diff --git a/.github/workflows/jepsen.yml b/.github/workflows/jepsen.yml index 50118eaf622..db837ac1ec7 100644 --- a/.github/workflows/jepsen.yml +++ b/.github/workflows/jepsen.yml @@ -8,13 +8,13 @@ on: # yamllint disable-line rule:truthy schedule: - cron: '0 */6 * * *' workflow_dispatch: - workflow_call: jobs: KeeperJepsenRelease: uses: ./.github/workflows/reusable_simple_job.yml with: test_name: Jepsen keeper check runner_type: style-checker + report_required: true run_command: | python3 jepsen_check.py keeper # ServerJepsenRelease: diff --git a/.github/workflows/master.yml b/.github/workflows/master.yml index d2865eb737d..6d150f37a27 100644 --- a/.github/workflows/master.yml +++ b/.github/workflows/master.yml @@ -15,6 +15,8 @@ jobs: outputs: data: ${{ steps.runconfig.outputs.CI_DATA }} steps: + - name: DebugInfo + uses: hmarr/debug-action@a701ed95a46e6f2fb0df25e1a558c16356fae35a - name: Check out repository code uses: ClickHouse/checkout@v1 with: @@ -33,11 +35,9 @@ jobs: - name: PrepareRunConfig id: runconfig run: | - echo "::group::configure CI run" python3 "$GITHUB_WORKSPACE/tests/ci/ci.py" --configure --rebuild-all-binaries --outfile ${{ runner.temp }}/ci_run_data.json - echo "::endgroup::" - echo "::group::CI run configure results" + echo "::group::CI configuration" python3 -m json.tool ${{ runner.temp }}/ci_run_data.json echo "::endgroup::" @@ -255,9 +255,9 @@ jobs: run_command: | cd "$GITHUB_WORKSPACE/tests/ci" python3 docker_server.py --release-type head \ - --image-repo clickhouse/clickhouse-server --image-path docker/server + --image-repo clickhouse/clickhouse-server --image-path docker/server --allow-build-reuse python3 docker_server.py --release-type head \ - --image-repo clickhouse/clickhouse-keeper --image-path docker/keeper + --image-repo clickhouse/clickhouse-keeper --image-path docker/keeper --allow-build-reuse ############################################################################################ ##################################### BUILD REPORTER ####################################### ############################################################################################ diff --git a/.github/workflows/pull_request.yml b/.github/workflows/pull_request.yml index bd2b2b60904..b3ac2135e50 100644 --- a/.github/workflows/pull_request.yml +++ b/.github/workflows/pull_request.yml @@ -22,6 +22,8 @@ jobs: outputs: data: ${{ steps.runconfig.outputs.CI_DATA }} steps: + - name: DebugInfo + uses: hmarr/debug-action@a701ed95a46e6f2fb0df25e1a558c16356fae35a - name: Check out repository code uses: ClickHouse/checkout@v1 with: @@ -44,11 +46,9 @@ jobs: - name: PrepareRunConfig id: runconfig run: | - echo "::group::configure CI run" python3 "$GITHUB_WORKSPACE/tests/ci/ci.py" --configure --outfile ${{ runner.temp }}/ci_run_data.json - echo "::endgroup::" - echo "::group::CI run configure results" + echo "::group::CI configuration" python3 -m json.tool ${{ runner.temp }}/ci_run_data.json echo "::endgroup::" @@ -67,6 +67,7 @@ jobs: DOCKER_TAG=$(echo '${{ toJson(fromJson(steps.runconfig.outputs.CI_DATA).docker_data.images) }}' | tr -d '\n') export DOCKER_TAG=$DOCKER_TAG python3 ./tests/ci/style_check.py --no-push + python3 "$GITHUB_WORKSPACE/tests/ci/ci.py" --infile ${{ runner.temp }}/ci_run_data.json --post --job-name 'Style check' BuildDockers: needs: [RunConfig] if: ${{ !failure() && !cancelled() }} @@ -796,7 +797,7 @@ jobs: test_name: Unit tests (asan) runner_type: fuzzer-unit-tester data: ${{ needs.RunConfig.outputs.data }} - UnitTestsReleaseClang: + UnitTestsRelease: needs: [RunConfig, BuilderBinRelease] if: ${{ !failure() && !cancelled() }} uses: ./.github/workflows/reusable_test.yml @@ -923,7 +924,7 @@ jobs: - UnitTestsTsan - UnitTestsMsan - UnitTestsUBsan - - UnitTestsReleaseClang + - UnitTestsRelease - CompatibilityCheckX86 - CompatibilityCheckAarch64 - SQLancerTestRelease @@ -966,13 +967,20 @@ jobs: ############################################################################################# ###################################### JEPSEN TESTS ######################################### ############################################################################################# + # This is special test NOT INCLUDED in FinishCheck + # When it's skipped, all dependent tasks will be skipped too. + # DO NOT add it there Jepsen: - # This is special test NOT INCLUDED in FinishCheck - # When it's skipped, all dependent tasks will be skipped too. - # DO NOT add it there - if: ${{ !failure() && !cancelled() && contains(github.event.pull_request.labels.*.name, 'jepsen-test') }} + # we need concurrency as the job uses dedicated instances in the cloud + concurrency: + group: jepsen + if: ${{ !failure() && !cancelled() }} needs: [RunConfig, BuilderBinRelease] - uses: ./.github/workflows/jepsen.yml + uses: ./.github/workflows/reusable_test.yml + with: + test_name: ClickHouse Keeper Jepsen + runner_type: style-checker + data: ${{ needs.RunConfig.outputs.data }} ############################################################################################# ####################################### libFuzzer ########################################### ############################################################################################# diff --git a/.github/workflows/reusable_build.yml b/.github/workflows/reusable_build.yml index e6aa04a3569..2371579692f 100644 --- a/.github/workflows/reusable_build.yml +++ b/.github/workflows/reusable_build.yml @@ -58,6 +58,7 @@ jobs: - name: Apply sparse checkout for contrib # in order to check that it doesn't break build # This step is done in GITHUB_WORKSPACE, # because it's broken in REPO_COPY for some reason + # See also update-submodules.sh if: ${{ env.BUILD_SPARSE_CHECKOUT == 'true' }} run: | rm -rf "$GITHUB_WORKSPACE/contrib" && echo 'removed' @@ -72,12 +73,15 @@ jobs: - name: Pre run: | python3 "$GITHUB_WORKSPACE/tests/ci/ci.py" --infile ${{ toJson(inputs.data) }} --pre --job-name '${{inputs.build_name}}' - - name: Build + - name: Run run: | - python3 "$GITHUB_WORKSPACE/tests/ci/build_check.py" "$BUILD_NAME" + python3 "$GITHUB_WORKSPACE/tests/ci/ci.py" \ + --infile ${{ toJson(inputs.data) }} \ + --job-name "$BUILD_NAME" \ + --run - name: Post # it still be build report to upload for failed build job - if: always() + if: ${{ !cancelled() }} run: | python3 "$GITHUB_WORKSPACE/tests/ci/ci.py" --infile ${{ toJson(inputs.data) }} --post --job-name '${{inputs.build_name}}' - name: Mark as done diff --git a/.github/workflows/reusable_simple_job.yml b/.github/workflows/reusable_simple_job.yml index ea196a32664..7b7084420a4 100644 --- a/.github/workflows/reusable_simple_job.yml +++ b/.github/workflows/reusable_simple_job.yml @@ -34,12 +34,16 @@ name: Simple job working-directory: description: sets custom working directory type: string - default: "" + default: "$GITHUB_WORKSPACE/tests/ci" git_ref: description: commit to use, merge commit for pr or head required: false type: string default: ${{ github.event.after }} # no merge commit + report_required: + description: set to true if job report with the commit status required + type: boolean + default: false secrets: secret_envs: description: if given, it's passed to the environments @@ -58,6 +62,8 @@ jobs: env: GITHUB_JOB_OVERRIDDEN: ${{inputs.test_name}} steps: + - name: DebugInfo + uses: hmarr/debug-action@a701ed95a46e6f2fb0df25e1a558c16356fae35a - name: Check out repository code uses: ClickHouse/checkout@v1 with: @@ -79,12 +85,12 @@ jobs: job_type: test - name: Run run: | - if [ -n '${{ inputs.working-directory }}' ]; then - cd "${{ inputs.working-directory }}" - else - cd "$GITHUB_WORKSPACE/tests/ci" - fi + cd "${{ inputs.working-directory }}" ${{ inputs.run_command }} + - name: Post + if: ${{ inputs.report_required }} + run: | + python3 "$GITHUB_WORKSPACE/tests/ci/ci.py" --post --job-name '${{inputs.test_name}}' - name: Clean if: always() uses: ./.github/actions/clean diff --git a/.github/workflows/reusable_test.yml b/.github/workflows/reusable_test.yml index 09177ad887a..749f64d434e 100644 --- a/.github/workflows/reusable_test.yml +++ b/.github/workflows/reusable_test.yml @@ -38,7 +38,7 @@ name: Testing workflow working-directory: description: sets custom working directory type: string - default: "" + default: "$GITHUB_WORKSPACE/tests/ci" secrets: secret_envs: description: if given, it's passed to the environments @@ -96,19 +96,14 @@ jobs: python3 "$GITHUB_WORKSPACE/tests/ci/ci.py" --infile ${{ toJson(inputs.data) }} --pre --job-name '${{inputs.test_name}}' - name: Run run: | - if [ -n "${{ inputs.working-directory }}" ]; then - cd "${{ inputs.working-directory }}" - else - cd "$GITHUB_WORKSPACE/tests/ci" - fi - if [ -n "$(echo '${{ inputs.run_command }}' | tr -d '\n')" ]; then - echo "Running command from workflow input" - ${{ inputs.run_command }} - else - echo "Running command from job config" - python3 "$GITHUB_WORKSPACE/tests/ci/ci.py" --infile ${{ toJson(inputs.data) }} --run --job-name '${{inputs.test_name}}' - fi + cd "${{ inputs.working-directory }}" + python3 "$GITHUB_WORKSPACE/tests/ci/ci.py" \ + --infile ${{ toJson(inputs.data) }} \ + --job-name '${{inputs.test_name}}' \ + --run \ + --run-command '''${{inputs.run_command}}''' - name: Post run + if: ${{ !cancelled() }} run: | python3 "$GITHUB_WORKSPACE/tests/ci/ci.py" --infile ${{ toJson(inputs.data) }} --post --job-name '${{inputs.test_name}}' - name: Mark as done diff --git a/.gitmessage b/.gitmessage index f4a25a837bc..098b66aab1c 100644 --- a/.gitmessage +++ b/.gitmessage @@ -1,9 +1,18 @@ -## To avoid merge commit in CI run (add a leading space to apply): -#no-merge-commit +### CI modificators (add a leading space to apply): -## Running specified job (add a leading space to apply): +## To avoid a merge commit in CI: +#no_merge_commit + +## To discard CI cache: +#no_ci_cache + +## To run specified set of tests in CI: +#ci_set_ +#ci_set_reduced + +## To run specified job in CI: #job_ #job_stateless_tests_release #job_package_debug diff --git a/base/poco/Foundation/include/Poco/BufferedStreamBuf.h b/base/poco/Foundation/include/Poco/BufferedStreamBuf.h index 9f4cbd4e4d8..d97e37eedf3 100644 --- a/base/poco/Foundation/include/Poco/BufferedStreamBuf.h +++ b/base/poco/Foundation/include/Poco/BufferedStreamBuf.h @@ -26,6 +26,11 @@ #include "Poco/StreamUtil.h" +namespace DB +{ +class ReadBufferFromIStream; +} + namespace Poco { @@ -120,6 +125,8 @@ protected: openmode getMode() const { return _mode; } private: + friend class DB::ReadBufferFromIStream; + virtual int readFromDevice(char_type * /*buffer*/, std::streamsize /*length*/) { return 0; } virtual int writeToDevice(const char_type * /*buffer*/, std::streamsize /*length*/) { return 0; } diff --git a/contrib/avro b/contrib/avro index 2fb8a8a6ec0..d43acc84d3d 160000 --- a/contrib/avro +++ b/contrib/avro @@ -1 +1 @@ -Subproject commit 2fb8a8a6ec0eab9109b68abf3b4857e8c476b918 +Subproject commit d43acc84d3d455b016f847d6666fbc3cd27f16a9 diff --git a/contrib/boost-cmake/CMakeLists.txt b/contrib/boost-cmake/CMakeLists.txt index 343e863e496..2c60fc0e552 100644 --- a/contrib/boost-cmake/CMakeLists.txt +++ b/contrib/boost-cmake/CMakeLists.txt @@ -44,12 +44,14 @@ set (SRCS_IOSTREAMS "${LIBRARY_DIR}/libs/iostreams/src/gzip.cpp" "${LIBRARY_DIR}/libs/iostreams/src/mapped_file.cpp" "${LIBRARY_DIR}/libs/iostreams/src/zlib.cpp" + "${LIBRARY_DIR}/libs/iostreams/src/zstd.cpp" ) add_library (_boost_iostreams ${SRCS_IOSTREAMS}) add_library (boost::iostreams ALIAS _boost_iostreams) target_include_directories (_boost_iostreams PRIVATE ${LIBRARY_DIR}) target_link_libraries (_boost_iostreams PRIVATE ch_contrib::zlib) +target_link_libraries (_boost_iostreams PRIVATE ch_contrib::zstd) # program_options diff --git a/contrib/jemalloc-cmake/CMakeLists.txt b/contrib/jemalloc-cmake/CMakeLists.txt index f85a38dcf8a..b633f0fda50 100644 --- a/contrib/jemalloc-cmake/CMakeLists.txt +++ b/contrib/jemalloc-cmake/CMakeLists.txt @@ -34,9 +34,9 @@ if (OS_LINUX) # avoid spurious latencies and additional work associated with # MADV_DONTNEED. See # https://github.com/ClickHouse/ClickHouse/issues/11121 for motivation. - set (JEMALLOC_CONFIG_MALLOC_CONF "percpu_arena:percpu,oversize_threshold:0,muzzy_decay_ms:5000,dirty_decay_ms:5000") + set (JEMALLOC_CONFIG_MALLOC_CONF "percpu_arena:percpu,oversize_threshold:0,muzzy_decay_ms:0,dirty_decay_ms:5000") else() - set (JEMALLOC_CONFIG_MALLOC_CONF "oversize_threshold:0,muzzy_decay_ms:5000,dirty_decay_ms:5000") + set (JEMALLOC_CONFIG_MALLOC_CONF "oversize_threshold:0,muzzy_decay_ms:0,dirty_decay_ms:5000") endif() # CACHE variable is empty to allow changing defaults without the necessity # to purge cache diff --git a/contrib/update-submodules.sh b/contrib/update-submodules.sh index b12f3f924dc..7195de020bd 100755 --- a/contrib/update-submodules.sh +++ b/contrib/update-submodules.sh @@ -6,9 +6,15 @@ SCRIPT_DIR=$(dirname "${SCRIPT_PATH}") GIT_DIR=$(git -C "$SCRIPT_DIR" rev-parse --show-toplevel) cd $GIT_DIR +# Exclude from contribs some garbage subdirs that we don't need. +# It reduces the checked out files size about 3 times and therefore speeds up indexing in IDEs and searching. +# NOTE .git/ still contains everything that we don't check out (although, it's compressed) +# See also https://git-scm.com/docs/git-sparse-checkout contrib/sparse-checkout/setup-sparse-checkout.sh + git submodule init git submodule sync + # NOTE: do not use --remote for `git submodule update`[1] command, since the submodule references to the specific commit SHA1 in the subproject. # It may cause unexpected behavior. Instead you need to commit a new SHA1 for a submodule. # diff --git a/docker/server/entrypoint.sh b/docker/server/entrypoint.sh index b9c7ea34a36..78f18f376f4 100755 --- a/docker/server/entrypoint.sh +++ b/docker/server/entrypoint.sh @@ -49,17 +49,10 @@ CLICKHOUSE_PASSWORD="${CLICKHOUSE_PASSWORD:-}" CLICKHOUSE_DB="${CLICKHOUSE_DB:-}" CLICKHOUSE_ACCESS_MANAGEMENT="${CLICKHOUSE_DEFAULT_ACCESS_MANAGEMENT:-0}" -for dir in "$DATA_DIR" \ - "$ERROR_LOG_DIR" \ - "$LOG_DIR" \ - "$TMP_DIR" \ - "$USER_PATH" \ - "$FORMAT_SCHEMA_PATH" \ - "${DISKS_PATHS[@]}" \ - "${DISKS_METADATA_PATHS[@]}" -do +function create_directory_and_do_chown() { + local dir=$1 # check if variable not empty - [ -z "$dir" ] && continue + [ -z "$dir" ] && return # ensure directories exist if [ "$DO_CHOWN" = "1" ]; then mkdir="mkdir" @@ -81,6 +74,23 @@ do chown -R "$USER:$GROUP" "$dir" fi fi +} + +create_directory_and_do_chown "$DATA_DIR" + +# Change working directory to $DATA_DIR in case there're paths relative to $DATA_DIR, also avoids running +# clickhouse-server at root directory. +cd "$DATA_DIR" + +for dir in "$ERROR_LOG_DIR" \ + "$LOG_DIR" \ + "$TMP_DIR" \ + "$USER_PATH" \ + "$FORMAT_SCHEMA_PATH" \ + "${DISKS_PATHS[@]}" \ + "${DISKS_METADATA_PATHS[@]}" +do + create_directory_and_do_chown "$dir" done # if clickhouse user is defined - create it (user "default" already exists out of box) diff --git a/docker/test/stateless/run.sh b/docker/test/stateless/run.sh index 4e9486d7286..08ee52e4f1b 100755 --- a/docker/test/stateless/run.sh +++ b/docker/test/stateless/run.sh @@ -99,6 +99,16 @@ if [[ -n "$USE_DATABASE_REPLICATED" ]] && [[ "$USE_DATABASE_REPLICATED" -eq 1 ]] > /etc/clickhouse-server2/config.d/filesystem_caches_path.xml.tmp mv /etc/clickhouse-server2/config.d/filesystem_caches_path.xml.tmp /etc/clickhouse-server2/config.d/filesystem_caches_path.xml + sudo cat /etc/clickhouse-server1/config.d/filesystem_caches_path.xml \ + | sed "s|/var/lib/clickhouse/filesystem_caches/|/var/lib/clickhouse/filesystem_caches_1/|" \ + > /etc/clickhouse-server1/config.d/filesystem_caches_path.xml.tmp + mv /etc/clickhouse-server1/config.d/filesystem_caches_path.xml.tmp /etc/clickhouse-server1/config.d/filesystem_caches_path.xml + + sudo cat /etc/clickhouse-server2/config.d/filesystem_caches_path.xml \ + | sed "s|/var/lib/clickhouse/filesystem_caches/|/var/lib/clickhouse/filesystem_caches_2/|" \ + > /etc/clickhouse-server2/config.d/filesystem_caches_path.xml.tmp + mv /etc/clickhouse-server2/config.d/filesystem_caches_path.xml.tmp /etc/clickhouse-server2/config.d/filesystem_caches_path.xml + mkdir -p /var/run/clickhouse-server1 sudo chown clickhouse:clickhouse /var/run/clickhouse-server1 sudo -E -u clickhouse /usr/bin/clickhouse server --config /etc/clickhouse-server1/config.xml --daemon \ @@ -235,6 +245,17 @@ clickhouse-client -q "system flush logs" ||: # stop logs replication to make it possible to dump logs tables via clickhouse-local stop_logs_replication +# Try to get logs while server is running +successfuly_saved=0 +for table in query_log zookeeper_log trace_log transactions_info_log +do + clickhouse-client -q "select * from system.$table format TSVWithNamesAndTypes" | zstd --threads=0 > /test_output/$table.tsv.zst || successfuly_saved=$((successfuly_saved+$?)) + if [[ -n "$USE_DATABASE_REPLICATED" ]] && [[ "$USE_DATABASE_REPLICATED" -eq 1 ]]; then + clickhouse-client -q "select * from system.$table format TSVWithNamesAndTypes" | zstd --threads=0 > /test_output/$table.1.tsv.zst || successfuly_saved=$((successfuly_saved+$?)) + clickhouse-client -q "select * from system.$table format TSVWithNamesAndTypes" | zstd --threads=0 > /test_output/$table.2.tsv.zst || successfuly_saved=$((successfuly_saved+$?)) + fi +done + # Stop server so we can safely read data with clickhouse-local. # Why do we read data with clickhouse-local? # Because it's the simplest way to read it when server has crashed. @@ -254,21 +275,25 @@ if [[ -n "$USE_S3_STORAGE_FOR_MERGE_TREE" ]] && [[ "$USE_S3_STORAGE_FOR_MERGE_TR data_path_config="--config-file=/etc/clickhouse-server/config.xml" fi -# 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 transactions_info_log -do - clickhouse-local "$data_path_config" --only-system-tables -q "select * from system.$table format TSVWithNamesAndTypes" | zstd --threads=0 > /test_output/$table.tsv.zst ||: - if [[ -n "$USE_DATABASE_REPLICATED" ]] && [[ "$USE_DATABASE_REPLICATED" -eq 1 ]]; then - clickhouse-local --path /var/lib/clickhouse1/ --only-system-tables -q "select * from system.$table format TSVWithNamesAndTypes" | zstd --threads=0 > /test_output/$table.1.tsv.zst ||: - clickhouse-local --path /var/lib/clickhouse2/ --only-system-tables -q "select * from system.$table format TSVWithNamesAndTypes" | zstd --threads=0 > /test_output/$table.2.tsv.zst ||: - fi -done + +# If server crashed dump system logs with clickhouse-local +if [ $successfuly_saved -ne 0 ]; then + # Compress tables. + # + # NOTE: + # - that due to tests with s3 storage we cannot use /var/lib/clickhouse/data + # directly + # - even though ci auto-compress some files (but not *.tsv) it does this only + # for files >64MB, we want this files to be compressed explicitly + for table in query_log zookeeper_log trace_log transactions_info_log + do + clickhouse-local "$data_path_config" --only-system-tables -q "select * from system.$table format TSVWithNamesAndTypes" | zstd --threads=0 > /test_output/$table.tsv.zst ||: + if [[ -n "$USE_DATABASE_REPLICATED" ]] && [[ "$USE_DATABASE_REPLICATED" -eq 1 ]]; then + clickhouse-local --path /var/lib/clickhouse1/ --only-system-tables -q "select * from system.$table format TSVWithNamesAndTypes" | zstd --threads=0 > /test_output/$table.1.tsv.zst ||: + clickhouse-local --path /var/lib/clickhouse2/ --only-system-tables -q "select * from system.$table format TSVWithNamesAndTypes" | zstd --threads=0 > /test_output/$table.2.tsv.zst ||: + fi + done +fi # Also export trace log in flamegraph-friendly format. for trace_type in CPU Memory Real diff --git a/docker/test/upgrade/run.sh b/docker/test/upgrade/run.sh index 861fd51a3fa..9c008209316 100644 --- a/docker/test/upgrade/run.sh +++ b/docker/test/upgrade/run.sh @@ -56,6 +56,9 @@ echo "ATTACH DATABASE system ENGINE=Ordinary" > /var/lib/clickhouse/metadata/sys # Install previous release packages install_packages previous_release_package_folder +# Save old settings from system table for settings changes check +clickhouse-local -q "select * from system.settings format Native" > old_settings.native + # Initial run without S3 to create system.*_log on local file system to make it # available for dump via clickhouse-local configure @@ -152,6 +155,63 @@ install_packages package_folder export ZOOKEEPER_FAULT_INJECTION=1 configure +# Check that all new/changed setting were added in settings changes history. +# Some settings can be different for builds with sanitizers, so we check +# settings changes only for non-sanitizer builds. +IS_SANITIZED=$(clickhouse-local --query "SELECT value LIKE '%-fsanitize=%' FROM system.build_options WHERE name = 'CXX_FLAGS'") +if [ "${IS_SANITIZED}" -eq "0" ] +then + clickhouse-local -q "select * from system.settings format Native" > new_settings.native + clickhouse-local -nmq " + CREATE TABLE old_settings AS file('old_settings.native'); + CREATE TABLE new_settings AS file('new_settings.native'); + + SELECT + name, + new_settings.value AS new_value, + old_settings.value AS old_value + FROM new_settings + LEFT JOIN old_settings ON new_settings.name = old_settings.name + WHERE (new_settings.value != old_settings.value) AND (name NOT IN ( + SELECT arrayJoin(tupleElement(changes, 'name')) + FROM system.settings_changes + WHERE version = extract(version(), '^(?:\\d+\\.\\d+)') + )) + SETTINGS join_use_nulls = 1 + INTO OUTFILE 'changed_settings.txt' + FORMAT PrettyCompactNoEscapes; + + SELECT name + FROM new_settings + WHERE (name NOT IN ( + SELECT name + FROM old_settings + )) AND (name NOT IN ( + SELECT arrayJoin(tupleElement(changes, 'name')) + FROM system.settings_changes + WHERE version = extract(version(), '^(?:\\d+\\.\\d+)') + )) + INTO OUTFILE 'new_settings.txt' + FORMAT PrettyCompactNoEscapes; + " + + if [ -s changed_settings.txt ] + then + mv changed_settings.txt /test_output/ + echo -e "Changed settings are not reflected in settings changes history (see changed_settings.txt)$FAIL$(head_escaped /test_output/changed_settings.txt)" >> /test_output/test_results.tsv + else + echo -e "There are no changed settings or they are reflected in settings changes history$OK" >> /test_output/test_results.tsv + fi + + if [ -s new_settings.txt ] + then + mv new_settings.txt /test_output/ + echo -e "New settings are not reflected in settings changes history (see new_settings.txt)$FAIL$(head_escaped /test_output/new_settings.txt)" >> /test_output/test_results.tsv + else + echo -e "There are no new settings or they are reflected in settings changes history$OK" >> /test_output/test_results.tsv + fi +fi + # Just in case previous version left some garbage in zk sudo cat /etc/clickhouse-server/config.d/lost_forever_check.xml \ | sed "s|>1<|>0<|g" \ @@ -257,6 +317,8 @@ clickhouse-local --structure "test String, res String, time Nullable(Float32), d (test like '%Fatal message%') DESC, (test like '%Error message%') DESC, (test like '%previous release%') DESC, +(test like '%Changed settings%') DESC, +(test like '%New settings%') DESC, rowNumberInAllBlocks() LIMIT 1" < /test_output/test_results.tsv > /test_output/check_status.tsv || echo "failure\tCannot parse test_results.tsv" > /test_output/check_status.tsv [ -s /test_output/check_status.tsv ] || echo -e "success\tNo errors found" > /test_output/check_status.tsv diff --git a/docs/en/engines/table-engines/integrations/mysql.md b/docs/en/engines/table-engines/integrations/mysql.md index e50ed8caedd..87d59a9394b 100644 --- a/docs/en/engines/table-engines/integrations/mysql.md +++ b/docs/en/engines/table-engines/integrations/mysql.md @@ -16,7 +16,7 @@ CREATE TABLE [IF NOT EXISTS] [db.]table_name [ON CLUSTER cluster] name1 [type1] [DEFAULT|MATERIALIZED|ALIAS expr1] [TTL expr1], name2 [type2] [DEFAULT|MATERIALIZED|ALIAS expr2] [TTL expr2], ... -) ENGINE = MySQL('host:port', 'database', 'table', 'user', 'password'[, replace_query, 'on_duplicate_clause']) +) ENGINE = MySQL({host:port, database, table, user, password[, replace_query, on_duplicate_clause] | named_collection[, option=value [,..]]}) SETTINGS [ connection_pool_size=16, ] [ connection_max_tries=3, ] @@ -42,23 +42,17 @@ The MySQL Table Engine is currently not available on the ClickHouse builds for M **Engine Parameters** - `host:port` — MySQL server address. - - `database` — Remote database name. - - `table` — Remote table name. - - `user` — MySQL user. - - `password` — User password. - - `replace_query` — Flag that converts `INSERT INTO` queries to `REPLACE INTO`. If `replace_query=1`, the query is substituted. - - `on_duplicate_clause` — The `ON DUPLICATE KEY on_duplicate_clause` expression that is added to the `INSERT` query. - Example: `INSERT INTO t (c1,c2) VALUES ('a', 2) ON DUPLICATE KEY UPDATE c2 = c2 + 1`, where `on_duplicate_clause` is `UPDATE c2 = c2 + 1`. See the [MySQL documentation](https://dev.mysql.com/doc/refman/8.0/en/insert-on-duplicate.html) to find which `on_duplicate_clause` you can use with the `ON DUPLICATE KEY` clause. - To specify `on_duplicate_clause` you need to pass `0` to the `replace_query` parameter. If you simultaneously pass `replace_query = 1` and `on_duplicate_clause`, ClickHouse generates an exception. +Arguments also can be passed using [named collections](/docs/en/operations/named-collections.md). In this case `host` and `port` should be specified separately. This approach is recommended for production environment. + Simple `WHERE` clauses such as `=, !=, >, >=, <, <=` are executed on the MySQL server. The rest of the conditions and the `LIMIT` sampling constraint are executed in ClickHouse only after the query to MySQL finishes. @@ -71,7 +65,7 @@ CREATE TABLE test_replicas (id UInt32, name String, age UInt32, money UInt32) EN ## Usage Example {#usage-example} -Table in MySQL: +Create table in MySQL: ``` text mysql> CREATE TABLE `test`.`test` ( @@ -94,7 +88,7 @@ mysql> select * from test; 1 row in set (0,00 sec) ``` -Table in ClickHouse, retrieving data from the MySQL table created above: +Create table in ClickHouse using plain arguments: ``` sql CREATE TABLE mysql_table @@ -105,6 +99,25 @@ CREATE TABLE mysql_table ENGINE = MySQL('localhost:3306', 'test', 'test', 'bayonet', '123') ``` +Or using [named collections](/docs/en/operations/named-collections.md): + +```sql +CREATE NAMED COLLECTION creds AS + host = 'localhost', + port = 3306, + database = 'test', + user = 'bayonet', + password = '123'; +CREATE TABLE mysql_table +( + `float_nullable` Nullable(Float32), + `int_id` Int32 +) +ENGINE = MySQL(creds, table='test') +``` + +Retrieving data from MySQL table: + ``` sql SELECT * FROM mysql_table ``` diff --git a/docs/en/engines/table-engines/integrations/postgresql.md b/docs/en/engines/table-engines/integrations/postgresql.md index aa3dc855537..131df1a435b 100644 --- a/docs/en/engines/table-engines/integrations/postgresql.md +++ b/docs/en/engines/table-engines/integrations/postgresql.md @@ -16,7 +16,7 @@ CREATE TABLE [IF NOT EXISTS] [db.]table_name [ON CLUSTER cluster] name1 type1 [DEFAULT|MATERIALIZED|ALIAS expr1] [TTL expr1], name2 type2 [DEFAULT|MATERIALIZED|ALIAS expr2] [TTL expr2], ... -) ENGINE = PostgreSQL('host:port', 'database', 'table', 'user', 'password'[, `schema`]); +) ENGINE = PostgreSQL({host:port, database, table, user, password[, schema, [, on_conflict]] | named_collection[, option=value [,..]]}) ``` See a detailed description of the [CREATE TABLE](../../../sql-reference/statements/create/table.md#create-table-query) query. @@ -35,31 +35,25 @@ The table structure can differ from the original PostgreSQL table structure: - `user` — PostgreSQL user. - `password` — User password. - `schema` — Non-default table schema. Optional. -- `on conflict ...` — example: `ON CONFLICT DO NOTHING`. Optional. Note: adding this option will make insertion less efficient. +- `on_conflict` — Conflict resolution strategy. Example: `ON CONFLICT DO NOTHING`. Optional. Note: adding this option will make insertion less efficient. -or via config (since version 21.11): +[Named collections](/docs/en/operations/named-collections.md) (available since version 21.11) are recommended for production environment. Here is an example: ``` - - - - - -
- - - - - - - + + localhost + 5432 + postgres + **** + schema1 + ``` Some parameters can be overridden by key value arguments: ``` sql -SELECT * FROM postgresql(postgres1, schema='schema1', table='table1'); +SELECT * FROM postgresql(postgres_creds, table='table1'); ``` ## Implementation Details {#implementation-details} diff --git a/docs/en/engines/table-engines/integrations/redis.md b/docs/en/engines/table-engines/integrations/redis.md index 8086a6503b8..3a07d150835 100644 --- a/docs/en/engines/table-engines/integrations/redis.md +++ b/docs/en/engines/table-engines/integrations/redis.md @@ -16,30 +16,32 @@ CREATE TABLE [IF NOT EXISTS] [db.]table_name name1 [type1], name2 [type2], ... -) ENGINE = Redis(host:port[, db_index[, password[, pool_size]]]) PRIMARY KEY(primary_key_name); +) ENGINE = Redis({host:port[, db_index[, password[, pool_size]]] | named_collection[, option=value [,..]] }) +PRIMARY KEY(primary_key_name); ``` **Engine Parameters** - `host:port` — Redis server address, you can ignore port and default Redis port 6379 will be used. - - `db_index` — Redis db index range from 0 to 15, default is 0. - - `password` — User password, default is blank string. - - `pool_size` — Redis max connection pool size, default is 16. - - `primary_key_name` - any column name in the column list. -- `primary` must be specified, it supports only one column in the primary key. The primary key will be serialized in binary as a Redis key. +:::note Serialization +`PRIMARY KEY` supports only one column. The primary key will be serialized in binary as a Redis key. +Columns other than the primary key will be serialized in binary as Redis value in corresponding order. +::: -- columns other than the primary key will be serialized in binary as Redis value in corresponding order. +Arguments also can be passed using [named collections](/docs/en/operations/named-collections.md). In this case `host` and `port` should be specified separately. This approach is recommended for production environment. At this moment, all parameters passed using named collections to redis are required. -- queries with key equals or in filtering will be optimized to multi keys lookup from Redis. If queries without filtering key full table scan will happen which is a heavy operation. +:::note Filtering +Queries with `key equals` or `in filtering` will be optimized to multi keys lookup from Redis. If queries without filtering key full table scan will happen which is a heavy operation. +::: ## Usage Example {#usage-example} -Create a table in ClickHouse which allows to read data from Redis: +Create a table in ClickHouse using `Redis` engine with plain arguments: ``` sql CREATE TABLE redis_table @@ -52,6 +54,31 @@ CREATE TABLE redis_table ENGINE = Redis('redis1:6379') PRIMARY KEY(key); ``` +Or using [named collections](/docs/en/operations/named-collections.md): + +``` + + + localhost + 6379 + **** + 16 + s0 + + +``` + +```sql +CREATE TABLE redis_table +( + `key` String, + `v1` UInt32, + `v2` String, + `v3` Float32 +) +ENGINE = Redis(redis_creds) PRIMARY KEY(key); +``` + Insert: ```sql diff --git a/docs/en/operations/named-collections.md b/docs/en/operations/named-collections.md index 06c05929ffa..c9d94dd95ee 100644 --- a/docs/en/operations/named-collections.md +++ b/docs/en/operations/named-collections.md @@ -5,9 +5,9 @@ sidebar_label: "Named collections" title: "Named collections" --- -Named collections provide a way to store collections of key-value pairs to be +Named collections provide a way to store collections of key-value pairs to be used to configure integrations with external sources. You can use named collections with -dictionaries, tables, table functions, and object storage. +dictionaries, tables, table functions, and object storage. Named collections can be configured with DDL or in configuration files and are applied when ClickHouse starts. They simplify the creation of objects and the hiding of credentials @@ -64,7 +64,7 @@ To manage named collections with DDL a user must have the `named_control_collect ``` :::tip -In the above example the `password_sha256_hex` value is the hexadecimal representation of the SHA256 hash of the password. This configuration for the user `default` has the attribute `replace=true` as in the default configuration has a plain text `password` set, and it is not possible to have both plain text and sha256 hex passwords set for a user. +In the above example the `password_sha256_hex` value is the hexadecimal representation of the SHA256 hash of the password. This configuration for the user `default` has the attribute `replace=true` as in the default configuration has a plain text `password` set, and it is not possible to have both plain text and sha256 hex passwords set for a user. ::: ## Storing named collections in configuration files @@ -296,7 +296,6 @@ host = '127.0.0.1', port = 5432, database = 'test', schema = 'test_schema', -connection_pool_size = 8 ``` Example of configuration: @@ -310,7 +309,6 @@ Example of configuration: 5432 test test_schema - 8 @@ -445,4 +443,3 @@ SELECT dictGet('dict', 'b', 1); │ a │ └─────────────────────────┘ ``` - diff --git a/docs/en/operations/settings/query-complexity.md b/docs/en/operations/settings/query-complexity.md index 1cb7ec9dced..9a80f977ed1 100644 --- a/docs/en/operations/settings/query-complexity.md +++ b/docs/en/operations/settings/query-complexity.md @@ -172,7 +172,7 @@ If you set `timeout_before_checking_execution_speed `to 0, ClickHouse will use c ## timeout_overflow_mode {#timeout-overflow-mode} -What to do if the query is run longer than `max_execution_time`: `throw` or `break`. By default, `throw`. +What to do if the query is run longer than `max_execution_time` or the estimated running time is longer than `max_estimated_execution_time`: `throw` or `break`. By default, `throw`. # max_execution_time_leaf @@ -214,6 +214,10 @@ A maximum number of execution bytes per second. Checked on every data block when Checks that execution speed is not too slow (no less than ‘min_execution_speed’), after the specified time in seconds has expired. +## max_estimated_execution_time {#max_estimated_execution_time} + +Maximum query estimate execution time in seconds. Checked on every data block when ‘timeout_before_checking_execution_speed’ expires. + ## max_columns_to_read {#max-columns-to-read} A maximum number of columns that can be read from a table in a single query. If a query requires reading a greater number of columns, it throws an exception. diff --git a/docs/en/operations/system-tables/replication_queue.md b/docs/en/operations/system-tables/replication_queue.md index dd8f6328688..d63517291a4 100644 --- a/docs/en/operations/system-tables/replication_queue.md +++ b/docs/en/operations/system-tables/replication_queue.md @@ -49,7 +49,7 @@ Columns: - `last_attempt_time` ([DateTime](../../sql-reference/data-types/datetime.md)) — Date and time when the task was last attempted. -- `num_postponed` ([UInt32](../../sql-reference/data-types/int-uint.md)) — The number of postponed tasks. +- `num_postponed` ([UInt32](../../sql-reference/data-types/int-uint.md)) — The number of times the action was postponed. - `postpone_reason` ([String](../../sql-reference/data-types/string.md)) — The reason why the task was postponed. diff --git a/docs/en/sql-reference/aggregate-functions/reference/index.md b/docs/en/sql-reference/aggregate-functions/reference/index.md index b1f2c5bacbb..10bd3e11064 100644 --- a/docs/en/sql-reference/aggregate-functions/reference/index.md +++ b/docs/en/sql-reference/aggregate-functions/reference/index.md @@ -88,6 +88,7 @@ ClickHouse-specific aggregate functions: - [quantileTDigestWeighted](/docs/en/sql-reference/aggregate-functions/reference/quantiletdigestweighted.md) - [quantileBFloat16](/docs/en/sql-reference/aggregate-functions/reference/quantilebfloat16.md#quantilebfloat16) - [quantileBFloat16Weighted](/docs/en/sql-reference/aggregate-functions/reference/quantilebfloat16.md#quantilebfloat16weighted) +- [quantileDDSketch](/docs/en/sql-reference/aggregate-functions/reference/quantileddsketch.md#quantileddsketch) - [simpleLinearRegression](/docs/en/sql-reference/aggregate-functions/reference/simplelinearregression.md) - [stochasticLinearRegression](/docs/en/sql-reference/aggregate-functions/reference/stochasticlinearregression.md) - [stochasticLogisticRegression](/docs/en/sql-reference/aggregate-functions/reference/stochasticlogisticregression.md) diff --git a/docs/en/sql-reference/aggregate-functions/reference/median.md b/docs/en/sql-reference/aggregate-functions/reference/median.md index f20b23a0c8b..7467a47cf5f 100644 --- a/docs/en/sql-reference/aggregate-functions/reference/median.md +++ b/docs/en/sql-reference/aggregate-functions/reference/median.md @@ -18,6 +18,7 @@ Functions: - `medianTDigest` — Alias for [quantileTDigest](../../../sql-reference/aggregate-functions/reference/quantiletdigest.md#quantiletdigest). - `medianTDigestWeighted` — Alias for [quantileTDigestWeighted](../../../sql-reference/aggregate-functions/reference/quantiletdigestweighted.md#quantiletdigestweighted). - `medianBFloat16` — Alias for [quantileBFloat16](../../../sql-reference/aggregate-functions/reference/quantilebfloat16.md#quantilebfloat16). +- `medianDDSketch` — Alias for [quantileDDSketch](../../../sql-reference/aggregate-functions/reference/quantileddsketch.md#quantileddsketch). **Example** diff --git a/docs/en/sql-reference/aggregate-functions/reference/quantileddsketch.md b/docs/en/sql-reference/aggregate-functions/reference/quantileddsketch.md new file mode 100644 index 00000000000..9cb73dfc9d8 --- /dev/null +++ b/docs/en/sql-reference/aggregate-functions/reference/quantileddsketch.md @@ -0,0 +1,61 @@ +--- +slug: /en/sql-reference/aggregate-functions/reference/quantileddsketch +sidebar_position: 211 +title: quantileDDSketch +--- + +Computes an approximate [quantile](https://en.wikipedia.org/wiki/Quantile) of a sample with relative-error guarantees. It works by building a [DDSketch](https://www.vldb.org/pvldb/vol12/p2195-masson.pdf). + +**Syntax** + +``` sql +quantileDDsketch[relative_accuracy, (level)](expr) +``` + +**Arguments** + +- `expr` — Column with numeric data. [Integer](../../../sql-reference/data-types/int-uint.md), [Float](../../../sql-reference/data-types/float.md). + +**Parameters** + +- `relative_accuracy` — Relative accuracy of the quantile. Possible values are in the range from 0 to 1. [Float](../../../sql-reference/data-types/float.md). The size of the sketch depends on the range of the data and the relative accuracy. The larger the range and the smaller the relative accuracy, the larger the sketch. The rough memory size of the of the sketch is `log(max_value/min_value)/relative_accuracy`. The recommended value is 0.001 or higher. + +- `level` — Level of quantile. Optional. Possible values are in the range from 0 to 1. Default value: 0.5. [Float](../../../sql-reference/data-types/float.md). + +**Returned value** + +- Approximate quantile of the specified level. + +Type: [Float64](../../../sql-reference/data-types/float.md#float32-float64). + +**Example** + +Input table has an integer and a float columns: + +``` text +┌─a─┬─────b─┐ +│ 1 │ 1.001 │ +│ 2 │ 1.002 │ +│ 3 │ 1.003 │ +│ 4 │ 1.004 │ +└───┴───────┘ +``` + +Query to calculate 0.75-quantile (third quartile): + +``` sql +SELECT quantileDDSketch(0.01, 0.75)(a), quantileDDSketch(0.01, 0.75)(b) FROM example_table; +``` + +Result: + +``` text +┌─quantileDDSketch(0.01, 0.75)(a)─┬─quantileDDSketch(0.01, 0.75)(b)─┐ +│ 2.974233423476717 │ 1.01 │ +└─────────────────────────────────┴─────────────────────────────────┘ +``` + +**See Also** + +- [median](../../../sql-reference/aggregate-functions/reference/median.md#median) +- [quantiles](../../../sql-reference/aggregate-functions/reference/quantiles.md#quantiles) diff --git a/docs/en/sql-reference/aggregate-functions/reference/quantiles.md b/docs/en/sql-reference/aggregate-functions/reference/quantiles.md index 38db39d2eec..e5da6a9c1de 100644 --- a/docs/en/sql-reference/aggregate-functions/reference/quantiles.md +++ b/docs/en/sql-reference/aggregate-functions/reference/quantiles.md @@ -9,7 +9,7 @@ sidebar_position: 201 Syntax: `quantiles(level1, level2, …)(x)` -All the quantile functions also have corresponding quantiles functions: `quantiles`, `quantilesDeterministic`, `quantilesTiming`, `quantilesTimingWeighted`, `quantilesExact`, `quantilesExactWeighted`, `quantileInterpolatedWeighted`, `quantilesTDigest`, `quantilesBFloat16`. These functions calculate all the quantiles of the listed levels in one pass, and return an array of the resulting values. +All the quantile functions also have corresponding quantiles functions: `quantiles`, `quantilesDeterministic`, `quantilesTiming`, `quantilesTimingWeighted`, `quantilesExact`, `quantilesExactWeighted`, `quantileInterpolatedWeighted`, `quantilesTDigest`, `quantilesBFloat16`, `quantilesDDSketch`. These functions calculate all the quantiles of the listed levels in one pass, and return an array of the resulting values. ## quantilesExactExclusive diff --git a/docs/en/sql-reference/data-types/nullable.md b/docs/en/sql-reference/data-types/nullable.md index 28180f7f991..5504765e4a0 100644 --- a/docs/en/sql-reference/data-types/nullable.md +++ b/docs/en/sql-reference/data-types/nullable.md @@ -4,11 +4,11 @@ sidebar_position: 55 sidebar_label: Nullable --- -# Nullable(typename) +# Nullable(T) -Allows to store special marker ([NULL](../../sql-reference/syntax.md)) that denotes “missing value” alongside normal values allowed by `TypeName`. For example, a `Nullable(Int8)` type column can store `Int8` type values, and the rows that do not have a value will store `NULL`. +Allows to store special marker ([NULL](../../sql-reference/syntax.md)) that denotes “missing value” alongside normal values allowed by `T`. For example, a `Nullable(Int8)` type column can store `Int8` type values, and the rows that do not have a value will store `NULL`. -For a `TypeName`, you can’t use composite data types [Array](../../sql-reference/data-types/array.md), [Map](../../sql-reference/data-types/map.md) and [Tuple](../../sql-reference/data-types/tuple.md). Composite data types can contain `Nullable` type values, such as `Array(Nullable(Int8))`. +`T` can’t be any of the composite data types [Array](../../sql-reference/data-types/array.md), [Map](../../sql-reference/data-types/map.md) and [Tuple](../../sql-reference/data-types/tuple.md) but composite data types can contain `Nullable` type values, e.g. `Array(Nullable(Int8))`. A `Nullable` type field can’t be included in table indexes. diff --git a/docs/en/sql-reference/functions/array-functions.md b/docs/en/sql-reference/functions/array-functions.md index 1639f45e66c..2120b675c73 100644 --- a/docs/en/sql-reference/functions/array-functions.md +++ b/docs/en/sql-reference/functions/array-functions.md @@ -657,6 +657,43 @@ SELECT arraySlice([1, 2, NULL, 4, 5], 2, 3) AS res; Array elements set to `NULL` are handled as normal values. +## arrayShingles + +Generates an array of "shingles", i.e. consecutive sub-arrays with specified length of the input array. + +**Syntax** + +``` sql +arrayShingles(array, length) +``` + +**Arguments** + +- `array` — Input array [Array](../../sql-reference/data-types/array.md). +- `length` — The length of each shingle. + +**Returned value** + +- An array of generated shingles. + +Type: [Array](../../sql-reference/data-types/array.md). + +**Examples** + +Query: + +``` sql +SELECT arrayShingles([1,2,3,4], 3) as res; +``` + +Result: + +``` text +┌─res───────────────┐ +│ [[1,2,3],[2,3,4]] │ +└───────────────────┘ +``` + ## arraySort(\[func,\] arr, …) {#sort} Sorts the elements of the `arr` array in ascending order. If the `func` function is specified, sorting order is determined by the result of the `func` function applied to the elements of the array. If `func` accepts multiple arguments, the `arraySort` function is passed several arrays that the arguments of `func` will correspond to. Detailed examples are shown at the end of `arraySort` description. diff --git a/docs/en/sql-reference/table-functions/executable.md b/docs/en/sql-reference/table-functions/executable.md index d377c5d4d0c..0f57fe943a7 100644 --- a/docs/en/sql-reference/table-functions/executable.md +++ b/docs/en/sql-reference/table-functions/executable.md @@ -63,7 +63,7 @@ if __name__ == "__main__": Let's invoke the script and have it generate 10 random strings: ```sql -SELECT * FROM executable('my_script.py', TabSeparated, 'id UInt32, random String', (SELECT 10)) +SELECT * FROM executable('generate_random.py', TabSeparated, 'id UInt32, random String', (SELECT 10)) ``` The response looks like: diff --git a/docs/en/sql-reference/table-functions/fuzzJSON.md b/docs/en/sql-reference/table-functions/fuzzJSON.md index a64f35691f6..ab7bd7f9f1b 100644 --- a/docs/en/sql-reference/table-functions/fuzzJSON.md +++ b/docs/en/sql-reference/table-functions/fuzzJSON.md @@ -9,7 +9,7 @@ sidebar_label: fuzzJSON Perturbs a JSON string with random variations. ``` sql -fuzzJSON({ named_collection [option=value [,..]] | json_str[, random_seed] }) +fuzzJSON({ named_collection [, option=value [,..]] | json_str[, random_seed] }) ``` **Arguments** diff --git a/docs/en/sql-reference/table-functions/gcs.md b/docs/en/sql-reference/table-functions/gcs.md index 5ffc20189da..80077ecdb33 100644 --- a/docs/en/sql-reference/table-functions/gcs.md +++ b/docs/en/sql-reference/table-functions/gcs.md @@ -16,7 +16,8 @@ If you have multiple replicas in your cluster, you can use the [s3Cluster functi **Syntax** ``` sql -gcs(path [,hmac_key, hmac_secret] [,format] [,structure] [,compression]) +gcs(url [, NOSIGN | hmac_key, hmac_secret] [,format] [,structure] [,compression_method]) +gcs(named_collection[, option=value [,..]]) ``` :::tip GCS @@ -24,10 +25,9 @@ The GCS Table Function integrates with Google Cloud Storage by using the GCS XML ::: -**Arguments** - -- `path` — Bucket url with path to file. Supports following wildcards in readonly mode: `*`, `**`, `?`, `{abc,def}` and `{N..M}` where `N`, `M` — numbers, `'abc'`, `'def'` — strings. +**Parameters** +- `url` — Bucket path to file. Supports following wildcards in readonly mode: `*`, `**`, `?`, `{abc,def}` and `{N..M}` where `N`, `M` — numbers, `'abc'`, `'def'` — strings. :::note GCS The GCS path is in this format as the endpoint for the Google XML API is different than the JSON API: ``` @@ -35,10 +35,21 @@ The GCS Table Function integrates with Google Cloud Storage by using the GCS XML ``` and not ~~https://storage.cloud.google.com~~. ::: +- `NOSIGN` — If this keyword is provided in place of credentials, all the requests will not be signed. +- `hmac_key` and `hmac_secret` — Keys that specify credentials to use with given endpoint. Optional. +- `format` — The [format](../../interfaces/formats.md#formats) of the file. +- `structure` — Structure of the table. Format `'column1_name column1_type, column2_name column2_type, ...'`. +- `compression_method` — Parameter is optional. Supported values: `none`, `gzip/gz`, `brotli/br`, `xz/LZMA`, `zstd/zst`. By default, it will autodetect compression method by file extension. + +Arguments can also be passed using [named collections](/docs/en/operations/named-collections.md). In this case `url`, `format`, `structure`, `compression_method` work in the same way, and some extra parameters are supported: + + - `access_key_id` — `hmac_key`, optional. + - `secret_access_key` — `hmac_secret`, optional. + - `filename` — appended to the url if specified. + - `use_environment_credentials` — enabled by default, allows passing extra parameters using environment variables `AWS_CONTAINER_CREDENTIALS_RELATIVE_URI`, `AWS_CONTAINER_CREDENTIALS_FULL_URI`, `AWS_CONTAINER_AUTHORIZATION_TOKEN`, `AWS_EC2_METADATA_DISABLED`. + - `no_sign_request` — disabled by default. + - `expiration_window_seconds` — default value is 120. -- `format` — The [format](../../interfaces/formats.md#formats) of the file. -- `structure` — Structure of the table. Format `'column1_name column1_type, column2_name column2_type, ...'`. -- `compression` — Parameter is optional. Supported values: `none`, `gzip/gz`, `brotli/br`, `xz/LZMA`, `zstd/zst`. By default, it will autodetect compression by file extension. **Returned value** @@ -61,7 +72,7 @@ LIMIT 2; └─────────┴─────────┴─────────┘ ``` -The similar but from file with `gzip` compression: +The similar but from file with `gzip` compression method: ``` sql SELECT * @@ -158,6 +169,16 @@ The below get data from all `test-data.csv.gz` files from any folder inside `my- SELECT * FROM gcs('https://storage.googleapis.com/my-test-bucket-768/**/test-data.csv.gz', 'CSV', 'name String, value UInt32', 'gzip'); ``` +For production use cases it is recommended to use [named collections](/docs/en/operations/named-collections.md). Here is the example: +``` sql + +CREATE NAMED COLLECTION creds AS + access_key_id = '***', + secret_access_key = '***'; +SELECT count(*) +FROM gcs(creds, url='https://s3-object-url.csv') +``` + ## Partitioned Write If you specify `PARTITION BY` expression when inserting data into `GCS` table, a separate file is created for each partition value. Splitting the data into separate files helps to improve reading operations efficiency. diff --git a/docs/en/sql-reference/table-functions/mysql.md b/docs/en/sql-reference/table-functions/mysql.md index 0e5b0f54d1c..5fd9708317c 100644 --- a/docs/en/sql-reference/table-functions/mysql.md +++ b/docs/en/sql-reference/table-functions/mysql.md @@ -11,31 +11,25 @@ Allows `SELECT` and `INSERT` queries to be performed on data that is stored on a **Syntax** ``` sql -mysql('host:port', 'database', 'table', 'user', 'password'[, replace_query, 'on_duplicate_clause']) +mysql({host:port, database, table, user, password[, replace_query, on_duplicate_clause] | named_collection[, option=value [,..]]}) ``` -**Arguments** +**Parameters** - `host:port` — MySQL server address. - - `database` — Remote database name. - - `table` — Remote table name. - - `user` — MySQL user. - - `password` — User password. - - `replace_query` — Flag that converts `INSERT INTO` queries to `REPLACE INTO`. Possible values: - `0` - The query is executed as `INSERT INTO`. - `1` - The query is executed as `REPLACE INTO`. - - `on_duplicate_clause` — The `ON DUPLICATE KEY on_duplicate_clause` expression that is added to the `INSERT` query. Can be specified only with `replace_query = 0` (if you simultaneously pass `replace_query = 1` and `on_duplicate_clause`, ClickHouse generates an exception). - Example: `INSERT INTO t (c1,c2) VALUES ('a', 2) ON DUPLICATE KEY UPDATE c2 = c2 + 1;` - `on_duplicate_clause` here is `UPDATE c2 = c2 + 1`. See the MySQL documentation to find which `on_duplicate_clause` you can use with the `ON DUPLICATE KEY` clause. +Arguments also can be passed using [named collections](/docs/en/operations/named-collections.md). In this case `host` and `port` should be specified separately. This approach is recommended for production environment. + Simple `WHERE` clauses such as `=, !=, >, >=, <, <=` are currently executed on the MySQL server. The rest of the conditions and the `LIMIT` sampling constraint are executed in ClickHouse only after the query to MySQL finishes. @@ -86,6 +80,18 @@ Selecting data from ClickHouse: SELECT * FROM mysql('localhost:3306', 'test', 'test', 'bayonet', '123'); ``` +Or using [named collections](/docs/en/operations/named-collections.md): + +```sql +CREATE NAMED COLLECTION creds AS + host = 'localhost', + port = 3306, + database = 'test', + user = 'bayonet', + password = '123'; +SELECT * FROM mysql(creds, table='test'); +``` + ``` text ┌─int_id─┬─float─┐ │ 1 │ 2 │ diff --git a/docs/en/sql-reference/table-functions/postgresql.md b/docs/en/sql-reference/table-functions/postgresql.md index b9211d70cdb..3fd0e5805e7 100644 --- a/docs/en/sql-reference/table-functions/postgresql.md +++ b/docs/en/sql-reference/table-functions/postgresql.md @@ -11,10 +11,10 @@ Allows `SELECT` and `INSERT` queries to be performed on data that is stored on a **Syntax** ``` sql -postgresql('host:port', 'database', 'table', 'user', 'password'[, `schema`]) +postgresql({host:port, database, table, user, password[, schema, [, on_conflict]] | named_collection[, option=value [,..]]}) ``` -**Arguments** +**Parameters** - `host:port` — PostgreSQL server address. - `database` — Remote database name. @@ -22,6 +22,9 @@ postgresql('host:port', 'database', 'table', 'user', 'password'[, `schema`]) - `user` — PostgreSQL user. - `password` — User password. - `schema` — Non-default table schema. Optional. +- `on_conflict` — Conflict resolution strategy. Example: `ON CONFLICT DO NOTHING`. Optional. + +Arguments also can be passed using [named collections](/docs/en/operations/named-collections.md). In this case `host` and `port` should be specified separately. This approach is recommended for production environment. **Returned Value** @@ -86,12 +89,24 @@ postgresql> SELECT * FROM test; (1 row) ``` -Selecting data from ClickHouse: +Selecting data from ClickHouse using plain arguments: ```sql SELECT * FROM postgresql('localhost:5432', 'test', 'test', 'postgresql_user', 'password') WHERE str IN ('test'); ``` +Or using [named collections](/docs/en/operations/named-collections.md): + +```sql +CREATE NAMED COLLECTION mypg AS + host = 'localhost', + port = 5432, + database = 'test', + user = 'postgresql_user', + password = 'password'; +SELECT * FROM postgresql(mypg, table='test') WHERE str IN ('test'); +``` + ``` text ┌─int_id─┬─int_nullable─┬─float─┬─str──┬─float_nullable─┐ │ 1 │ ᴺᵁᴸᴸ │ 2 │ test │ ᴺᵁᴸᴸ │ diff --git a/docs/en/sql-reference/table-functions/redis.md b/docs/en/sql-reference/table-functions/redis.md index 98d9a647cee..09841642210 100644 --- a/docs/en/sql-reference/table-functions/redis.md +++ b/docs/en/sql-reference/table-functions/redis.md @@ -34,6 +34,7 @@ redis(host:port, key, structure[, db_index[, password[, pool_size]]]) - queries with key equals or in filtering will be optimized to multi keys lookup from Redis. If queries without filtering key full table scan will happen which is a heavy operation. +[Named collections](/docs/en/operations/named-collections.md) are not supported for `redis` table function at the moment. **Returned Value** @@ -41,17 +42,7 @@ A table object with key as Redis key, other columns packaged together as Redis v ## Usage Example {#usage-example} -Create a table in ClickHouse which allows to read data from Redis: - -``` sql -CREATE TABLE redis_table -( - `k` String, - `m` String, - `n` UInt32 -) -ENGINE = Redis('redis1:6379') PRIMARY KEY(k); -``` +Read from Redis: ```sql SELECT * FROM redis( @@ -61,6 +52,15 @@ SELECT * FROM redis( ) ``` +Insert into Redis: + +```sql +INSERT INTO TABLE FUNCTION redis( + 'redis1:6379', + 'key', + 'key String, v1 String, v2 UInt32') values ('1', '1', 1); +``` + **See Also** - [The `Redis` table engine](/docs/en/engines/table-engines/integrations/redis.md) diff --git a/docs/en/sql-reference/table-functions/remote.md b/docs/en/sql-reference/table-functions/remote.md index 228f4a4c7e1..f6e49099d99 100644 --- a/docs/en/sql-reference/table-functions/remote.md +++ b/docs/en/sql-reference/table-functions/remote.md @@ -13,10 +13,12 @@ Both functions can be used in `SELECT` and `INSERT` queries. ## Syntax ``` sql -remote('addresses_expr', [db, table, 'user'[, 'password'], sharding_key]) -remote('addresses_expr', [db.table, 'user'[, 'password'], sharding_key]) -remoteSecure('addresses_expr', [db, table, 'user'[, 'password'], sharding_key]) -remoteSecure('addresses_expr', [db.table, 'user'[, 'password'], sharding_key]) +remote(addresses_expr, [db, table, user [, password], sharding_key]) +remote(addresses_expr, [db.table, user [, password], sharding_key]) +remote(named_collection[, option=value [,..]]) +remoteSecure(addresses_expr, [db, table, user [, password], sharding_key]) +remoteSecure(addresses_expr, [db.table, user [, password], sharding_key]) +remoteSecure(named_collection[, option=value [,..]]) ``` ## Parameters @@ -39,6 +41,8 @@ remoteSecure('addresses_expr', [db.table, 'user'[, 'password'], sharding_key]) - `password` — User password. If not specified, an empty password is used. Type: [String](../../sql-reference/data-types/string.md). - `sharding_key` — Sharding key to support distributing data across nodes. For example: `insert into remote('127.0.0.1:9000,127.0.0.2', db, table, 'default', rand())`. Type: [UInt32](../../sql-reference/data-types/int-uint.md). +Arguments also can be passed using [named collections](/docs/en/operations/named-collections.md). + ## Returned value A table located on a remote server. @@ -82,7 +86,16 @@ example01-01-1,example01-02-1 SELECT * FROM remote('127.0.0.1', db.remote_engine_table) LIMIT 3; ``` -### Inserting data from a remote server into a table: +Or using [named collections](/docs/en/operations/named-collections.md): + +```sql +CREATE NAMED COLLECTION creds AS + host = '127.0.0.1', + database = 'db'; +SELECT * FROM remote(creds, table='remote_engine_table') LIMIT 3; +``` + +### Inserting data into a table on a remote server: ``` sql CREATE TABLE remote_table (name String, value UInt32) ENGINE=Memory; diff --git a/docs/en/sql-reference/table-functions/s3.md b/docs/en/sql-reference/table-functions/s3.md index 8065f066666..970b3e52882 100644 --- a/docs/en/sql-reference/table-functions/s3.md +++ b/docs/en/sql-reference/table-functions/s3.md @@ -16,33 +16,41 @@ When using the `s3 table function` with [`INSERT INTO...SELECT`](../../sql-refer **Syntax** ``` sql -s3(path [, NOSIGN | aws_access_key_id, aws_secret_access_key [,session_token]] [,format] [,structure] [,compression]) +s3(url [, NOSIGN | access_key_id, secret_access_key, [session_token]] [,format] [,structure] [,compression_method]) +s3(named_collection[, option=value [,..]]) ``` :::tip GCS The S3 Table Function integrates with Google Cloud Storage by using the GCS XML API and HMAC keys. See the [Google interoperability docs]( https://cloud.google.com/storage/docs/interoperability) for more details about the endpoint and HMAC. -For GCS, substitute your HMAC key and HMAC secret where you see `aws_access_key_id` and `aws_secret_access_key`. +For GCS, substitute your HMAC key and HMAC secret where you see `access_key_id` and `secret_access_key`. ::: -**Arguments** +**Parameters** -- `path` — Bucket url with path to file. Supports following wildcards in readonly mode: `*`, `**`, `?`, `{abc,def}` and `{N..M}` where `N`, `M` — numbers, `'abc'`, `'def'` — strings. For more information see [here](../../engines/table-engines/integrations/s3.md#wildcards-in-path). +`s3` table function supports the following plain parameters: +- `url` — Bucket url with path to file. Supports following wildcards in readonly mode: `*`, `**`, `?`, `{abc,def}` and `{N..M}` where `N`, `M` — numbers, `'abc'`, `'def'` — strings. For more information see [here](../../engines/table-engines/integrations/s3.md#wildcards-in-path). :::note GCS - The GCS path is in this format as the endpoint for the Google XML API is different than the JSON API: + The GCS url is in this format as the endpoint for the Google XML API is different than the JSON API: ``` https://storage.googleapis.com/// ``` and not ~~https://storage.cloud.google.com~~. ::: - -- `NOSIGN` - If this keyword is provided in place of credentials, all the requests will not be signed. -- `access_key_id`, `secret_access_key` — Keys that specify credentials to use with given endpoint. Optional. +- `NOSIGN` — If this keyword is provided in place of credentials, all the requests will not be signed. +- `access_key_id` and `secret_access_key` — Keys that specify credentials to use with given endpoint. Optional. - `session_token` - Session token to use with the given keys. Optional when passing keys. - `format` — The [format](../../interfaces/formats.md#formats) of the file. - `structure` — Structure of the table. Format `'column1_name column1_type, column2_name column2_type, ...'`. -- `compression` — Parameter is optional. Supported values: `none`, `gzip/gz`, `brotli/br`, `xz/LZMA`, `zstd/zst`. By default, it will autodetect compression by file extension. +- `compression_method` — Parameter is optional. Supported values: `none`, `gzip/gz`, `brotli/br`, `xz/LZMA`, `zstd/zst`. By default, it will autodetect compression method by file extension. + +Arguments can also be passed using [named collections](/docs/en/operations/named-collections.md). In this case `url`, `access_key_id`, `secret_access_key`, `format`, `structure`, `compression_method` work in the same way, and some extra parameters are supported: + + - `filename` — appended to the url if specified. + - `use_environment_credentials` — enabled by default, allows passing extra parameters using environment variables `AWS_CONTAINER_CREDENTIALS_RELATIVE_URI`, `AWS_CONTAINER_CREDENTIALS_FULL_URI`, `AWS_CONTAINER_AUTHORIZATION_TOKEN`, `AWS_EC2_METADATA_DISABLED`. + - `no_sign_request` — disabled by default. + - `expiration_window_seconds` — default value is 120. **Returned value** @@ -82,7 +90,7 @@ FROM s3( LIMIT 5; ``` -ClickHouse also can determine the compression of the file. For example, if the file was zipped up with a `.csv.gz` extension, ClickHouse would decompress the file automatically. +ClickHouse also can determine the compression method of the file. For example, if the file was zipped up with a `.csv.gz` extension, ClickHouse would decompress the file automatically. ::: @@ -168,7 +176,7 @@ The below get data from all `test-data.csv.gz` files from any folder inside `my- SELECT * FROM s3('https://clickhouse-public-datasets.s3.amazonaws.com/my-test-bucket-768/**/test-data.csv.gz', 'CSV', 'name String, value UInt32', 'gzip'); ``` -Note. It is possible to specify custom URL mappers in the server configuration file. Example: +Note. It is possible to specify custom URL mappers in the server configuration file. Example: ``` sql SELECT * FROM s3('s3://clickhouse-public-datasets/my-test-bucket-768/**/test-data.csv.gz', 'CSV', 'name String, value UInt32', 'gzip'); ``` @@ -190,6 +198,16 @@ Custom mapper can be added into `config.xml`: ``` +For production use cases it is recommended to use [named collections](/docs/en/operations/named-collections.md). Here is the example: +``` sql + +CREATE NAMED COLLECTION creds AS + access_key_id = '***', + secret_access_key = '***'; +SELECT count(*) +FROM s3(creds, url='https://s3-object-url.csv') +``` + ## Partitioned Write If you specify `PARTITION BY` expression when inserting data into `S3` table, a separate file is created for each partition value. Splitting the data into separate files helps to improve reading operations efficiency. diff --git a/docs/en/sql-reference/table-functions/s3Cluster.md b/docs/en/sql-reference/table-functions/s3Cluster.md index 080c9860519..92d9527df82 100644 --- a/docs/en/sql-reference/table-functions/s3Cluster.md +++ b/docs/en/sql-reference/table-functions/s3Cluster.md @@ -4,23 +4,34 @@ sidebar_position: 181 sidebar_label: s3Cluster title: "s3Cluster Table Function" --- +This is an extension to the [s3](/docs/en/sql-reference/table-functions/s3.md) table function. Allows processing files from [Amazon S3](https://aws.amazon.com/s3/) and Google Cloud Storage [Google Cloud Storage](https://cloud.google.com/storage/) in parallel from many nodes in a specified cluster. On initiator it creates a connection to all nodes in the cluster, discloses asterisks in S3 file path, and dispatches each file dynamically. On the worker node it asks the initiator about the next task to process and processes it. This is repeated until all tasks are finished. **Syntax** ``` sql -s3Cluster(cluster_name, source, [,access_key_id, secret_access_key, [session_token]] [,format] [,structure]) +s3Cluster(cluster_name, url [, NOSIGN | access_key_id, secret_access_key, [session_token]] [,format] [,structure] [,compression_method]) +s3Cluster(cluster_name, named_collection[, option=value [,..]]) ``` **Arguments** - `cluster_name` — Name of a cluster that is used to build a set of addresses and connection parameters to remote and local servers. -- `source` — URL to a file or a bunch of files. Supports following wildcards in readonly mode: `*`, `**`, `?`, `{'abc','def'}` and `{N..M}` where `N`, `M` — numbers, `abc`, `def` — strings. For more information see [Wildcards In Path](../../engines/table-engines/integrations/s3.md#wildcards-in-path). -- `access_key_id`, `secret_access_key` — Keys that specify credentials to use with given endpoint. Optional. +- `url` — path to a file or a bunch of files. Supports following wildcards in readonly mode: `*`, `**`, `?`, `{'abc','def'}` and `{N..M}` where `N`, `M` — numbers, `abc`, `def` — strings. For more information see [Wildcards In Path](../../engines/table-engines/integrations/s3.md#wildcards-in-path). +- `NOSIGN` — If this keyword is provided in place of credentials, all the requests will not be signed. +- `access_key_id` and `secret_access_key` — Keys that specify credentials to use with given endpoint. Optional. - `session_token` - Session token to use with the given keys. Optional when passing keys. - `format` — The [format](../../interfaces/formats.md#formats) of the file. - `structure` — Structure of the table. Format `'column1_name column1_type, column2_name column2_type, ...'`. +- `compression_method` — Parameter is optional. Supported values: `none`, `gzip/gz`, `brotli/br`, `xz/LZMA`, `zstd/zst`. By default, it will autodetect compression method by file extension. + +Arguments can also be passed using [named collections](/docs/en/operations/named-collections.md). In this case `url`, `access_key_id`, `secret_access_key`, `format`, `structure`, `compression_method` work in the same way, and some extra parameters are supported: + + - `filename` — appended to the url if specified. + - `use_environment_credentials` — enabled by default, allows passing extra parameters using environment variables `AWS_CONTAINER_CREDENTIALS_RELATIVE_URI`, `AWS_CONTAINER_CREDENTIALS_FULL_URI`, `AWS_CONTAINER_AUTHORIZATION_TOKEN`, `AWS_EC2_METADATA_DISABLED`. + - `no_sign_request` — disabled by default. + - `expiration_window_seconds` — default value is 120. **Returned value** @@ -47,6 +58,18 @@ Count the total amount of rows in all files in the cluster `cluster_simple`: If your listing of files contains number ranges with leading zeros, use the construction with braces for each digit separately or use `?`. ::: +For production use cases it is recommended to use [named collections](/docs/en/operations/named-collections.md). Here is the example: +``` sql + +CREATE NAMED COLLECTION creds AS + access_key_id = 'minio' + secret_access_key = 'minio123'; +SELECT count(*) FROM s3Cluster( + 'cluster_simple', creds, url='https://s3-object-url.csv', + format='CSV', structure='name String, value UInt32, polygon Array(Array(Tuple(Float64, Float64)))' +) +``` + **See Also** - [S3 engine](../../engines/table-engines/integrations/s3.md) diff --git a/docs/ru/getting-started/tutorial.md b/docs/ru/getting-started/tutorial.md index a2ddb103bc3..34064b6cf2f 100644 --- a/docs/ru/getting-started/tutorial.md +++ b/docs/ru/getting-started/tutorial.md @@ -670,4 +670,4 @@ ENGINE = ReplicatedMergeTree( INSERT INTO tutorial.hits_replica SELECT * FROM tutorial.hits_local; ``` -Репликация работает в режиме мультимастера. Это означает, что данные могут быть загружены на любую из реплик и система автоматически синхронизирует данные между остальными репликами. Репликация асинхронна, то есть в конкретный момент времнени не все реплики могут содержать недавно добавленные данные. Как минимум одна реплика должна быть в строю для приёма данных. Прочие реплики синхронизируются и восстановят согласованное состояния как только снова станут активными. Заметим, что при таком подходе есть вероятность утраты недавно добавленных данных. +Репликация работает в режиме мультимастера. Это означает, что данные могут быть загружены на любую из реплик и система автоматически синхронизирует данные между остальными репликами. Репликация асинхронна, то есть в конкретный момент времени не все реплики могут содержать недавно добавленные данные. Как минимум одна реплика должна быть в строю для приёма данных. Прочие реплики синхронизируются и восстановят согласованное состояния как только снова станут активными. Заметим, что при таком подходе есть вероятность утраты недавно добавленных данных. diff --git a/docs/ru/operations/system-tables/replication_queue.md b/docs/ru/operations/system-tables/replication_queue.md index 60d42133153..31bd0bf50fd 100644 --- a/docs/ru/operations/system-tables/replication_queue.md +++ b/docs/ru/operations/system-tables/replication_queue.md @@ -49,7 +49,7 @@ slug: /ru/operations/system-tables/replication_queue - `last_attempt_time` ([DateTime](../../sql-reference/data-types/datetime.md)) — дата и время последней попытки выполнить задачу. -- `num_postponed` ([UInt32](../../sql-reference/data-types/int-uint.md)) — количество отложенных задач. +- `num_postponed` ([UInt32](../../sql-reference/data-types/int-uint.md)) — количество откладываний запуска задачи. - `postpone_reason` ([String](../../sql-reference/data-types/string.md)) — причина, по которой была отложена задача. diff --git a/docs/zh/faq/general/ne-tormozit.md b/docs/zh/faq/general/ne-tormozit.md index c4149655108..f397f6bb1d6 100644 --- a/docs/zh/faq/general/ne-tormozit.md +++ b/docs/zh/faq/general/ne-tormozit.md @@ -1,27 +1,27 @@ --- slug: /zh/faq/general/ne-tormozit -title: "What does \u201C\u043D\u0435 \u0442\u043E\u0440\u043C\u043E\u0437\u0438\u0442\ - \u201D mean?" +title: "\u201C\u043D\u0435 \u0442\u043E\u0440\u043C\u043E\u0437\u0438\u0442\ + \u201D 是什么意思？" toc_hidden: true sidebar_position: 11 --- -# What Does “Не тормозит” Mean? {#what-does-ne-tormozit-mean} +# “Не тормозит” 是什么意思？ {#what-does-ne-tormozit-mean} -This question usually arises when people see official ClickHouse t-shirts. They have large words **“ClickHouse не тормозит”** on the front. +这个问题通常出现在人们看到官方 ClickHouse T恤时。它们的正面印有大字**“ClickHouse не тормозит”**。 -Before ClickHouse became open-source, it has been developed as an in-house storage system by the largest Russian IT company, [Yandex](https://yandex.com/company/). That’s why it initially got its slogan in Russian, which is “не тормозит” (pronounced as “ne tormozit”). After the open-source release we first produced some of those t-shirts for events in Russia and it was a no-brainer to use the slogan as-is. +在 ClickHouse 开源之前，它作为俄罗斯最大的 IT 公司 [Yandex](https://yandex.com/company/) 的内部存储系统而开发。这就是为什么它最初获得了俄文口号“не тормозит”（发音为“ne tormozit”）。在开源发布后，我们首先为俄罗斯的活动制作了一些这样的T恤，使用原汁原味的口号是理所当然的。 -One of the following batches of those t-shirts was supposed to be given away on events outside of Russia and we tried to make the English version of the slogan. Unfortunately, the Russian language is kind of elegant in terms of expressing stuff and there was a restriction of limited space on a t-shirt, so we failed to come up with good enough translation (most options appeared to be either long or inaccurate) and decided to keep the slogan in Russian even on t-shirts produced for international events. It appeared to be a great decision because people all over the world get positively surprised and curious when they see it. +其中一批这样的T恤原本打算在俄罗斯之外的活动中赠送，我们尝试制作口号的英文版本。不幸的是，俄语在表达方面有些优雅，而且T恤上的空间有限，所以我们未能提出足够好的翻译（大多数选项要么太长，要么不够准确），并决定即使在为国际活动制作的T恤上也保留俄文口号。这被证明是一个绝妙的决定，因为全世界的人们看到它时都会感到惊喜和好奇。 -So, what does it mean? Here are some ways to translate *“не тормозит”*: +那么，它是什么意思呢？以下是翻译“не тормозит”的一些方式： -- If you translate it literally, it’d be something like *“ClickHouse does not press the brake pedal”*. -- If you’d want to express it as close to how it sounds to a Russian person with IT background, it’d be something like *“If your larger system lags, it’s not because it uses ClickHouse”*. -- Shorter, but not so precise versions could be *“ClickHouse is not slow”*, *“ClickHouse does not lag”* or just *“ClickHouse is fast”*. +- 如果你直译，那就是“ClickHouse 不踩刹车”。 +- 如果你想尽可能接近一个有 IT 背景的俄罗斯人的听觉感受，那就是“如果你的大型系统延迟，不是因为它使用了 ClickHouse”。 +- 更短，但不那么精确的版本可能是“ClickHouse 不慢”，“ClickHouse 不卡顿”或仅仅“ClickHouse 很快”。 -If you haven’t seen one of those t-shirts in person, you can check them out online in many ClickHouse-related videos. For example, this one: +如果您还没有亲眼见过这些 T恤，可以在许多与 ClickHouse 相关的视频中在线查看。例如，这个： ![iframe](https://www.youtube.com/embed/bSyQahMVZ7w) -P.S. These t-shirts are not for sale, they are given away for free on most [ClickHouse Meetups](https://clickhouse.com/#meet), usually for best questions or other forms of active participation. +附言：这些 T恤不出售，它们在大多数 [ClickHouse 聚会](https://clickhouse.com/#meet)上免费赠送，通常是给出最佳问题或其他形式的积极参与者。 diff --git a/docs/zh/faq/general/why-clickhouse-is-so-fast.md b/docs/zh/faq/general/why-clickhouse-is-so-fast.md index a7df6aec207..ddfda87abb4 100644 --- a/docs/zh/faq/general/why-clickhouse-is-so-fast.md +++ b/docs/zh/faq/general/why-clickhouse-is-so-fast.md @@ -1,63 +1,63 @@ --- slug: /zh/faq/general/why-clickhouse-is-so-fast -title: Why is ClickHouse so fast? +title: 为什么 ClickHouse 如此快速？ toc_hidden: true sidebar_position: 8 --- -# Why ClickHouse Is So Fast? {#why-clickhouse-is-so-fast} +# 为什么 ClickHouse 如此快速？ {#why-clickhouse-is-so-fast} -It was designed to be fast. Query execution performance has always been a top priority during the development process, but other important characteristics like user-friendliness, scalability, and security were also considered so ClickHouse could become a real production system. +它被设计成一个快速的系统。在开发过程中，查询执行性能一直是首要考虑的优先级，但也考虑了其他重要特性，如用户友好性、可扩展性和安全性，使 ClickHouse 成为一个真正的生产系统。 -ClickHouse was initially built as a prototype to do just a single task well: to filter and aggregate data as fast as possible. That’s what needs to be done to build a typical analytical report and that’s what a typical [GROUP BY](../../sql-reference/statements/select/group-by.md) query does. ClickHouse team has made several high-level decisions that combined made achieving this task possible: +ClickHouse 最初是作为一个原型构建的，它的单一任务就是尽可能快速地过滤和聚合数据。这正是构建典型分析报告所需做的，也是典型 [GROUP BY](../../sql-reference/statements/select/group-by.md) 查询所做的。ClickHouse 团队做出了几个高层次的决策，这些决策组合在一起使得实现这一任务成为可能： -Column-oriented storage -: Source data often contain hundreds or even thousands of columns, while a report can use just a few of them. The system needs to avoid reading unnecessary columns, or most expensive disk read operations would be wasted. +列式存储 +: 源数据通常包含数百甚至数千列，而报告可能只使用其中的几列。系统需要避免读取不必要的列，否则大部分昂贵的磁盘读取操作将被浪费。 -Indexes -: ClickHouse keeps data structures in memory that allows reading not only used columns but only necessary row ranges of those columns. +索引 +: ClickHouse 在内存中保留数据结构，允许不仅读取使用的列，而且只读取这些列的必要行范围。 -Data compression -: Storing different values of the same column together often leads to better compression ratios (compared to row-oriented systems) because in real data column often has the same or not so many different values for neighboring rows. In addition to general-purpose compression, ClickHouse supports [specialized codecs](../../sql-reference/statements/create/table.mdx/#create-query-specialized-codecs) that can make data even more compact. +数据压缩 +: 将同一列的不同值存储在一起通常会导致更好的压缩比（与行式系统相比），因为在实际数据中列通常对相邻行有相同或不太多的不同值。除了通用压缩之外，ClickHouse 还支持 [专用编解码器](../../sql-reference/statements/create/table.mdx/#create-query-specialized-codecs)，可以使数据更加紧凑。 -Vectorized query execution -: ClickHouse not only stores data in columns but also processes data in columns. It leads to better CPU cache utilization and allows for [SIMD](https://en.wikipedia.org/wiki/SIMD) CPU instructions usage. +向量化查询执行 +: ClickHouse 不仅以列的形式存储数据，而且以列的形式处理数据。这导致更好的 CPU 缓存利用率，并允许使用 [SIMD](https://en.wikipedia.org/wiki/SIMD) CPU 指令。 -Scalability -: ClickHouse can leverage all available CPU cores and disks to execute even a single query. Not only on a single server but all CPU cores and disks of a cluster as well. +可扩展性 +: ClickHouse 可以利用所有可用的 CPU 核心和磁盘来执行甚至是单个查询。不仅在单个服务器上，而且在集群的所有 CPU 核心和磁盘上。 -But many other database management systems use similar techniques. What really makes ClickHouse stand out is **attention to low-level details**. Most programming languages provide implementations for most common algorithms and data structures, but they tend to be too generic to be effective. Every task can be considered as a landscape with various characteristics, instead of just throwing in random implementation. For example, if you need a hash table, here are some key questions to consider: +但许多其他数据库管理系统也使用类似的技术。真正使 ClickHouse 脱颖而出的是 **对底层细节的关注**。大多数编程语言为最常见的算法和数据结构提供了实现，但它们往往过于通用而无法高效。每个任务都可以被视为具有各种特征的景观，而不是仅仅随意投入某个实现。例如，如果您需要一个哈希表，这里有一些关键问题需要考虑： -- Which hash function to choose? -- Collision resolution algorithm: [open addressing](https://en.wikipedia.org/wiki/Open_addressing) vs [chaining](https://en.wikipedia.org/wiki/Hash_table#Separate_chaining)? -- Memory layout: one array for keys and values or separate arrays? Will it store small or large values? -- Fill factor: when and how to resize? How to move values around on resize? -- Will values be removed and which algorithm will work better if they will? -- Will we need fast probing with bitmaps, inline placement of string keys, support for non-movable values, prefetch, and batching? +- 选择哪种哈希函数？ +- 冲突解决算法：[开放寻址](https://en.wikipedia.org/wiki/Open_addressing)还是[链接](https://en.wikipedia.org/wiki/Hash_table#Separate_chaining)？ +- 内存布局：一个数组用于键和值还是分开的数组？它会存储小值还是大值？ +- 填充因子：何时以及如何调整大小？在调整大小时如何移动值？ +- 是否会移除值，如果会，哪种算法会更好？ +- 我们是否需要使用位图进行快速探测，字符串键的内联放置，对不可移动值的支持，预取和批处理？ -Hash table is a key data structure for `GROUP BY` implementation and ClickHouse automatically chooses one of [30+ variations](https://github.com/ClickHouse/ClickHouse/blob/master/src/Interpreters/Aggregator.h) for each specific query. +哈希表是 `GROUP BY` 实现的关键数据结构，ClickHouse 会根据每个特定查询自动选择 [30 多种变体](https://github.com/ClickHouse/ClickHouse/blob/master/src/Interpreters/Aggregator.h) 中的一种。 -The same goes for algorithms, for example, in sorting you might consider: +算法也是如此，例如，在排序中，您可能会考虑： -- What will be sorted: an array of numbers, tuples, strings, or structures? -- Is all data available completely in RAM? -- Do we need a stable sort? -- Do we need a full sort? Maybe partial sort or n-th element will suffice? -- How to implement comparisons? -- Are we sorting data that has already been partially sorted? +- 将要排序的是数字数组、元组、字符串还是结构？ +- 所有数据是否完全可用于 RAM？ +- 我们需要稳定排序吗？ +- 我们需要完全排序吗？也许部分排序或第 n 个元素就足够了？ +- 如何实现比较？ +- 我们正在对已经部分排序的数据进行排序吗？ -Algorithms that they rely on characteristics of data they are working with can often do better than their generic counterparts. If it is not really known in advance, the system can try various implementations and choose the one that works best in runtime. For example, see an [article on how LZ4 decompression is implemented in ClickHouse](https://habr.com/en/company/yandex/blog/457612/). +他们所依赖的算法根据其所处理的数据特性，往往可以比通用算法做得更好。如果事先真的不知道，系统可以尝试各种实现，并在运行时选择最佳的一种。例如，看一篇关于 [ClickHouse 中 LZ4 解压缩是如何实现的文章](https://habr.com/en/company/yandex/blog/457612/)。 -Last but not least, the ClickHouse team always monitors the Internet on people claiming that they came up with the best implementation, algorithm, or data structure to do something and tries it out. Those claims mostly appear to be false, but from time to time you’ll indeed find a gem. +最后但同样重要的是，ClickHouse 团队始终关注互联网上人们声称他们提出了最佳的实现、算法或数据结构来做某事，并尝试它。这些声称大多是虚假的，但有时你确实会找到一颗宝石。 -:::info Tips for building your own high-performance software -- Keep in mind low-level details when designing your system. -- Design based on hardware capabilities. -- Choose data structures and abstractions based on the needs of the task. -- Provide specializations for special cases. -- Try new, “best” algorithms, that you read about yesterday. -- Choose an algorithm in runtime based on statistics. -- Benchmark on real datasets. -- Test for performance regressions in CI. -- Measure and observe everything. +:::info 构建高性能软件的提示 +- 设计系统时要考虑到底层细节。 +- 基于硬件能力进行设计。 +- 根据任务的需求选择数据结构和抽象。 +- 为特殊情况提供专门化。 +- 尝试您昨天阅读的关于新的“最佳”算法。 +- 根据统计数据在运行时选择算法。 +- 在真实数据集上进行基准测试。 +- 在 CI 中测试性能回归。 +- 测量并观察一切。 ::: diff --git a/docs/zh/faq/integration/json-import.md b/docs/zh/faq/integration/json-import.md index 2d5c687316d..730af8cc6da 100644 --- a/docs/zh/faq/integration/json-import.md +++ b/docs/zh/faq/integration/json-import.md @@ -1,35 +1,35 @@ --- slug: /zh/faq/integration/json-import -title: How to import JSON into ClickHouse? +title: 如何将 JSON 导入到 ClickHouse？ toc_hidden: true sidebar_position: 11 --- -# How to Import JSON Into ClickHouse? {#how-to-import-json-into-clickhouse} +# 如何将 JSON 导入到 ClickHouse？ {#how-to-import-json-into-clickhouse} -ClickHouse supports a wide range of [data formats for input and output](../../interfaces/formats.md). There are multiple JSON variations among them, but the most commonly used for data ingestion is [JSONEachRow](../../interfaces/formats.md#jsoneachrow). It expects one JSON object per row, each object separated by a newline. +ClickHouse 支持多种[输入和输出的数据格式](../../interfaces/formats.md)。其中包括多种 JSON 变体，但最常用于数据导入的是 [JSONEachRow](../../interfaces/formats.md#jsoneachrow)。它期望每行一个 JSON 对象，每个对象由一个新行分隔。 -## Examples {#examples} +## 示例 {#examples} -Using [HTTP interface](../../interfaces/http.md): +使用 [HTTP 接口](../../interfaces/http.md)： ``` bash $ echo '{"foo":"bar"}' | curl 'http://localhost:8123/?query=INSERT%20INTO%20test%20FORMAT%20JSONEachRow' --data-binary @- ``` -Using [CLI interface](../../interfaces/cli.md): +使用 [CLI接口](../../interfaces/cli.md): ``` bash $ echo '{"foo":"bar"}' | clickhouse-client --query="INSERT INTO test FORMAT JSONEachRow" ``` -Instead of inserting data manually, you might consider to use one of [client libraries](../../interfaces/index.md) instead. +除了手动插入数据外，您可能会考虑使用 [客户端库](../../interfaces/index.md) 之一。 -## Useful Settings {#useful-settings} +## 实用设置 {#useful-settings} -- `input_format_skip_unknown_fields` allows to insert JSON even if there were additional fields not present in table schema (by discarding them). -- `input_format_import_nested_json` allows to insert nested JSON objects into columns of [Nested](../../sql-reference/data-types/nested-data-structures/nested.md) type. +- `input_format_skip_unknown_fields` 允许插入 JSON，即使存在表格架构中未出现的额外字段（通过丢弃它们）。 +- `input_format_import_nested_json` 允许将嵌套 JSON 对象插入到 [Nested](../../sql-reference/data-types/nested-data-structures/nested.md) 类型的列中。 :::note -Settings are specified as `GET` parameters for the HTTP interface or as additional command-line arguments prefixed with `--` for the `CLI` interface. +对于 HTTP 接口，设置作为 `GET` 参数指定；对于 `CLI` 接口，则作为前缀为 -- 的附加命令行参数。 ::: \ No newline at end of file diff --git a/docs/zh/faq/integration/oracle-odbc.md b/docs/zh/faq/integration/oracle-odbc.md index e22db1d8960..ca65f08686c 100644 --- a/docs/zh/faq/integration/oracle-odbc.md +++ b/docs/zh/faq/integration/oracle-odbc.md @@ -1,16 +1,16 @@ --- slug: /zh/faq/integration/oracle-odbc -title: What if I have a problem with encodings when using Oracle via ODBC? +title: 使用 Oracle ODBC 时遇到编码问题怎么办？ toc_hidden: true sidebar_position: 20 --- -# What If I Have a Problem with Encodings When Using Oracle Via ODBC? {#oracle-odbc-encodings} +# 使用 Oracle ODBC 时遇到编码问题怎么办？ {#oracle-odbc-encodings} -If you use Oracle as a source of ClickHouse external dictionaries via Oracle ODBC driver, you need to set the correct value for the `NLS_LANG` environment variable in `/etc/default/clickhouse`. For more information, see the [Oracle NLS_LANG FAQ](https://www.oracle.com/technetwork/products/globalization/nls-lang-099431.html). +如果您使用 Oracle 作为 ClickHouse 外部字典的数据源，并通过 Oracle ODBC 驱动程序，您需要在 `/etc/default/clickhouse` 中为 `NLS_LANG` 环境变量设置正确的值。更多信息，请参阅 [Oracle NLS_LANG FAQ](https://www.oracle.com/technetwork/products/globalization/nls-lang-099431.html)。 -**Example** +**示例** ``` sql NLS_LANG=RUSSIAN_RUSSIA.UTF8 -``` +``` \ No newline at end of file diff --git a/docs/zh/faq/operations/delete-old-data.md b/docs/zh/faq/operations/delete-old-data.md index 24181116bab..293ba8069fa 100644 --- a/docs/zh/faq/operations/delete-old-data.md +++ b/docs/zh/faq/operations/delete-old-data.md @@ -1,44 +1,44 @@ --- slug: /zh/faq/operations/delete-old-data -title: Is it possible to delete old records from a ClickHouse table? +title: 是否可以从ClickHouse表中删除旧记录？ toc_hidden: true sidebar_position: 20 --- -# Is It Possible to Delete Old Records from a ClickHouse Table? {#is-it-possible-to-delete-old-records-from-a-clickhouse-table} +# 是否可以从ClickHouse表中删除旧记录？ {#is-it-possible-to-delete-old-records-from-a-clickhouse-table} -The short answer is “yes”. ClickHouse has multiple mechanisms that allow freeing up disk space by removing old data. Each mechanism is aimed for different scenarios. +简短的答案是“可以”。ClickHouse具有多种机制，允许通过删除旧数据来释放磁盘空间。每种机制都针对不同的场景。 ## TTL {#ttl} -ClickHouse allows to automatically drop values when some condition happens. This condition is configured as an expression based on any columns, usually just static offset for any timestamp column. +ClickHouse 允许在某些条件发生时自动删除值。这个条件被配置为基于任何列的表达式，通常只是针对任何时间戳列的静态偏移量。 -The key advantage of this approach is that it does not need any external system to trigger, once TTL is configured, data removal happens automatically in background. +这种方法的主要优势是它不需要任何外部系统来触发，一旦配置了 TTL，数据删除就会自动在后台发生。 :::note -TTL can also be used to move data not only to [/dev/null](https://en.wikipedia.org/wiki/Null_device), but also between different storage systems, like from SSD to HDD. +TTL 也可以用来将数据移动到非 [/dev/null](https://en.wikipedia.org/wiki/Null_device) 的不同存储系统，例如从 SSD 到 HDD。 ::: -More details on [configuring TTL](../../engines/table-engines/mergetree-family/mergetree.md#table_engine-mergetree-ttl). +有关 [配置 TTL](../../engines/table-engines/mergetree-family/mergetree.md#table_engine-mergetree-ttl) 的更多详细信息。 ## ALTER DELETE {#alter-delete} -ClickHouse does not have real-time point deletes like in [OLTP](https://en.wikipedia.org/wiki/Online_transaction_processing) databases. The closest thing to them are mutations. They are issued as `ALTER ... DELETE` or `ALTER ... UPDATE` queries to distinguish from normal `DELETE` or `UPDATE` as they are asynchronous batch operations, not immediate modifications. The rest of syntax after `ALTER TABLE` prefix is similar. +ClickHouse没有像[OLTP](https://en.wikipedia.org/wiki/Online_transaction_processing)数据库那样的实时点删除。最接近的东西是 `Mutation`，执行 `ALTER ... DELETE` 或 `ALTER ... UPDATE` 查询，以区别于普通的`DELETE`或`UPDATE`。因为它们是异步批处理操作，而不是立即修改。`ALTER TABLE`前缀后的其余语法相似。 -`ALTER DELETE` can be issued to flexibly remove old data. If you need to do it regularly, the main downside will be the need to have an external system to submit the query. There are also some performance considerations since mutation rewrite complete parts even there’s only a single row to be deleted. +`ALTER DELETE`可以灵活地用来删除旧数据。如果你需要定期这样做，主要缺点将是需要有一个外部系统来提交查询。还有一些性能方面的考虑，因为即使只有一行要被删除，突变也会重写完整部分。 -This is the most common approach to make your system based on ClickHouse [GDPR](https://gdpr-info.eu)-compliant. +这是使基于ClickHouse的系统符合[GDPR](https://gdpr-info.eu)的最常见方法。 -More details on [mutations](../../sql-reference/statements/alter.md/#alter-mutations). +有关[mutations](../../sql-reference/statements/alter.md/#alter-mutations)的更多详细信息。 ## DROP PARTITION {#drop-partition} -`ALTER TABLE ... DROP PARTITION` provides a cost-efficient way to drop a whole partition. It’s not that flexible and needs proper partitioning scheme configured on table creation, but still covers most common cases. Like mutations need to be executed from an external system for regular use. +`ALTER TABLE ... DROP PARTITION`提供了一种成本效率高的方式来删除整个分区。它不是那么灵活，需要在创建表时配置适当的分区方案，但仍然涵盖了大多数常见情况。像 mutations 一样，需要从外部系统执行以进行常规使用。 -More details on [manipulating partitions](../../sql-reference/statements/alter/partition.mdx/#alter_drop-partition). +有关[操作分区](../../sql-reference/statements/alter/partition.mdx/#alter_drop-partition)的更多详细信息。 ## TRUNCATE {#truncate} -It’s rather radical to drop all data from a table, but in some cases it might be exactly what you need. +从表中删除所有数据是相当激进的，但在某些情况下可能正是您所需要的。 -More details on [table truncation](../../sql-reference/statements/truncate.md). \ No newline at end of file +有关[truncate](../../sql-reference/statements/truncate.md)的更多详细信息。 diff --git a/programs/local/LocalServer.cpp b/programs/local/LocalServer.cpp index c9841277b6d..4e0b9eeb731 100644 --- a/programs/local/LocalServer.cpp +++ b/programs/local/LocalServer.cpp @@ -290,6 +290,11 @@ void LocalServer::cleanup() { connection.reset(); + /// Suggestions are loaded async in a separate thread and it can use global context. + /// We should reset it before resetting global_context. + if (suggest) + suggest.reset(); + if (global_context) { global_context->shutdown(); diff --git a/programs/server/Server.cpp b/programs/server/Server.cpp index a996ed6e34c..8281afbf320 100644 --- a/programs/server/Server.cpp +++ b/programs/server/Server.cpp @@ -76,8 +76,8 @@ #include #include #include -#include -#include +#include +#include #include #include #include "MetricsTransmitter.h" @@ -1467,6 +1467,8 @@ try global_context->reloadAuxiliaryZooKeepersConfigIfChanged(config); + global_context->reloadQueryMaskingRulesIfChanged(config); + std::lock_guard lock(servers_lock); updateServers(*config, server_pool, async_metrics, servers, servers_to_start_before_tables); } @@ -2002,6 +2004,12 @@ try else LOG_INFO(log, "Closed all listening sockets."); + /// Wait for unfinished backups and restores. + /// This must be done after closing listening sockets (no more backups/restores) but before ProcessList::killAllQueries + /// (because killAllQueries() will cancel all running backups/restores). + if (server_settings.shutdown_wait_backups_and_restores) + global_context->waitAllBackupsAndRestores(); + /// Killing remaining queries. if (!server_settings.shutdown_wait_unfinished_queries) global_context->getProcessList().killAllQueries(); diff --git a/programs/server/config.xml b/programs/server/config.xml index e1428b17084..0dc271692b8 100644 --- a/programs/server/config.xml +++ b/programs/server/config.xml @@ -440,6 +440,9 @@ 10000 + + /var/lib/clickhouse/caches/ + false diff --git a/src/Access/ExternalAuthenticators.cpp b/src/Access/ExternalAuthenticators.cpp index 981a7aafc6f..351bcb95c73 100644 --- a/src/Access/ExternalAuthenticators.cpp +++ b/src/Access/ExternalAuthenticators.cpp @@ -242,7 +242,10 @@ HTTPAuthClientParams parseHTTPAuthParams(const Poco::Util::AbstractConfiguration size_t connection_timeout_ms = config.getInt(prefix + ".connection_timeout_ms", 1000); size_t receive_timeout_ms = config.getInt(prefix + ".receive_timeout_ms", 1000); size_t send_timeout_ms = config.getInt(prefix + ".send_timeout_ms", 1000); - http_auth_params.timeouts = ConnectionTimeouts{connection_timeout_ms, receive_timeout_ms, send_timeout_ms}; + http_auth_params.timeouts = ConnectionTimeouts() + .withConnectionTimeout(Poco::Timespan(connection_timeout_ms * 1000)) + .withReceiveTimeout(Poco::Timespan(receive_timeout_ms * 1000)) + .withSendTimeout(Poco::Timespan(send_timeout_ms * 1000)); http_auth_params.max_tries = config.getInt(prefix + ".max_tries", 3); http_auth_params.retry_initial_backoff_ms = config.getInt(prefix + ".retry_initial_backoff_ms", 50); diff --git a/src/AggregateFunctions/AggregateFunctionQuantile.h b/src/AggregateFunctions/AggregateFunctionQuantile.h index 21e174a9e7f..5a0509a5b8d 100644 --- a/src/AggregateFunctions/AggregateFunctionQuantile.h +++ b/src/AggregateFunctions/AggregateFunctionQuantile.h @@ -31,7 +31,7 @@ namespace ErrorCodes template class QuantileTiming; template class QuantileGK; - +template class QuantileDDSketch; /** Generic aggregate function for calculation of quantiles. * It depends on quantile calculation data structure. Look at Quantile*.h for various implementations. @@ -64,6 +64,7 @@ private: using ColVecType = ColumnVectorOrDecimal; static constexpr bool returns_float = !(std::is_same_v); + static constexpr bool is_quantile_ddsketch = std::is_same_v>; static_assert(!is_decimal || !returns_float); QuantileLevels levels; @@ -74,6 +75,9 @@ private: /// Used for the approximate version of the algorithm (Greenwald-Khanna) ssize_t accuracy = 10000; + /// Used for the quantile sketch + Float64 relative_accuracy = 0.01; + DataTypePtr & argument_type; public: @@ -87,7 +91,36 @@ public: if (!returns_many && levels.size() > 1) throw Exception(ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Aggregate function {} requires one level parameter or less", getName()); - if constexpr (has_accuracy_parameter) + if constexpr (is_quantile_ddsketch) + { + if (params.empty()) + throw Exception( + ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH, "Aggregate function {} requires at least one param", getName()); + + const auto & relative_accuracy_field = params[0]; + if (relative_accuracy_field.getType() != Field::Types::Float64) + throw Exception( + ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT, "Aggregate function {} requires relative accuracy parameter with Float64 type", getName()); + + relative_accuracy = relative_accuracy_field.get(); + + if (relative_accuracy <= 0 || relative_accuracy >= 1 || isNaN(relative_accuracy)) + throw Exception( + ErrorCodes::BAD_ARGUMENTS, + "Aggregate function {} requires relative accuracy parameter with value between 0 and 1 but is {}", + getName(), + relative_accuracy); + // Throw exception if the relative accuracy is too small. + // This is to avoid the case where the user specifies a relative accuracy that is too small + // and the sketch is not able to allocate enough memory to satisfy the accuracy requirement. + if (relative_accuracy < 1e-6) + throw Exception( + ErrorCodes::BAD_ARGUMENTS, + "Aggregate function {} requires relative accuracy parameter with value greater than 1e-6 but is {}", + getName(), + relative_accuracy); + } + else if constexpr (has_accuracy_parameter) { if (params.empty()) throw Exception( @@ -116,7 +149,9 @@ public: void create(AggregateDataPtr __restrict place) const override /// NOLINT { - if constexpr (has_accuracy_parameter) + if constexpr (is_quantile_ddsketch) + new (place) Data(relative_accuracy); + else if constexpr (has_accuracy_parameter) new (place) Data(accuracy); else new (place) Data; @@ -147,6 +182,10 @@ public: { /// Return normalized state type: quantiles*(1)(...) Array params{1}; + if constexpr (is_quantile_ddsketch) + params = {relative_accuracy, 1}; + else if constexpr (has_accuracy_parameter) + params = {accuracy, 1}; AggregateFunctionProperties properties; return std::make_shared( AggregateFunctionFactory::instance().get( @@ -295,4 +334,7 @@ struct NameQuantilesBFloat16Weighted { static constexpr auto name = "quantilesBF struct NameQuantileGK { static constexpr auto name = "quantileGK"; }; struct NameQuantilesGK { static constexpr auto name = "quantilesGK"; }; +struct NameQuantileDDSketch { static constexpr auto name = "quantileDDSketch"; }; +struct NameQuantilesDDSketch { static constexpr auto name = "quantilesDDSketch"; }; + } diff --git a/src/AggregateFunctions/AggregateFunctionQuantileDDSketch.cpp b/src/AggregateFunctions/AggregateFunctionQuantileDDSketch.cpp new file mode 100644 index 00000000000..0c5c992920c --- /dev/null +++ b/src/AggregateFunctions/AggregateFunctionQuantileDDSketch.cpp @@ -0,0 +1,61 @@ +#include +#include +#include +#include +#include +#include +#include + + +namespace DB +{ +struct Settings; + +namespace ErrorCodes +{ + extern const int ILLEGAL_TYPE_OF_ARGUMENT; +} + +namespace +{ + +template using FuncQuantileDDSketch = AggregateFunctionQuantile, NameQuantileDDSketch, false, std::conditional_t, false, true>; +template using FuncQuantilesDDSketch = AggregateFunctionQuantile, NameQuantilesDDSketch, false, std::conditional_t, true, true>; + + +template