header);
+JSONBuilder::ItemPtr explainSortDescription(const SortDescription & description);
-
 }

src/DataTypes/DataTypeLowCardinalityHelpers.cpp

@@ -36,7 +36,7 @@ DataTypePtr recursiveRemoveLowCardinality(const DataTypePtr & type)
             element = recursiveRemoveLowCardinality(element);
 
         if (tuple_type->haveExplicitNames())
-            return std::make_shared<DataTypeTuple>(elements, tuple_type->getElementNames(), tuple_type->serializeNames());
+            return std::make_shared<DataTypeTuple>(elements, tuple_type->getElementNames());
         else
             return std::make_shared<DataTypeTuple>(elements);
     }

src/DataTypes/DataTypeTuple.cpp

@@ -64,8 +64,8 @@ static std::optional<Exception> checkTupleNames(const Strings & names)
     return {};
 }
 
-DataTypeTuple::DataTypeTuple(const DataTypes & elems_, const Strings & names_, bool serialize_names_)
+DataTypeTuple::DataTypeTuple(const DataTypes & elems_, const Strings & names_)
-    : elems(elems_), names(names_), have_explicit_names(true), serialize_names(serialize_names_)
+    : elems(elems_), names(names_), have_explicit_names(true)
 {
     size_t size = elems.size();
     if (names.size() != size)
@@ -75,11 +75,6 @@ DataTypeTuple::DataTypeTuple(const DataTypes & elems_, const Strings & names_, b
         throw std::move(*exception);
 }
 
-bool DataTypeTuple::canBeCreatedWithNames(const Strings & names)
-{
-    return checkTupleNames(names) == std::nullopt;
-}
-
 std::string DataTypeTuple::doGetName() const
 {
     size_t size = elems.size();
@@ -91,7 +86,7 @@ std::string DataTypeTuple::doGetName() const
         if (i != 0)
             s << ", ";
 
-        if (have_explicit_names && serialize_names)
+        if (have_explicit_names)
             s << backQuoteIfNeed(names[i]) << ' ';
 
         s << elems[i]->getName();
@@ -206,7 +201,7 @@ bool DataTypeTuple::equals(const IDataType & rhs) const
         return false;
 
     for (size_t i = 0; i < size; ++i)
-        if (!elems[i]->equals(*rhs_tuple.elems[i]))
+        if (!elems[i]->equals(*rhs_tuple.elems[i]) || names[i] != rhs_tuple.names[i])
             return false;
 
     return true;
@@ -265,31 +260,29 @@ size_t DataTypeTuple::getSizeOfValueInMemory() const
 SerializationPtr DataTypeTuple::doGetDefaultSerialization() const
 {
     SerializationTuple::ElementSerializations serializations(elems.size());
-    bool use_explicit_names = have_explicit_names && serialize_names;
     for (size_t i = 0; i < elems.size(); ++i)
     {
-        String elem_name = use_explicit_names ? names[i] : toString(i + 1);
+        String elem_name = have_explicit_names ? names[i] : toString(i + 1);
         auto serialization = elems[i]->getDefaultSerialization();
         serializations[i] = std::make_shared<SerializationNamed>(serialization, elem_name);
     }
 
-    return std::make_shared<SerializationTuple>(std::move(serializations), use_explicit_names);
+    return std::make_shared<SerializationTuple>(std::move(serializations), have_explicit_names);
 }
 
 SerializationPtr DataTypeTuple::getSerialization(const SerializationInfo & info) const
 {
     SerializationTuple::ElementSerializations serializations(elems.size());
     const auto & info_tuple = assert_cast<const SerializationInfoTuple &>(info);
-    bool use_explicit_names = have_explicit_names && serialize_names;
 
     for (size_t i = 0; i < elems.size(); ++i)
     {
-        String elem_name = use_explicit_names ? names[i] : toString(i + 1);
+        String elem_name = have_explicit_names ? names[i] : toString(i + 1);
         auto serialization = elems[i]->getSerialization(*info_tuple.getElementInfo(i));
         serializations[i] = std::make_shared<SerializationNamed>(serialization, elem_name);
     }
 
-    return std::make_shared<SerializationTuple>(std::move(serializations), use_explicit_names);
+    return std::make_shared<SerializationTuple>(std::move(serializations), have_explicit_names);
 }
 
 MutableSerializationInfoPtr DataTypeTuple::createSerializationInfo(const SerializationInfo::Settings & settings) const