Merge branch 'ClickHouse:master' into map-combinator-for-non-map-columns

This commit is contained in:
UnamedRus 2024-06-24 00:20:02 +03:00 committed by GitHub
commit 5f31b3cf59
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
436 changed files with 11460 additions and 4271 deletions

View File

@ -48,9 +48,7 @@ At a minimum, the following information should be added (but add more as needed)
- [ ] <!---ci_include_stateful--> Allow: Stateful tests
- [ ] <!---ci_include_integration--> Allow: Integration Tests
- [ ] <!---ci_include_performance--> Allow: Performance tests
- [ ] <!---ci_set_normal_builds--> Allow: Normal Builds
- [ ] <!---ci_set_special_builds--> Allow: Special Builds
- [ ] <!---ci_set_non_required--> Allow: All NOT Required Checks
- [ ] <!---ci_set_builds--> Allow: All Builds
- [ ] <!---batch_0_1--> Allow: batch 1, 2 for multi-batch jobs
- [ ] <!---batch_2_3--> Allow: batch 3, 4, 5, 6 for multi-batch jobs
---
@ -61,6 +59,7 @@ At a minimum, the following information should be added (but add more as needed)
- [ ] <!---ci_exclude_aarch64|release|debug--> Exclude: All with aarch64, release, debug
---
- [ ] <!---do_not_test--> Do not test
- [ ] <!---woolen_wolfdog--> Woolen Wolfdog
- [ ] <!---upload_all--> Upload binaries for special builds
- [ ] <!---no_merge_commit--> Disable merge-commit
- [ ] <!---no_ci_cache--> Disable CI cache

View File

@ -70,7 +70,7 @@ jobs:
if: ${{ !failure() && !cancelled() }}
uses: ./.github/workflows/reusable_test.yml
with:
test_name: Compatibility check (amd64)
test_name: Compatibility check (release)
runner_type: style-checker
data: ${{ needs.RunConfig.outputs.data }}
CompatibilityCheckAarch64:
@ -194,7 +194,7 @@ jobs:
if: ${{ !failure() && !cancelled() }}
uses: ./.github/workflows/reusable_test.yml
with:
test_name: Install packages (amd64)
test_name: Install packages (release)
runner_type: style-checker
data: ${{ needs.RunConfig.outputs.data }}
run_command: |
@ -204,7 +204,7 @@ jobs:
if: ${{ !failure() && !cancelled() }}
uses: ./.github/workflows/reusable_test.yml
with:
test_name: Install packages (arm64)
test_name: Install packages (aarch64)
runner_type: style-checker-aarch64
data: ${{ needs.RunConfig.outputs.data }}
run_command: |

View File

@ -104,10 +104,9 @@ jobs:
with:
stage: Tests_2
data: ${{ needs.RunConfig.outputs.data }}
# stage for jobs that do not prohibit merge
Tests_3:
# Test_3 should not wait for Test_1/Test_2 and should not be blocked by them on master branch since all jobs need to run there.
needs: [RunConfig, Builds_1, Builds_2]
needs: [RunConfig, Builds_1]
if: ${{ !failure() && !cancelled() && contains(fromJson(needs.RunConfig.outputs.data).stages_data.stages_to_do, 'Tests_3') }}
uses: ./.github/workflows/reusable_test_stage.yml
with:
@ -115,25 +114,16 @@ jobs:
data: ${{ needs.RunConfig.outputs.data }}
################################# Reports #################################
# Reports should be run even if Builds_1/2 failed - put them separately in wf (not in Tests_1/2)
Builds_1_Report:
# Reports should run even if Builds_1/2 fail - run them separately, not in Tests_1/2/3
Builds_Report:
# run report check for failed builds to indicate the CI error
if: ${{ !cancelled() && needs.RunConfig.result == 'success' && contains(fromJson(needs.RunConfig.outputs.data).jobs_data.jobs_to_do, 'ClickHouse build check') }}
needs: [RunConfig, Builds_1]
needs: [RunConfig, Builds_1, Builds_2]
uses: ./.github/workflows/reusable_test.yml
with:
test_name: ClickHouse build check
runner_type: style-checker-aarch64
data: ${{ needs.RunConfig.outputs.data }}
Builds_2_Report:
# run report check for failed builds to indicate the CI error
if: ${{ !cancelled() && needs.RunConfig.result == 'success' && contains(fromJson(needs.RunConfig.outputs.data).jobs_data.jobs_to_do, 'ClickHouse special build check') }}
needs: [RunConfig, Builds_2]
uses: ./.github/workflows/reusable_test.yml
with:
test_name: ClickHouse special build check
runner_type: style-checker-aarch64
data: ${{ needs.RunConfig.outputs.data }}
MarkReleaseReady:
if: ${{ !failure() && !cancelled() }}
@ -165,7 +155,7 @@ jobs:
FinishCheck:
if: ${{ !cancelled() }}
needs: [RunConfig, Builds_1, Builds_2, Builds_1_Report, Builds_2_Report, Tests_1, Tests_2, Tests_3]
needs: [RunConfig, Builds_1, Builds_2, Builds_Report, Tests_1, Tests_2, Tests_3]
runs-on: [self-hosted, style-checker-aarch64]
steps:
- name: Check out repository code

View File

@ -126,16 +126,16 @@ jobs:
with:
stage: Builds_2
data: ${{ needs.RunConfig.outputs.data }}
# stage for running non-required checks without being blocked by required checks (Test_1) if corresponding settings is selected
Tests_2:
needs: [RunConfig, Builds_2]
needs: [RunConfig, Builds_1]
if: ${{ !failure() && !cancelled() && contains(fromJson(needs.RunConfig.outputs.data).stages_data.stages_to_do, 'Tests_2') }}
uses: ./.github/workflows/reusable_test_stage.yml
with:
stage: Tests_2
data: ${{ needs.RunConfig.outputs.data }}
# stage for jobs that do not prohibit merge
Tests_3:
needs: [RunConfig, Builds_1, Tests_1, Builds_2, Tests_2]
needs: [RunConfig, Builds_1, Tests_1]
if: ${{ !failure() && !cancelled() && contains(fromJson(needs.RunConfig.outputs.data).stages_data.stages_to_do, 'Tests_3') }}
uses: ./.github/workflows/reusable_test_stage.yml
with:
@ -143,29 +143,21 @@ jobs:
data: ${{ needs.RunConfig.outputs.data }}
################################# Reports #################################
# Reports should by run even if Builds_1/2 fail, so put them separately in wf (not in Tests_1/2)
Builds_1_Report:
# Reports should run even if Builds_1/2 fail - run them separately (not in Tests_1/2/3)
Builds_Report:
# run report check for failed builds to indicate the CI error
if: ${{ !cancelled() && needs.StyleCheck.result == 'success' && contains(fromJson(needs.RunConfig.outputs.data).jobs_data.jobs_to_do, 'ClickHouse build check') }}
needs: [RunConfig, StyleCheck, Builds_1]
if: ${{ !cancelled() && needs.RunConfig.result == 'success' && contains(fromJson(needs.RunConfig.outputs.data).jobs_data.jobs_to_do, 'ClickHouse build check') }}
needs: [RunConfig, StyleCheck, Builds_1, Builds_2]
uses: ./.github/workflows/reusable_test.yml
with:
test_name: ClickHouse build check
runner_type: style-checker-aarch64
data: ${{ needs.RunConfig.outputs.data }}
Builds_2_Report:
# run report check for failed builds to indicate the CI error
if: ${{ !cancelled() && needs.StyleCheck.result == 'success' && contains(fromJson(needs.RunConfig.outputs.data).jobs_data.jobs_to_do, 'ClickHouse special build check') }}
needs: [RunConfig, StyleCheck, Builds_2]
uses: ./.github/workflows/reusable_test.yml
with:
test_name: ClickHouse special build check
runner_type: style-checker-aarch64
data: ${{ needs.RunConfig.outputs.data }}
CheckReadyForMerge:
if: ${{ !cancelled() && needs.StyleCheck.result == 'success' }}
needs: [RunConfig, BuildDockers, StyleCheck, FastTest, Builds_1, Builds_2, Builds_1_Report, Builds_2_Report, Tests_1, Tests_2]
# Test_2 or Test_3 must not have jobs required for Mergeable check
needs: [RunConfig, BuildDockers, StyleCheck, FastTest, Builds_1, Builds_2, Builds_Report, Tests_1]
runs-on: [self-hosted, style-checker-aarch64]
steps:
- name: Check out repository code
@ -181,7 +173,7 @@ jobs:
#
FinishCheck:
if: ${{ !cancelled() }}
needs: [RunConfig, BuildDockers, StyleCheck, FastTest, Builds_1, Builds_2, Builds_1_Report, Builds_2_Report, Tests_1, Tests_2, Tests_3]
needs: [RunConfig, BuildDockers, StyleCheck, FastTest, Builds_1, Builds_2, Builds_Report, Tests_1, Tests_2, Tests_3]
runs-on: [self-hosted, style-checker-aarch64]
steps:
- name: Check out repository code
@ -204,8 +196,7 @@ jobs:
concurrency:
group: jepsen
if: ${{ !failure() && !cancelled() && contains(fromJson(needs.RunConfig.outputs.data).jobs_data.jobs_to_do, 'ClickHouse Keeper Jepsen') }}
# jepsen needs binary_release build which is in Builds_2
needs: [RunConfig, Builds_2]
needs: [RunConfig, Builds_1]
uses: ./.github/workflows/reusable_test.yml
with:
test_name: ClickHouse Keeper Jepsen

View File

@ -65,7 +65,7 @@ jobs:
if: ${{ !failure() && !cancelled() }}
uses: ./.github/workflows/reusable_test.yml
with:
test_name: Compatibility check (amd64)
test_name: Compatibility check (release)
runner_type: style-checker
data: ${{ needs.RunConfig.outputs.data }}
CompatibilityCheckAarch64:
@ -244,7 +244,7 @@ jobs:
if: ${{ !failure() && !cancelled() }}
uses: ./.github/workflows/reusable_test.yml
with:
test_name: Install packages (amd64)
test_name: Install packages (release)
runner_type: style-checker
data: ${{ needs.RunConfig.outputs.data }}
run_command: |
@ -254,7 +254,7 @@ jobs:
if: ${{ !failure() && !cancelled() }}
uses: ./.github/workflows/reusable_test.yml
with:
test_name: Install packages (arm64)
test_name: Install packages (aarch64)
runner_type: style-checker-aarch64
data: ${{ needs.RunConfig.outputs.data }}
run_command: |

View File

@ -34,20 +34,18 @@ curl https://clickhouse.com/ | sh
Every month we get together with the community (users, contributors, customers, those interested in learning more about ClickHouse) to discuss what is coming in the latest release. If you are interested in sharing what you've built on ClickHouse, let us know.
* [v24.5 Community Call](https://clickhouse.com/company/events/v24-5-community-release-call) - May 30
* [v24.6 Community Call](https://clickhouse.com/company/events/v24-6-community-release-call) - Jul 2
## Upcoming Events
Keep an eye out for upcoming meetups and events around the world. Somewhere else you want us to be? Please feel free to reach out to tyler `<at>` clickhouse `<dot>` com. You can also peruse [ClickHouse Events](https://clickhouse.com/company/news-events) for a list of all upcoming trainings, meetups, speaking engagements, etc.
* [ClickHouse Happy Hour @ Tom's Watch Bar - Los Angeles](https://www.meetup.com/clickhouse-los-angeles-user-group/events/300740584/) - May 22
* [ClickHouse & Confluent Meetup in Dubai](https://www.meetup.com/clickhouse-dubai-meetup-group/events/299629189/) - May 28
* [ClickHouse Meetup in Stockholm](https://www.meetup.com/clickhouse-stockholm-user-group/events/299752651/) - Jun 3
* [ClickHouse Meetup @ Cloudflare - San Francisco](https://www.meetup.com/clickhouse-silicon-valley-meetup-group/events/300523061/) - Jun 4
* [ClickHouse (クリックハウス) Meetup Tokyo](https://www.meetup.com/clickhouse-tokyo-user-group/events/300798053/) - Jun 5
* [AWS Summit in DC](https://clickhouse.com/company/events/2024-06-aws-summit-dc) - Jun 26
* [ClickHouse Meetup in Amsterdam](https://www.meetup.com/clickhouse-netherlands-user-group/events/300781068/) - Jun 27
* [ClickHouse Meetup in Paris](https://www.meetup.com/clickhouse-france-user-group/events/300783448/) - Jul 9
* [ClickHouse Cloud - Live Update Call](https://clickhouse.com/company/events/202407-cloud-update-live) - Jul 9
* [ClickHouse Meetup @ Ramp - New York City](https://www.meetup.com/clickhouse-new-york-user-group/events/300595845/) - Jul 9
* [AWS Summit in New York](https://clickhouse.com/company/events/2024-07-awssummit-nyc) - Jul 10
* [ClickHouse Meetup @ Klaviyo - Boston](https://www.meetup.com/clickhouse-boston-user-group/events/300907870) - Jul 11
## Recent Recordings

View File

@ -23,9 +23,6 @@
#include <openssl/conf.h>
#endif
#if __has_feature(address_sanitizer)
#include <sanitizer/lsan_interface.h>
#endif
using Poco::RandomInputStream;
using Poco::Thread;
@ -70,18 +67,12 @@ void OpenSSLInitializer::initialize()
SSL_library_init();
SSL_load_error_strings();
OpenSSL_add_all_algorithms();
char seed[SEEDSIZE];
RandomInputStream rnd;
rnd.read(seed, sizeof(seed));
{
# if __has_feature(address_sanitizer)
/// Leak sanitizer (part of address sanitizer) thinks that a few bytes of memory in OpenSSL are allocated during but never released.
__lsan::ScopedDisabler lsan_disabler;
#endif
RAND_seed(seed, SEEDSIZE);
}
RAND_seed(seed, SEEDSIZE);
int nMutexes = CRYPTO_num_locks();
_mutexes = new Poco::FastMutex[nMutexes];
CRYPTO_set_locking_callback(&OpenSSLInitializer::lock);
@ -89,8 +80,8 @@ void OpenSSLInitializer::initialize()
// https://sourceforge.net/p/poco/bugs/110/
//
// From http://www.openssl.org/docs/crypto/threads.html :
// "If the application does not register such a callback using CRYPTO_THREADID_set_callback(),
// then a default implementation is used - on Windows and BeOS this uses the system's
// "If the application does not register such a callback using CRYPTO_THREADID_set_callback(),
// then a default implementation is used - on Windows and BeOS this uses the system's
// default thread identifying APIs"
CRYPTO_set_id_callback(&OpenSSLInitializer::id);
CRYPTO_set_dynlock_create_callback(&OpenSSLInitializer::dynlockCreate);
@ -109,7 +100,7 @@ void OpenSSLInitializer::uninitialize()
CRYPTO_set_locking_callback(0);
CRYPTO_set_id_callback(0);
delete [] _mutexes;
CONF_modules_free();
}
}

View File

@ -213,6 +213,7 @@ target_compile_definitions (_poco_foundation
)
target_include_directories (_poco_foundation SYSTEM PUBLIC "include")
target_link_libraries (_poco_foundation PRIVATE clickhouse_common_io)
target_link_libraries (_poco_foundation
PRIVATE

View File

@ -48,7 +48,13 @@ class Foundation_API ThreadPool
/// from the pool.
{
public:
ThreadPool(int minCapacity = 2, int maxCapacity = 16, int idleTime = 60, int stackSize = POCO_THREAD_STACK_SIZE);
explicit ThreadPool(
int minCapacity = 2,
int maxCapacity = 16,
int idleTime = 60,
int stackSize = POCO_THREAD_STACK_SIZE,
size_t global_profiler_real_time_period_ns_ = 0,
size_t global_profiler_cpu_time_period_ns_ = 0);
/// Creates a thread pool with minCapacity threads.
/// If required, up to maxCapacity threads are created
/// a NoThreadAvailableException exception is thrown.
@ -56,8 +62,14 @@ public:
/// and more than minCapacity threads are running, the thread
/// is killed. Threads are created with given stack size.
ThreadPool(
const std::string & name, int minCapacity = 2, int maxCapacity = 16, int idleTime = 60, int stackSize = POCO_THREAD_STACK_SIZE);
explicit ThreadPool(
const std::string & name,
int minCapacity = 2,
int maxCapacity = 16,
int idleTime = 60,
int stackSize = POCO_THREAD_STACK_SIZE,
size_t global_profiler_real_time_period_ns_ = 0,
size_t global_profiler_cpu_time_period_ns_ = 0);
/// Creates a thread pool with the given name and minCapacity threads.
/// If required, up to maxCapacity threads are created
/// a NoThreadAvailableException exception is thrown.
@ -171,6 +183,8 @@ private:
int _serial;
int _age;
int _stackSize;
size_t _globalProfilerRealTimePeriodNs;
size_t _globalProfilerCPUTimePeriodNs;
ThreadVec _threads;
mutable FastMutex _mutex;
};

View File

@ -20,6 +20,7 @@
#include "Poco/ErrorHandler.h"
#include <sstream>
#include <ctime>
#include <Common/ThreadPool.h>
namespace Poco {
@ -28,7 +29,11 @@ namespace Poco {
class PooledThread: public Runnable
{
public:
PooledThread(const std::string& name, int stackSize = POCO_THREAD_STACK_SIZE);
explicit PooledThread(
const std::string& name,
int stackSize = POCO_THREAD_STACK_SIZE,
size_t globalProfilerRealTimePeriodNs_ = 0,
size_t globalProfilerCPUTimePeriodNs_ = 0);
~PooledThread();
void start();
@ -51,16 +56,24 @@ private:
Event _targetCompleted;
Event _started;
FastMutex _mutex;
size_t _globalProfilerRealTimePeriodNs;
size_t _globalProfilerCPUTimePeriodNs;
};
PooledThread::PooledThread(const std::string& name, int stackSize):
_idle(true),
_idleTime(0),
_pTarget(0),
_name(name),
PooledThread::PooledThread(
const std::string& name,
int stackSize,
size_t globalProfilerRealTimePeriodNs_,
size_t globalProfilerCPUTimePeriodNs_) :
_idle(true),
_idleTime(0),
_pTarget(0),
_name(name),
_thread(name),
_targetCompleted(false)
_targetCompleted(false),
_globalProfilerRealTimePeriodNs(globalProfilerRealTimePeriodNs_),
_globalProfilerCPUTimePeriodNs(globalProfilerCPUTimePeriodNs_)
{
poco_assert_dbg (stackSize >= 0);
_thread.setStackSize(stackSize);
@ -83,7 +96,7 @@ void PooledThread::start()
void PooledThread::start(Thread::Priority priority, Runnable& target)
{
FastMutex::ScopedLock lock(_mutex);
poco_assert (_pTarget == 0);
_pTarget = &target;
@ -109,7 +122,7 @@ void PooledThread::start(Thread::Priority priority, Runnable& target, const std:
}
_thread.setName(fullName);
_thread.setPriority(priority);
poco_assert (_pTarget == 0);
_pTarget = &target;
@ -145,7 +158,7 @@ void PooledThread::join()
void PooledThread::activate()
{
FastMutex::ScopedLock lock(_mutex);
poco_assert (_idle);
_idle = false;
_targetCompleted.reset();
@ -155,7 +168,7 @@ void PooledThread::activate()
void PooledThread::release()
{
const long JOIN_TIMEOUT = 10000;
_mutex.lock();
_pTarget = 0;
_mutex.unlock();
@ -174,6 +187,10 @@ void PooledThread::release()
void PooledThread::run()
{
DB::ThreadStatus thread_status;
if (unlikely(_globalProfilerRealTimePeriodNs != 0 || _globalProfilerCPUTimePeriodNs != 0))
thread_status.initGlobalProfiler(_globalProfilerRealTimePeriodNs, _globalProfilerCPUTimePeriodNs);
_started.set();
for (;;)
{
@ -220,13 +237,17 @@ void PooledThread::run()
ThreadPool::ThreadPool(int minCapacity,
int maxCapacity,
int idleTime,
int stackSize):
_minCapacity(minCapacity),
_maxCapacity(maxCapacity),
int stackSize,
size_t globalProfilerRealTimePeriodNs_,
size_t globalProfilerCPUTimePeriodNs_) :
_minCapacity(minCapacity),
_maxCapacity(maxCapacity),
_idleTime(idleTime),
_serial(0),
_age(0),
_stackSize(stackSize)
_stackSize(stackSize),
_globalProfilerRealTimePeriodNs(globalProfilerRealTimePeriodNs_),
_globalProfilerCPUTimePeriodNs(globalProfilerCPUTimePeriodNs_)
{
poco_assert (minCapacity >= 1 && maxCapacity >= minCapacity && idleTime > 0);
@ -243,14 +264,18 @@ ThreadPool::ThreadPool(const std::string& name,
int minCapacity,
int maxCapacity,
int idleTime,
int stackSize):
int stackSize,
size_t globalProfilerRealTimePeriodNs_,
size_t globalProfilerCPUTimePeriodNs_) :
_name(name),
_minCapacity(minCapacity),
_maxCapacity(maxCapacity),
_minCapacity(minCapacity),
_maxCapacity(maxCapacity),
_idleTime(idleTime),
_serial(0),
_age(0),
_stackSize(stackSize)
_stackSize(stackSize),
_globalProfilerRealTimePeriodNs(globalProfilerRealTimePeriodNs_),
_globalProfilerCPUTimePeriodNs(globalProfilerCPUTimePeriodNs_)
{
poco_assert (minCapacity >= 1 && maxCapacity >= minCapacity && idleTime > 0);
@ -393,15 +418,15 @@ void ThreadPool::housekeep()
ThreadVec activeThreads;
idleThreads.reserve(_threads.size());
activeThreads.reserve(_threads.size());
for (ThreadVec::iterator it = _threads.begin(); it != _threads.end(); ++it)
{
if ((*it)->idle())
{
if ((*it)->idleTime() < _idleTime)
idleThreads.push_back(*it);
else
expiredThreads.push_back(*it);
else
expiredThreads.push_back(*it);
}
else activeThreads.push_back(*it);
}
@ -463,7 +488,7 @@ PooledThread* ThreadPool::createThread()
{
std::ostringstream name;
name << _name << "[#" << ++_serial << "]";
return new PooledThread(name.str(), _stackSize);
return new PooledThread(name.str(), _stackSize, _globalProfilerRealTimePeriodNs, _globalProfilerCPUTimePeriodNs);
}
@ -481,7 +506,7 @@ public:
ThreadPool* pool()
{
FastMutex::ScopedLock lock(_mutex);
if (!_pPool)
{
_pPool = new ThreadPool("default");
@ -490,7 +515,7 @@ public:
}
return _pPool;
}
private:
ThreadPool* _pPool;
FastMutex _mutex;

View File

@ -330,27 +330,26 @@ void SSLManager::initDefaultContext(bool server)
else
_ptrDefaultClientContext->disableProtocols(disabledProtocols);
/// Temporarily disabled during the transition from boringssl to OpenSSL due to tsan issues.
/// bool cacheSessions = config.getBool(prefix + CFG_CACHE_SESSIONS, false);
/// if (server)
/// {
/// std::string sessionIdContext = config.getString(prefix + CFG_SESSION_ID_CONTEXT, config.getString("application.name", ""));
/// _ptrDefaultServerContext->enableSessionCache(cacheSessions, sessionIdContext);
/// if (config.hasProperty(prefix + CFG_SESSION_CACHE_SIZE))
/// {
/// int cacheSize = config.getInt(prefix + CFG_SESSION_CACHE_SIZE);
/// _ptrDefaultServerContext->setSessionCacheSize(cacheSize);
/// }
/// if (config.hasProperty(prefix + CFG_SESSION_TIMEOUT))
/// {
/// int timeout = config.getInt(prefix + CFG_SESSION_TIMEOUT);
/// _ptrDefaultServerContext->setSessionTimeout(timeout);
/// }
/// }
/// else
/// {
/// _ptrDefaultClientContext->enableSessionCache(cacheSessions);
/// }
bool cacheSessions = config.getBool(prefix + CFG_CACHE_SESSIONS, false);
if (server)
{
std::string sessionIdContext = config.getString(prefix + CFG_SESSION_ID_CONTEXT, config.getString("application.name", ""));
_ptrDefaultServerContext->enableSessionCache(cacheSessions, sessionIdContext);
if (config.hasProperty(prefix + CFG_SESSION_CACHE_SIZE))
{
int cacheSize = config.getInt(prefix + CFG_SESSION_CACHE_SIZE);
_ptrDefaultServerContext->setSessionCacheSize(cacheSize);
}
if (config.hasProperty(prefix + CFG_SESSION_TIMEOUT))
{
int timeout = config.getInt(prefix + CFG_SESSION_TIMEOUT);
_ptrDefaultServerContext->setSessionTimeout(timeout);
}
}
else
{
_ptrDefaultClientContext->enableSessionCache(cacheSessions);
}
bool extendedVerification = config.getBool(prefix + CFG_EXTENDED_VERIFICATION, false);
if (server)
_ptrDefaultServerContext->enableExtendedCertificateVerification(extendedVerification);

View File

@ -1,12 +1,12 @@
# This variables autochanged by tests/ci/version_helper.py:
# NOTE: has nothing common with DBMS_TCP_PROTOCOL_VERSION,
# NOTE: VERSION_REVISION has nothing common with DBMS_TCP_PROTOCOL_VERSION,
# only DBMS_TCP_PROTOCOL_VERSION should be incremented on protocol changes.
SET(VERSION_REVISION 54487)
SET(VERSION_REVISION 54488)
SET(VERSION_MAJOR 24)
SET(VERSION_MINOR 6)
SET(VERSION_MINOR 7)
SET(VERSION_PATCH 1)
SET(VERSION_GITHASH 70a1d3a63d47f0be077d67b8deb907230fc7cfb0)
SET(VERSION_DESCRIBE v24.6.1.1-testing)
SET(VERSION_STRING 24.6.1.1)
SET(VERSION_GITHASH aa023477a9265e403982fca5ee29a714db5133d9)
SET(VERSION_DESCRIBE v24.7.1.1-testing)
SET(VERSION_STRING 24.7.1.1)
# end of autochange

2
contrib/openssl vendored

@ -1 +1 @@
Subproject commit 67c0b63e578e4c751ac9edf490f5a96124fff8dc
Subproject commit 5d81fa7068fc8c07f4d0997d5b703f3c541a637c

2
contrib/re2 vendored

@ -1 +1 @@
Subproject commit a807e8a3aac2cc33c77b7071efea54fcabe38e0c
Subproject commit 85dd7ad833a73095ecf3e3baea608ba051bbe2c7

View File

@ -28,16 +28,20 @@ set(RE2_SOURCES
add_library(_re2 ${RE2_SOURCES})
target_include_directories(_re2 PUBLIC "${SRC_DIR}")
target_link_libraries(_re2 PRIVATE
absl::absl_check
absl::absl_log
absl::base
absl::core_headers
absl::fixed_array
absl::flags
absl::flat_hash_map
absl::flat_hash_set
absl::hash
absl::inlined_vector
absl::strings
absl::str_format
absl::synchronization
absl::optional
absl::span)
absl::span
absl::str_format
absl::strings
absl::synchronization)
add_library(ch_contrib::re2 ALIAS _re2)

View File

@ -254,7 +254,7 @@ function run_tests()
set +e
clickhouse-test --testname --shard --zookeeper --check-zookeeper-session --hung-check --print-time \
--test-runs "$NUM_TRIES" "${ADDITIONAL_OPTIONS[@]}" 2>&1 \
--no-drop-if-fail --test-runs "$NUM_TRIES" "${ADDITIONAL_OPTIONS[@]}" 2>&1 \
| ts '%Y-%m-%d %H:%M:%S' \
| tee -a test_output/test_result.txt
set -e
@ -285,7 +285,7 @@ stop_logs_replication
# Try to get logs while server is running
failed_to_save_logs=0
for table in query_log zookeeper_log trace_log transactions_info_log metric_log blob_storage_log
for table in query_log zookeeper_log trace_log transactions_info_log metric_log blob_storage_log error_log
do
err=$(clickhouse-client -q "select * from system.$table into outfile '/test_output/$table.tsv.gz' format TSVWithNamesAndTypes")
echo "$err"
@ -339,7 +339,7 @@ if [ $failed_to_save_logs -ne 0 ]; then
# directly
# - even though ci auto-compress some files (but not *.tsv) it does this only
# for files >64MB, we want this files to be compressed explicitly
for table in query_log zookeeper_log trace_log transactions_info_log metric_log blob_storage_log
for table in query_log zookeeper_log trace_log transactions_info_log metric_log blob_storage_log error_log
do
clickhouse-local "$data_path_config" --only-system-tables --stacktrace -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
@ -379,6 +379,10 @@ fi
tar -chf /test_output/coordination.tar /var/lib/clickhouse/coordination ||:
rm -rf /var/lib/clickhouse/data/system/*/
tar -chf /test_output/store.tar /var/lib/clickhouse/store ||:
tar -chf /test_output/metadata.tar /var/lib/clickhouse/metadata/*.sql ||:
if [[ -n "$USE_DATABASE_REPLICATED" ]] && [[ "$USE_DATABASE_REPLICATED" -eq 1 ]]; then
rg -Fa "<Fatal>" /var/log/clickhouse-server/clickhouse-server1.log ||:
rg -Fa "<Fatal>" /var/log/clickhouse-server/clickhouse-server2.log ||:

View File

@ -89,10 +89,6 @@ function configure()
# since we run clickhouse from root
sudo chown root: /var/lib/clickhouse
# Set more frequent update period of asynchronous metrics to more frequently update information about real memory usage (less chance of OOM).
echo "<clickhouse><asynchronous_metrics_update_period_s>1</asynchronous_metrics_update_period_s></clickhouse>" \
> /etc/clickhouse-server/config.d/asynchronous_metrics_update_period_s.xml
local total_mem
total_mem=$(awk '/MemTotal/ { print $(NF-1) }' /proc/meminfo) # KiB
total_mem=$(( total_mem*1024 )) # bytes

View File

@ -37,7 +37,7 @@ Using named collections:
<named_collections>
<iceberg_conf>
<url>http://test.s3.amazonaws.com/clickhouse-bucket/</url>
<access_key_id>test<access_key_id>
<access_key_id>test</access_key_id>
<secret_access_key>test</secret_access_key>
</iceberg_conf>
</named_collections>

View File

@ -13,7 +13,7 @@ This engine provides integration with [Amazon S3](https://aws.amazon.com/s3/) ec
CREATE TABLE s3_queue_engine_table (name String, value UInt32)
ENGINE = S3Queue(path, [NOSIGN, | aws_access_key_id, aws_secret_access_key,] format, [compression])
[SETTINGS]
[mode = 'unordered',]
[mode = '',]
[after_processing = 'keep',]
[keeper_path = '',]
[s3queue_loading_retries = 0,]

View File

@ -1490,6 +1490,8 @@ Differs from [PrettySpaceNoEscapes](#prettyspacenoescapes) in that up to 10,000
- [output_format_pretty_color](/docs/en/operations/settings/settings-formats.md/#output_format_pretty_color) - use ANSI escape sequences to paint colors in Pretty formats. Default value - `true`.
- [output_format_pretty_grid_charset](/docs/en/operations/settings/settings-formats.md/#output_format_pretty_grid_charset) - Charset for printing grid borders. Available charsets: ASCII, UTF-8. Default value - `UTF-8`.
- [output_format_pretty_row_numbers](/docs/en/operations/settings/settings-formats.md/#output_format_pretty_row_numbers) - Add row numbers before each row for pretty output format. Default value - `true`.
- [output_format_pretty_display_footer_column_names](/docs/en/operations/settings/settings-formats.md/#output_format_pretty_display_footer_column_names) - Display column names in the footer if table contains many rows. Default value - `true`.
- [output_format_pretty_display_footer_column_names_min_rows](/docs/en/operations/settings/settings-formats.md/#output_format_pretty_display_footer_column_names_min_rows) - Sets the minimum number of rows for which a footer will be displayed if [output_format_pretty_display_footer_column_names](/docs/en/operations/settings/settings-formats.md/#output_format_pretty_display_footer_column_names) is enabled. Default value - 50.
## RowBinary {#rowbinary}

View File

@ -508,7 +508,7 @@ Now `rule` can configure `method`, `headers`, `url`, `handler`:
- `headers` are responsible for matching the header part of the HTTP request. It is compatible with RE2s regular expressions. It is an optional configuration. If it is not defined in the configuration file, it does not match the header portion of the HTTP request.
- `handler` contains the main processing part. Now `handler` can configure `type`, `status`, `content_type`, `response_content`, `query`, `query_param_name`.
- `handler` contains the main processing part. Now `handler` can configure `type`, `status`, `content_type`, `http_response_headers`, `response_content`, `query`, `query_param_name`.
`type` currently supports three types: [predefined_query_handler](#predefined_query_handler), [dynamic_query_handler](#dynamic_query_handler), [static](#static).
- `query` — use with `predefined_query_handler` type, executes query when the handler is called.
@ -519,6 +519,8 @@ Now `rule` can configure `method`, `headers`, `url`, `handler`:
- `content_type` — use with any type, response [content-type](https://developer.mozilla.org/en-US/docs/Web/HTTP/Headers/Content-Type).
- `http_response_headers` — use with any type, response headers map. Could be used to set content type as well.
- `response_content` — use with `static` type, response content sent to client, when using the prefix file:// or config://, find the content from the file or configuration sends to client.
Next are the configuration methods for different `type`.
@ -616,6 +618,33 @@ Return a message.
<type>static</type>
<status>402</status>
<content_type>text/html; charset=UTF-8</content_type>
<http_response_headers>
<Content-Language>en</Content-Language>
<X-My-Custom-Header>43</X-My-Custom-Header>
</http_response_headers>
<response_content>Say Hi!</response_content>
</handler>
</rule>
<defaults/>
</http_handlers>
```
`http_response_headers` could be used to set content type instead of `content_type`.
``` xml
<http_handlers>
<rule>
<methods>GET</methods>
<headers><XXX>xxx</XXX></headers>
<url>/hi</url>
<handler>
<type>static</type>
<status>402</status>
<http_response_headers>
<Content-Type>text/html; charset=UTF-8</Content-Type>
<Content-Language>en</Content-Language>
<X-My-Custom-Header>43</X-My-Custom-Header>
</http_response_headers>
<response_content>Say Hi!</response_content>
</handler>
</rule>
@ -696,6 +725,9 @@ Find the content from the file send to client.
<handler>
<type>static</type>
<content_type>text/html; charset=UTF-8</content_type>
<http_response_headers>
<ETag>737060cd8c284d8af7ad3082f209582d</ETag>
</http_response_headers>
<response_content>file:///absolute_path_file.html</response_content>
</handler>
</rule>
@ -706,6 +738,9 @@ Find the content from the file send to client.
<handler>
<type>static</type>
<content_type>text/html; charset=UTF-8</content_type>
<http_response_headers>
<ETag>737060cd8c284d8af7ad3082f209582d</ETag>
</http_response_headers>
<response_content>file://./relative_path_file.html</response_content>
</handler>
</rule>

View File

@ -31,6 +31,56 @@ Alternatively, in order to enable the MySQL interface for an existing service:
3. After entering the password, you will get prompted the MySQL connection string for this service
![Connection screen - MySQL Enabled](./images/mysql5.png)
## Creating multiple MySQL users in ClickHouse Cloud
By default, there is a built-in `mysql4<subdomain>` user, which uses the same password as the `default` one. The `<subdomain>` part is the first segment of your ClickHouse Cloud hostname. This format is necessary to work with the tools that implement secure connection, but don't provide [SNI information in their TLS handshake](https://www.cloudflare.com/learning/ssl/what-is-sni), which makes it impossible to do the internal routing without an extra hint in the username (MySQL console client is one of such tools).
Because of this, we _highly recommend_ following the `mysql4<subdomain>_<username>` format when creating a new user intended to be used with the MySQL interface, where `<subdomain>` is a hint to identify your Cloud service, and `<username>` is an arbitrary suffix of your choice.
:::tip
For ClickHouse Cloud hostname like `foobar.us-east1.aws.clickhouse.cloud`, the `<subdomain>` part equals to `foobar`, and a custom MySQL username could look like `mysql4foobar_team1`.
:::
You can create extra users to use with the MySQL interface if, for example, you need to apply extra settings.
1. Optional - create a [settings profile](https://clickhouse.com/docs/en/sql-reference/statements/create/settings-profile) to apply for your custom user. For example, `my_custom_profile` with an extra setting which will be applied by default when we connect with the user we create later:
```sql
CREATE SETTINGS PROFILE my_custom_profile SETTINGS prefer_column_name_to_alias=1;
```
`prefer_column_name_to_alias` is used just as an example, you can use other settings there.
2. [Create a user](https://clickhouse.com/docs/en/sql-reference/statements/create/user) using the following format: `mysql4<subdomain>_<username>` ([see above](#creating-multiple-mysql-users-in-clickhouse-cloud)). The password must be in double SHA1 format. For example:
```sql
CREATE USER mysql4foobar_team1 IDENTIFIED WITH double_sha1_password BY 'YourPassword42$';
```
or if you want to use a custom profile for this user:
```sql
CREATE USER mysql4foobar_team1 IDENTIFIED WITH double_sha1_password BY 'YourPassword42$' SETTINGS PROFILE 'my_custom_profile';
```
where `my_custom_profile` is the name of the profile you created earlier.
3. [Grant](https://clickhouse.com/docs/en/sql-reference/statements/grant) the new user the necessary permissions to interact with the desired tables or databases. For example, if you want to grant access to `system.query_log` only:
```sql
GRANT SELECT ON system.query_log TO mysql4foobar_team1;
```
4. Use the created user to connect to your ClickHouse Cloud service with the MySQL interface.
### Troubleshooting multiple MySQL users in ClickHouse Cloud
If you created a new MySQL user, and you see the following error while connecting via MySQL CLI client:
```
ERROR 2013 (HY000): Lost connection to MySQL server at 'reading authorization packet', system error: 54
```
In this case, ensure that the username follows the `mysql4<subdomain>_<username>` format, as described ([above](#creating-multiple-mysql-users-in-clickhouse-cloud)).
## Enabling the MySQL Interface On Self-managed ClickHouse
Add the [mysql_port](../operations/server-configuration-parameters/settings.md#server_configuration_parameters-mysql_port) setting to your server's configuration file. For example, you could define the port in a new XML file in your `config.d/` [folder](../operations/configuration-files):

View File

@ -59,10 +59,10 @@ For that, we need to use `jemalloc`'s tool called [jeprof](https://github.com/je
If thats the case, we recommend installing an [alternative implementation](https://github.com/gimli-rs/addr2line) of the tool.
```
git clone https://github.com/gimli-rs/addr2line
git clone https://github.com/gimli-rs/addr2line.git --depth=1 --branch=0.23.0
cd addr2line
cargo b --examples -r
cp ./target/release/examples/addr2line path/to/current/addr2line
cargo build --features bin --release
cp ./target/release/addr2line path/to/current/addr2line
```
:::

View File

@ -591,6 +591,22 @@ Default value: 100000
<max_part_num_to_warn>400</max_part_num_to_warn>
```
## max\_table\_num\_to\_throw {#max-table-num-to-throw}
If number of tables is greater than this value, server will throw an exception. 0 means no limitation. View, remote tables, dictionary, system tables are not counted. Only count table in Atomic/Ordinary/Replicated/Lazy database engine.Default value: 0
**Example**
```xml
<max_table_num_to_throw>400</max_table_num_to_throw>
```
## max\_database\_num\_to\_throw {#max-table-num-to-throw}
If number of _database is greater than this value, server will throw an exception. 0 means no limitation.
Default value: 0
**Example**
```xml
<max_database_num_to_throw>400</max_database_num_to_throw>
```
## max_temporary_data_on_disk_size
@ -938,6 +954,38 @@ Or it can be set in hex:
Everything mentioned above can be applied for `aes_256_gcm_siv` (but the key must be 32 bytes long).
## error_log {#error_log}
It is disabled by default.
**Enabling**
To manually turn on error history collection [`system.error_log`](../../operations/system-tables/error_log.md), create `/etc/clickhouse-server/config.d/error_log.xml` with the following content:
``` xml
<clickhouse>
<error_log>
<database>system</database>
<table>error_log</table>
<flush_interval_milliseconds>7500</flush_interval_milliseconds>
<collect_interval_milliseconds>1000</collect_interval_milliseconds>
<max_size_rows>1048576</max_size_rows>
<reserved_size_rows>8192</reserved_size_rows>
<buffer_size_rows_flush_threshold>524288</buffer_size_rows_flush_threshold>
<flush_on_crash>false</flush_on_crash>
</error_log>
</clickhouse>
```
**Disabling**
To disable `error_log` setting, you should create the following file `/etc/clickhouse-server/config.d/disable_error_log.xml` with the following content:
``` xml
<clickhouse>
<error_log remove="1" />
</clickhouse>
```
## custom_settings_prefixes {#custom_settings_prefixes}
@ -1901,7 +1949,7 @@ For more information, see the MergeTreeSettings.h header file.
## metric_log {#metric_log}
It is enabled by default. If it`s not, you can do this manually.
It is disabled by default.
**Enabling**
@ -3084,3 +3132,21 @@ This setting is only necessary for the migration period and will become obsolete
Type: Bool
Default: 1
## merge_workload {#merge_workload}
Used to regulate how resources are utilized and shared between merges and other workloads. Specified value is used as `workload` setting value for all background merges. Can be overridden by a merge tree setting.
Default value: "default"
**See Also**
- [Workload Scheduling](/docs/en/operations/workload-scheduling.md)
## mutation_workload {#mutation_workload}
Used to regulate how resources are utilized and shared between mutations and other workloads. Specified value is used as `workload` setting value for all background mutations. Can be overridden by a merge tree setting.
Default value: "default"
**See Also**
- [Workload Scheduling](/docs/en/operations/workload-scheduling.md)

View File

@ -974,6 +974,24 @@ Default value: false
- [exclude_deleted_rows_for_part_size_in_merge](#exclude_deleted_rows_for_part_size_in_merge) setting
## merge_workload
Used to regulate how resources are utilized and shared between merges and other workloads. Specified value is used as `workload` setting value for background merges of this table. If not specified (empty string), then server setting `merge_workload` is used instead.
Default value: an empty string
**See Also**
- [Workload Scheduling](/docs/en/operations/workload-scheduling.md)
## mutation_workload
Used to regulate how resources are utilized and shared between mutations and other workloads. Specified value is used as `workload` setting value for background mutations of this table. If not specified (empty string), then server setting `mutation_workload` is used instead.
Default value: an empty string
**See Also**
- [Workload Scheduling](/docs/en/operations/workload-scheduling.md)
### optimize_row_order
Controls if the row order should be optimized during inserts to improve the compressability of the newly inserted table part.

View File

@ -1706,6 +1706,43 @@ Result:
└────────────┘
```
## output_format_pretty_display_footer_column_names
Display column names in the footer if there are many table rows.
Possible values:
- 0 — No column names are displayed in the footer.
- 1 — Column names are displayed in the footer if row count is greater than or equal to the threshold value set by [output_format_pretty_display_footer_column_names_min_rows](#output_format_pretty_display_footer_column_names_min_rows) (50 by default).
Default value: `1`.
**Example**
Query:
```sql
SELECT *, toTypeName(*) FROM (SELECT * FROM system.numbers LIMIT 1000);
```
Result:
```response
┌─number─┬─toTypeName(number)─┐
1. │ 0 │ UInt64 │
2. │ 1 │ UInt64 │
3. │ 2 │ UInt64 │
...
999. │ 998 │ UInt64 │
1000. │ 999 │ UInt64 │
└─number─┴─toTypeName(number)─┘
```
## output_format_pretty_display_footer_column_names_min_rows
Sets the minimum number of rows for which a footer with column names will be displayed if setting [output_format_pretty_display_footer_column_names](#output_format_pretty_display_footer_column_names) is enabled.
Default value: `50`.
## Template format settings {#template-format-settings}
### format_template_resultset {#format_template_resultset}

View File

@ -1592,19 +1592,19 @@ Default value: `default`.
## parallel_replicas_custom_key_range_lower {#parallel_replicas_custom_key_range_lower}
Allows the filter type `range` to split the work evenly between replicas based on the custom range `[parallel_replicas_custom_key_range_lower, INT_MAX]`.
Allows the filter type `range` to split the work evenly between replicas based on the custom range `[parallel_replicas_custom_key_range_lower, INT_MAX]`.
When used in conjuction with [parallel_replicas_custom_key_range_upper](#parallel_replicas_custom_key_range_upper), it lets the filter evenly split the work over replicas for the range `[parallel_replicas_custom_key_range_lower, parallel_replicas_custom_key_range_upper]`.
When used in conjuction with [parallel_replicas_custom_key_range_upper](#parallel_replicas_custom_key_range_upper), it lets the filter evenly split the work over replicas for the range `[parallel_replicas_custom_key_range_lower, parallel_replicas_custom_key_range_upper]`.
Note: This setting will not cause any additional data to be filtered during query processing, rather it changes the points at which the range filter breaks up the range `[0, INT_MAX]` for parallel processing.
Note: This setting will not cause any additional data to be filtered during query processing, rather it changes the points at which the range filter breaks up the range `[0, INT_MAX]` for parallel processing.
## parallel_replicas_custom_key_range_upper {#parallel_replicas_custom_key_range_upper}
Allows the filter type `range` to split the work evenly between replicas based on the custom range `[0, parallel_replicas_custom_key_range_upper]`. A value of 0 disables the upper bound, setting it the max value of the custom key expression.
When used in conjuction with [parallel_replicas_custom_key_range_lower](#parallel_replicas_custom_key_range_lower), it lets the filter evenly split the work over replicas for the range `[parallel_replicas_custom_key_range_lower, parallel_replicas_custom_key_range_upper]`.
When used in conjuction with [parallel_replicas_custom_key_range_lower](#parallel_replicas_custom_key_range_lower), it lets the filter evenly split the work over replicas for the range `[parallel_replicas_custom_key_range_lower, parallel_replicas_custom_key_range_upper]`.
Note: This setting will not cause any additional data to be filtered during query processing, rather it changes the points at which the range filter breaks up the range `[0, INT_MAX]` for parallel processing.
Note: This setting will not cause any additional data to be filtered during query processing, rather it changes the points at which the range filter breaks up the range `[0, INT_MAX]` for parallel processing.
## allow_experimental_parallel_reading_from_replicas
@ -3188,7 +3188,7 @@ Default value: `0`.
## lightweight_deletes_sync {#lightweight_deletes_sync}
The same as 'mutation_sync', but controls only execution of lightweight deletes.
The same as 'mutation_sync', but controls only execution of lightweight deletes.
Possible values:
@ -5150,7 +5150,7 @@ Allows using statistic to optimize the order of [prewhere conditions](../../sql-
## analyze_index_with_space_filling_curves
If a table has a space-filling curve in its index, e.g. `ORDER BY mortonEncode(x, y)`, and the query has conditions on its arguments, e.g. `x >= 10 AND x <= 20 AND y >= 20 AND y <= 30`, use the space-filling curve for index analysis.
If a table has a space-filling curve in its index, e.g. `ORDER BY mortonEncode(x, y)` or `ORDER BY hilbertEncode(x, y)`, and the query has conditions on its arguments, e.g. `x >= 10 AND x <= 20 AND y >= 20 AND y <= 30`, use the space-filling curve for index analysis.
## query_plan_enable_optimizations {#query_plan_enable_optimizations}
@ -5418,6 +5418,15 @@ When set to `false` than all attempts are made with identical timeouts.
Default value: `true`.
## allow_deprecated_snowflake_conversion_functions {#allow_deprecated_snowflake_conversion_functions}
Functions `snowflakeToDateTime`, `snowflakeToDateTime64`, `dateTimeToSnowflake`, and `dateTime64ToSnowflake` are deprecated and disabled by default.
Please use functions `snowflakeIDToDateTime`, `snowflakeIDToDateTime64`, `dateTimeToSnowflakeID`, and `dateTime64ToSnowflakeID` instead.
To re-enable the deprecated functions (e.g., during a transition period), please set this setting to `true`.
Default value: `false`
## allow_experimental_variant_type {#allow_experimental_variant_type}
Allows creation of experimental [Variant](../../sql-reference/data-types/variant.md).

View File

@ -639,6 +639,10 @@ An internal metric of the low-level memory allocator (jemalloc). See https://jem
An internal metric of the low-level memory allocator (jemalloc). See https://jemalloc.net/jemalloc.3.html
### jemalloc.prof.active
An internal metric of the low-level memory allocator (jemalloc). See https://jemalloc.net/jemalloc.3.html
**See Also**
- [Monitoring](../../operations/monitoring.md) — Base concepts of ClickHouse monitoring.

View File

@ -0,0 +1,39 @@
---
slug: /en/operations/system-tables/error_log
---
# error_log
Contains history of error values from table `system.errors`, periodically flushed to disk.
Columns:
- `hostname` ([LowCardinality(String)](../../sql-reference/data-types/string.md)) — Hostname of the server executing the query.
- `event_date` ([Date](../../sql-reference/data-types/date.md)) — Event date.
- `event_time` ([DateTime](../../sql-reference/data-types/datetime.md)) — Event time.
- `code` ([Int32](../../sql-reference/data-types/int-uint.md)) — Code number of the error.
- `error` ([LowCardinality(String)](../../sql-reference/data-types/string.md)) - Name of the error.
- `value` ([UInt64](../../sql-reference/data-types/int-uint.md)) — The number of times this error happened.
- `remote` ([UInt8](../../sql-reference/data-types/int-uint.md)) — Remote exception (i.e. received during one of the distributed queries).
**Example**
``` sql
SELECT * FROM system.error_log LIMIT 1 FORMAT Vertical;
```
``` text
Row 1:
──────
hostname: clickhouse.eu-central1.internal
event_date: 2024-06-18
event_time: 2024-06-18 07:32:39
code: 999
error: KEEPER_EXCEPTION
value: 2
remote: 0
```
**See also**
- [error_log setting](../../operations/server-configuration-parameters/settings.md#error_log) — Enabling and disabling the setting.
- [system.errors](../../operations/system-tables/errors.md) — Contains error codes with the number of times they have been triggered.
- [Monitoring](../../operations/monitoring.md) — Base concepts of ClickHouse monitoring.

View File

@ -113,6 +113,8 @@ Columns:
- `used_functions` ([Array(String)](../../sql-reference/data-types/array.md)) — Canonical names of `functions`, which were used during query execution.
- `used_storages` ([Array(String)](../../sql-reference/data-types/array.md)) — Canonical names of `storages`, which were used during query execution.
- `used_table_functions` ([Array(String)](../../sql-reference/data-types/array.md)) — Canonical names of `table functions`, which were used during query execution.
- `used_privileges` ([Array(String)](../../sql-reference/data-types/array.md)) - Privileges which were successfully checked during query execution.
- `missing_privileges` ([Array(String)](../../sql-reference/data-types/array.md)) - Privileges that are missing during query execution.
- `query_cache_usage` ([Enum8](../../sql-reference/data-types/enum.md)) — Usage of the [query cache](../query-cache.md) during query execution. Values:
- `'Unknown'` = Status unknown.
- `'None'` = The query result was neither written into nor read from the query cache.
@ -194,6 +196,8 @@ used_formats: []
used_functions: []
used_storages: []
used_table_functions: []
used_privileges: []
missing_privileges: []
query_cache_usage: None
```

View File

@ -47,6 +47,8 @@ Example:
Queries can be marked with setting `workload` to distinguish different workloads. If `workload` is not set, than value "default" is used. Note that you are able to specify the other value using settings profiles. Setting constraints can be used to make `workload` constant if you want all queries from the user to be marked with fixed value of `workload` setting.
It is possible to assign a `workload` setting for background activities. Merges and mutations are using `merge_workload` and `mutation_workload` server settings correspondingly. These values can also be overridden for specific tables using `merge_workload` and `mutation_workload` merge tree settings
Let's consider an example of a system with two different workloads: "production" and "development".
```sql
@ -151,6 +153,9 @@ Example:
</clickhouse>
```
## See also
- [system.scheduler](/docs/en/operations/system-tables/scheduler.md)
- [merge_workload](/docs/en/operations/settings/merge-tree-settings.md#merge_workload) merge tree setting
- [merge_workload](/docs/en/operations/server-configuration-parameters/settings.md#merge_workload) global server setting
- [mutation_workload](/docs/en/operations/settings/merge-tree-settings.md#mutation_workload) merge tree setting
- [mutation_workload](/docs/en/operations/server-configuration-parameters/settings.md#mutation_workload) global server setting

View File

@ -25,7 +25,7 @@ stddevPop(x)
**Returned value**
Square root of standard deviation of `x`. [Float64](../../data-types/float.md).
- Square root of standard deviation of `x`. [Float64](../../data-types/float.md).
**Example**

View File

@ -4,30 +4,25 @@ slug: "/en/sql-reference/aggregate-functions/reference/varpop"
sidebar_position: 32
---
This page covers the `varPop` and `varPopStable` functions available in ClickHouse.
## varPop
Calculates the population covariance between two data columns. The population covariance measures the degree to which two variables vary together. Calculates the amount `Σ((x - x̅)^2) / n`, where `n` is the sample size and `x̅`is the average value of `x`.
Calculates the population variance.
**Syntax**
```sql
covarPop(x, y)
varPop(x)
```
Alias: `VAR_POP`.
**Parameters**
- `x`: The first data column. [Numeric](../../../native-protocol/columns.md)
- `y`: The second data column. [Numeric](../../../native-protocol/columns.md)
- `x`: Population of values to find the population variance of. [(U)Int*](../../data-types/int-uint.md), [Float*](../../data-types/float.md), [Decimal*](../../data-types/decimal.md).
**Returned value**
Returns an integer of type `Float64`.
**Implementation details**
This function uses a numerically unstable algorithm. If you need numerical stability in calculations, use the slower but more stable [`varPopStable`](#varpopstable) function.
- Returns the population variance of `x`. [`Float64`](../../data-types/float.md).
**Example**
@ -37,69 +32,21 @@ Query:
DROP TABLE IF EXISTS test_data;
CREATE TABLE test_data
(
x Int32,
y Int32
x UInt8,
)
ENGINE = Memory;
INSERT INTO test_data VALUES (1, 2), (2, 3), (3, 5), (4, 6), (5, 8);
INSERT INTO test_data VALUES (3), (3), (3), (4), (4), (5), (5), (7), (11), (15);
SELECT
covarPop(x, y) AS covar_pop
varPop(x) AS var_pop
FROM test_data;
```
Result:
```response
3
```
## varPopStable
Calculates population covariance between two data columns using a stable, numerically accurate method to calculate the variance. This function is designed to provide reliable results even with large datasets or values that might cause numerical instability in other implementations.
**Syntax**
```sql
covarPopStable(x, y)
```
**Parameters**
- `x`: The first data column. [String literal](../../syntax#syntax-string-literal)
- `y`: The second data column. [Expression](../../syntax#syntax-expressions)
**Returned value**
Returns an integer of type `Float64`.
**Implementation details**
Unlike [`varPop`](#varpop), this function uses a stable, numerically accurate algorithm to calculate the population variance to avoid issues like catastrophic cancellation or loss of precision. This function also handles `NaN` and `Inf` values correctly, excluding them from calculations.
**Example**
Query:
```sql
DROP TABLE IF EXISTS test_data;
CREATE TABLE test_data
(
x Int32,
y Int32
)
ENGINE = Memory;
INSERT INTO test_data VALUES (1, 2), (2, 9), (9, 5), (4, 6), (5, 8);
SELECT
covarPopStable(x, y) AS covar_pop_stable
FROM test_data;
```
Result:
```response
0.5999999999999999
┌─var_pop─┐
│ 14.4 │
└─────────┘
```

View File

@ -0,0 +1,52 @@
---
title: "varPopStable"
slug: "/en/sql-reference/aggregate-functions/reference/varpopstable"
sidebar_position: 32
---
## varPopStable
Returns the population variance. Unlike [`varPop`](../reference/varpop.md), this function uses a [numerically stable](https://en.wikipedia.org/wiki/Numerical_stability) algorithm. It works slower but provides a lower computational error.
**Syntax**
```sql
varPopStable(x)
```
Alias: `VAR_POP_STABLE`.
**Parameters**
- `x`: Population of values to find the population variance of. [(U)Int*](../../data-types/int-uint.md), [Float*](../../data-types/float.md), [Decimal*](../../data-types/decimal.md).
**Returned value**
- Returns the population variance of `x`. [Float64](../../data-types/float.md).
**Example**
Query:
```sql
DROP TABLE IF EXISTS test_data;
CREATE TABLE test_data
(
x UInt8,
)
ENGINE = Memory;
INSERT INTO test_data VALUES (3),(3),(3),(4),(4),(5),(5),(7),(11),(15);
SELECT
varPopStable(x) AS var_pop_stable
FROM test_data;
```
Result:
```response
┌─var_pop_stable─┐
│ 14.4 │
└────────────────┘
```

View File

@ -4,8 +4,6 @@ slug: /en/sql-reference/aggregate-functions/reference/varsamp
sidebar_position: 33
---
This page contains information on the `varSamp` and `varSampStable` ClickHouse functions.
## varSamp
Calculate the sample variance of a data set.
@ -13,24 +11,27 @@ Calculate the sample variance of a data set.
**Syntax**
```sql
varSamp(expr)
varSamp(x)
```
Alias: `VAR_SAMP`.
**Parameters**
- `expr`: An expression representing the data set for which you want to calculate the sample variance. [Expression](../../syntax#syntax-expressions)
- `x`: The population for which you want to calculate the sample variance. [(U)Int*](../../data-types/int-uint.md), [Float*](../../data-types/float.md), [Decimal*](../../data-types/decimal.md).
**Returned value**
Returns a Float64 value representing the sample variance of the input data set.
- Returns the sample variance of the input data set `x`. [Float64](../../data-types/float.md).
**Implementation details**
The `varSamp()` function calculates the sample variance using the following formula:
The `varSamp` function calculates the sample variance using the following formula:
```plaintext
∑(x - mean(x))^2 / (n - 1)
```
$$
\sum\frac{(x - \text{mean}(x))^2}{(n - 1)}
$$
Where:
@ -38,91 +39,29 @@ Where:
- `mean(x)` is the arithmetic mean of the data set.
- `n` is the number of data points in the data set.
The function assumes that the input data set represents a sample from a larger population. If you want to calculate the variance of the entire population (when you have the complete data set), you should use the [`varPop()` function](./varpop#varpop) instead.
This function uses a numerically unstable algorithm. If you need numerical stability in calculations, use the slower but more stable [`varSampStable`](#varsampstable) function.
The function assumes that the input data set represents a sample from a larger population. If you want to calculate the variance of the entire population (when you have the complete data set), you should use [`varPop`](../reference/varpop.md) instead.
**Example**
Query:
```sql
CREATE TABLE example_table
DROP TABLE IF EXISTS test_data;
CREATE TABLE test_data
(
id UInt64,
value Float64
x Float64
)
ENGINE = MergeTree
ORDER BY id;
ENGINE = Memory;
INSERT INTO example_table VALUES (1, 10.5), (2, 12.3), (3, 9.8), (4, 11.2), (5, 10.7);
INSERT INTO test_data VALUES (10.5), (12.3), (9.8), (11.2), (10.7);
SELECT varSamp(value) FROM example_table;
SELECT round(varSamp(x),3) AS var_samp FROM test_data;
```
Response:
```response
0.8650000000000091
┌─var_samp─┐
│ 0.865 │
└──────────┘
```
## varSampStable
Calculate the sample variance of a data set using a numerically stable algorithm.
**Syntax**
```sql
varSampStable(expr)
```
**Parameters**
- `expr`: An expression representing the data set for which you want to calculate the sample variance. [Expression](../../syntax#syntax-expressions)
**Returned value**
The `varSampStable` function returns a Float64 value representing the sample variance of the input data set.
**Implementation details**
The `varSampStable` function calculates the sample variance using the same formula as the [`varSamp`](#varsamp) function:
```plaintext
∑(x - mean(x))^2 / (n - 1)
```
Where:
- `x` is each individual data point in the data set.
- `mean(x)` is the arithmetic mean of the data set.
- `n` is the number of data points in the data set.
The difference between `varSampStable` and `varSamp` is that `varSampStable` is designed to provide a more deterministic and stable result when dealing with floating-point arithmetic. It uses an algorithm that minimizes the accumulation of rounding errors, which can be particularly important when dealing with large data sets or data with a wide range of values.
Like `varSamp`, the `varSampStable` function assumes that the input data set represents a sample from a larger population. If you want to calculate the variance of the entire population (when you have the complete data set), you should use the [`varPopStable`](./varpop#varpopstable) function instead.
**Example**
Query:
```sql
CREATE TABLE example_table
(
id UInt64,
value Float64
)
ENGINE = MergeTree
ORDER BY id;
INSERT INTO example_table VALUES (1, 10.5), (2, 12.3), (3, 9.8), (4, 11.2), (5, 10.7);
SELECT varSampStable(value) FROM example_table;
```
Response:
```response
0.865
```
This query calculates the sample variance of the `value` column in the `example_table` using the `varSampStable()` function. The result shows that the sample variance of the values `[10.5, 12.3, 9.8, 11.2, 10.7]` is approximately 0.865, which may differ slightly from the result of `varSamp` due to the more precise handling of floating-point arithmetic.

View File

@ -0,0 +1,63 @@
---
title: "varSampStable"
slug: /en/sql-reference/aggregate-functions/reference/varsampstable
sidebar_position: 33
---
## varSampStable
Calculate the sample variance of a data set. Unlike [`varSamp`](../reference/varsamp.md), this function uses a numerically stable algorithm. It works slower but provides a lower computational error.
**Syntax**
```sql
varSampStable(x)
```
Alias: `VAR_SAMP_STABLE`
**Parameters**
- `x`: The population for which you want to calculate the sample variance. [(U)Int*](../../data-types/int-uint.md), [Float*](../../data-types/float.md), [Decimal*](../../data-types/decimal.md).
**Returned value**
- Returns the sample variance of the input data set. [Float64](../../data-types/float.md).
**Implementation details**
The `varSampStable` function calculates the sample variance using the same formula as the [`varSamp`](../reference/varsamp.md):
$$
\sum\frac{(x - \text{mean}(x))^2}{(n - 1)}
$$
Where:
- `x` is each individual data point in the data set.
- `mean(x)` is the arithmetic mean of the data set.
- `n` is the number of data points in the data set.
**Example**
Query:
```sql
DROP TABLE IF EXISTS test_data;
CREATE TABLE test_data
(
x Float64
)
ENGINE = Memory;
INSERT INTO test_data VALUES (10.5), (12.3), (9.8), (11.2), (10.7);
SELECT round(varSampStable(x),3) AS var_samp_stable FROM test_data;
```
Response:
```response
┌─var_samp_stable─┐
│ 0.865 │
└─────────────────┘
```

View File

@ -137,7 +137,7 @@ If the time transition (due to daylight saving time or for other reasons) was pe
Non-monotonic calendar dates. For example, in Happy Valley - Goose Bay, the time was transitioned one hour backwards at 00:01:00 7 Nov 2010 (one minute after midnight). So after 6th Nov has ended, people observed a whole one minute of 7th Nov, then time was changed back to 23:01 6th Nov and after another 59 minutes the 7th Nov started again. ClickHouse does not (yet) support this kind of fun. During these days the results of time processing functions may be slightly incorrect.
Similar issue exists for Casey Antarctic station in year 2010. They changed time three hours back at 5 Mar, 02:00. If you are working in antarctic station, please don't afraid to use ClickHouse. Just make sure you set timezone to UTC or be aware of inaccuracies.
Similar issue exists for Casey Antarctic station in year 2010. They changed time three hours back at 5 Mar, 02:00. If you are working in antarctic station, please don't be afraid to use ClickHouse. Just make sure you set timezone to UTC or be aware of inaccuracies.
Time shifts for multiple days. Some pacific islands changed their timezone offset from UTC+14 to UTC-12. That's alright but some inaccuracies may present if you do calculations with their timezone for historical time points at the days of conversion.

View File

@ -7,33 +7,43 @@ sidebar_label: Float32, Float64
# Float32, Float64
:::note
If you need accurate calculations, in particular if you work with financial or business data requiring a high precision you should consider using Decimal instead. Floats might lead to inaccurate results as illustrated below:
If you need accurate calculations, in particular if you work with financial or business data requiring a high precision, you should consider using [Decimal](../data-types/decimal.md) instead.
```
[Floating Point Numbers](https://en.wikipedia.org/wiki/IEEE_754) might lead to inaccurate results as illustrated below:
```sql
CREATE TABLE IF NOT EXISTS float_vs_decimal
(
my_float Float64,
my_decimal Decimal64(3)
)Engine=MergeTree ORDER BY tuple()
INSERT INTO float_vs_decimal SELECT round(randCanonical(), 3) AS res, res FROM system.numbers LIMIT 1000000; # Generate 1 000 000 random number with 2 decimal places and store them as a float and as a decimal
)
Engine=MergeTree
ORDER BY tuple();
# Generate 1 000 000 random numbers with 2 decimal places and store them as a float and as a decimal
INSERT INTO float_vs_decimal SELECT round(randCanonical(), 3) AS res, res FROM system.numbers LIMIT 1000000;
```
```
SELECT sum(my_float), sum(my_decimal) FROM float_vs_decimal;
> 500279.56300000014 500279.563
┌──────sum(my_float)─┬─sum(my_decimal)─┐
│ 499693.60500000004 │ 499693.605 │
└────────────────────┴─────────────────┘
SELECT sumKahan(my_float), sumKahan(my_decimal) FROM float_vs_decimal;
> 500279.563 500279.563
┌─sumKahan(my_float)─┬─sumKahan(my_decimal)─┐
│ 499693.605 │ 499693.605 │
└────────────────────┴──────────────────────┘
```
:::
[Floating point numbers](https://en.wikipedia.org/wiki/IEEE_754).
Types are equivalent to types of C:
The equivalent types in ClickHouse and in C are given below:
- `Float32``float`.
- `Float64``double`.
Aliases:
Float types in ClickHouse have the following aliases:
- `Float32``FLOAT`, `REAL`, `SINGLE`.
- `Float64``DOUBLE`, `DOUBLE PRECISION`.

View File

@ -2178,6 +2178,32 @@ Result:
Alias: levenshteinDistance
## editDistanceUTF8
Calculates the [edit distance](https://en.wikipedia.org/wiki/Edit_distance) between two UTF8 strings.
**Syntax**
```sql
editDistanceUTF8(string1, string2)
```
**Examples**
``` sql
SELECT editDistanceUTF8('我是谁', '我是我');
```
Result:
``` text
┌─editDistanceUTF8('我是谁', '我是我')──┐
│ 1 │
└─────────────────────────────────────┘
```
Alias: levenshteinDistanceUTF8
## damerauLevenshteinDistance
Calculates the [Damerau-Levenshtein distance](https://en.wikipedia.org/wiki/Damerau%E2%80%93Levenshtein_distance) between two byte strings.

View File

@ -543,12 +543,17 @@ serverUUID()
Generates a [Snowflake ID](https://en.wikipedia.org/wiki/Snowflake_ID).
The generated Snowflake ID contains the current Unix timestamp in milliseconds 41 (+ 1 top zero bit) bits, followed by machine id (10 bits), a counter (12 bits) to distinguish IDs within a millisecond.
The generated Snowflake ID contains the current Unix timestamp in milliseconds (41 + 1 top zero bits), followed by a machine id (10 bits), and a counter (12 bits) to distinguish IDs within a millisecond.
For any given timestamp (unix_ts_ms), the counter starts at 0 and is incremented by 1 for each new Snowflake ID until the timestamp changes.
In case the counter overflows, the timestamp field is incremented by 1 and the counter is reset to 0.
Function `generateSnowflakeID` guarantees that the counter field within a timestamp increments monotonically across all function invocations in concurrently running threads and queries.
:::note
The generated Snowflake IDs are based on the UNIX epoch 1970-01-01.
While no standard or recommendation exists for the epoch of Snowflake IDs, implementations in other systems may use a different epoch, e.g. Twitter/X (2010-11-04) or Mastodon (2015-01-01).
:::
```
0 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
@ -605,6 +610,11 @@ SELECT generateSnowflakeID(1), generateSnowflakeID(2);
## snowflakeToDateTime
:::warning
This function is deprecated and can only be used if setting [allow_deprecated_snowflake_conversion_functions](../../operations/settings/settings.md#allow_deprecated_snowflake_conversion_functions) is enabled.
The function will be removed at some point in future.
:::
Extracts the timestamp component of a [Snowflake ID](https://en.wikipedia.org/wiki/Snowflake_ID) in [DateTime](../data-types/datetime.md) format.
**Syntax**
@ -641,6 +651,11 @@ Result:
## snowflakeToDateTime64
:::warning
This function is deprecated and can only be used if setting [allow_deprecated_snowflake_conversion_functions](../../operations/settings/settings.md#allow_deprecated_snowflake_conversion_functions) is enabled.
The function will be removed at some point in future.
:::
Extracts the timestamp component of a [Snowflake ID](https://en.wikipedia.org/wiki/Snowflake_ID) in [DateTime64](../data-types/datetime64.md) format.
**Syntax**
@ -677,6 +692,11 @@ Result:
## dateTimeToSnowflake
:::warning
This function is deprecated and can only be used if setting [allow_deprecated_snowflake_conversion_functions](../../operations/settings/settings.md#allow_deprecated_snowflake_conversion_functions) is enabled.
The function will be removed at some point in future.
:::
Converts a [DateTime](../data-types/datetime.md) value to the first [Snowflake ID](https://en.wikipedia.org/wiki/Snowflake_ID) at the giving time.
**Syntax**
@ -711,6 +731,11 @@ Result:
## dateTime64ToSnowflake
:::warning
This function is deprecated and can only be used if setting [allow_deprecated_snowflake_conversion_functions](../../operations/settings/settings.md#allow_deprecated_snowflake_conversion_functions) is enabled.
The function will be removed at some point in future.
:::
Convert a [DateTime64](../data-types/datetime64.md) to the first [Snowflake ID](https://en.wikipedia.org/wiki/Snowflake_ID) at the giving time.
**Syntax**
@ -743,6 +768,148 @@ Result:
└─────────────────────────────┘
```
## snowflakeIDToDateTime
Returns the timestamp component of a [Snowflake ID](https://en.wikipedia.org/wiki/Snowflake_ID) as a value of type [DateTime](../data-types/datetime.md).
**Syntax**
``` sql
snowflakeIDToDateTime(value[, epoch[, time_zone]])
```
**Arguments**
- `value` — Snowflake ID. [UInt64](../data-types/int-uint.md).
- `epoch` - Epoch of the Snowflake ID in milliseconds since 1970-01-01. Defaults to 0 (1970-01-01). For the Twitter/X epoch (2015-01-01), provide 1288834974657. Optional. [UInt*](../data-types/int-uint.md).
- `time_zone` — [Timezone](/docs/en/operations/server-configuration-parameters/settings.md/#server_configuration_parameters-timezone). The function parses `time_string` according to the timezone. Optional. [String](../data-types/string.md).
**Returned value**
- The timestamp component of `value` as a [DateTime](../data-types/datetime.md) value.
**Example**
Query:
```sql
SELECT snowflakeIDToDateTime(7204436857747984384) AS res
```
Result:
```
┌─────────────────res─┐
│ 2024-06-06 10:59:58 │
└─────────────────────┘
```
## snowflakeIDToDateTime64
Returns the timestamp component of a [Snowflake ID](https://en.wikipedia.org/wiki/Snowflake_ID) as a value of type [DateTime64](../data-types/datetime64.md).
**Syntax**
``` sql
snowflakeIDToDateTime64(value[, epoch[, time_zone]])
```
**Arguments**
- `value` — Snowflake ID. [UInt64](../data-types/int-uint.md).
- `epoch` - Epoch of the Snowflake ID in milliseconds since 1970-01-01. Defaults to 0 (1970-01-01). For the Twitter/X epoch (2015-01-01), provide 1288834974657. Optional. [UInt*](../data-types/int-uint.md).
- `time_zone` — [Timezone](/docs/en/operations/server-configuration-parameters/settings.md/#server_configuration_parameters-timezone). The function parses `time_string` according to the timezone. Optional. [String](../data-types/string.md).
**Returned value**
- The timestamp component of `value` as a [DateTime64](../data-types/datetime64.md) with scale = 3, i.e. millisecond precision.
**Example**
Query:
```sql
SELECT snowflakeIDToDateTime64(7204436857747984384) AS res
```
Result:
```
┌─────────────────res─┐
│ 2024-06-06 10:59:58 │
└─────────────────────┘
```
## dateTimeToSnowflakeID
Converts a [DateTime](../data-types/datetime.md) value to the first [Snowflake ID](https://en.wikipedia.org/wiki/Snowflake_ID) at the giving time.
**Syntax**
``` sql
dateTimeToSnowflakeID(value[, epoch])
```
**Arguments**
- `value` — Date with time. [DateTime](../data-types/datetime.md).
- `epoch` - Epoch of the Snowflake ID in milliseconds since 1970-01-01. Defaults to 0 (1970-01-01). For the Twitter/X epoch (2015-01-01), provide 1288834974657. Optional. [UInt*](../data-types/int-uint.md).
**Returned value**
- Input value converted to [UInt64](../data-types/int-uint.md) as the first Snowflake ID at that time.
**Example**
Query:
```sql
SELECT toDateTime('2021-08-15 18:57:56', 'Asia/Shanghai') AS dt, dateTimeToSnowflakeID(dt) AS res;
```
Result:
```
┌──────────────────dt─┬─────────────────res─┐
│ 2021-08-15 18:57:56 │ 6832626392367104000 │
└─────────────────────┴─────────────────────┘
```
## dateTime64ToSnowflakeID
Convert a [DateTime64](../data-types/datetime64.md) to the first [Snowflake ID](https://en.wikipedia.org/wiki/Snowflake_ID) at the giving time.
**Syntax**
``` sql
dateTime64ToSnowflakeID(value[, epoch])
```
**Arguments**
- `value` — Date with time. [DateTime64](../data-types/datetime64.md).
- `epoch` - Epoch of the Snowflake ID in milliseconds since 1970-01-01. Defaults to 0 (1970-01-01). For the Twitter/X epoch (2015-01-01), provide 1288834974657. Optional. [UInt*](../data-types/int-uint.md).
**Returned value**
- Input value converted to [UInt64](../data-types/int-uint.md) as the first Snowflake ID at that time.
**Example**
Query:
```sql
SELECT toDateTime('2021-08-15 18:57:56.493', 3, 'Asia/Shanghai') AS dt, dateTime64ToSnowflakeID(dt) AS res;
```
Result:
```
┌──────────────────────dt─┬─────────────────res─┐
│ 2021-08-15 18:57:56.493 │ 6832626394434895872 │
└─────────────────────────┴─────────────────────┘
```
## See also
- [dictGetUUID](../functions/ext-dict-functions.md#ext_dict_functions-other)

View File

@ -414,6 +414,8 @@ $ curl -v 'http://localhost:8123/predefined_query'
- `content_type` — используется со всеми типами, возвращает [content-type](https://developer.mozilla.org/en-US/docs/Web/HTTP/Headers/Content-Type).
- `http_response_headers` — используется со всеми типами чтобы добавить кастомные хедеры в ответ. Может использоваться в том числе для задания хедера `Content-Type` вместо `content_type`.
- `response_content` — используется с типом`static`, содержимое ответа, отправленное клиенту, при использовании префикса file:// or config://, находит содержимое из файла или конфигурации, отправленного клиенту.
Далее приведены методы настройки для различных типов.
@ -509,6 +511,33 @@ max_final_threads 2
<type>static</type>
<status>402</status>
<content_type>text/html; charset=UTF-8</content_type>
<http_response_headers>
<Content-Language>en</Content-Language>
<X-My-Custom-Header>43</X-My-Custom-Header>
</http_response_headers>
<response_content>Say Hi!</response_content>
</handler>
</rule>
<defaults/>
</http_handlers>
```
`http_response_headers` так же может использоваться для определения `Content-Type` вместо `content_type`.
``` xml
<http_handlers>
<rule>
<methods>GET</methods>
<headers><XXX>xxx</XXX></headers>
<url>/hi</url>
<handler>
<type>static</type>
<status>402</status>
<http_response_headers>
<Content-Type>text/html; charset=UTF-8</Content-Type>
<Content-Language>en</Content-Language>
<X-My-Custom-Header>43</X-My-Custom-Header>
</http_response_headers>
<response_content>Say Hi!</response_content>
</handler>
</rule>
@ -589,6 +618,9 @@ $ curl -v -H 'XXX:xxx' 'http://localhost:8123/get_config_static_handler'
<handler>
<type>static</type>
<content_type>text/html; charset=UTF-8</content_type>
<http_response_headers>
<ETag>737060cd8c284d8af7ad3082f209582d</ETag>
</http_response_headers>
<response_content>file:///absolute_path_file.html</response_content>
</handler>
</rule>
@ -599,6 +631,9 @@ $ curl -v -H 'XXX:xxx' 'http://localhost:8123/get_config_static_handler'
<handler>
<type>static</type>
<content_type>text/html; charset=UTF-8</content_type>
<http_response_headers>
<ETag>737060cd8c284d8af7ad3082f209582d</ETag>
</http_response_headers>
<response_content>file://./relative_path_file.html</response_content>
</handler>
</rule>

View File

@ -368,7 +368,7 @@ int KeeperClient::main(const std::vector<String> & /* args */)
return 0;
}
DB::ConfigProcessor config_processor(config().getString("config-file", "config.xml"));
ConfigProcessor config_processor(config().getString("config-file", "config.xml"));
/// This will handle a situation when clickhouse is running on the embedded config, but config.d folder is also present.
ConfigProcessor::registerEmbeddedConfig("config.xml", "<clickhouse/>");

View File

@ -12,8 +12,7 @@ bool parseKeeperArg(IParser::Pos & pos, Expected & expected, String & result)
if (!parseIdentifierOrStringLiteral(pos, expected, result))
return false;
}
while (pos->type != TokenType::Whitespace && pos->type != TokenType::EndOfStream && pos->type != TokenType::Semicolon)
else if (pos->type == TokenType::Number)
{
result.append(pos->begin, pos->end);
++pos;
@ -40,8 +39,8 @@ bool KeeperParser::parseImpl(Pos & pos, ASTPtr & node, Expected & expected)
for (const auto & pair : KeeperClient::commands)
expected.add(pos, pair.first.data());
for (const auto & flwc : four_letter_word_commands)
expected.add(pos, flwc.data());
for (const auto & four_letter_word_command : four_letter_word_commands)
expected.add(pos, four_letter_word_command.data());
if (pos->type != TokenType::BareWord)
return false;

View File

@ -11,7 +11,6 @@ namespace DB
{
bool parseKeeperArg(IParser::Pos & pos, Expected & expected, String & result);
bool parseKeeperPath(IParser::Pos & pos, Expected & expected, String & path);

View File

@ -732,11 +732,8 @@ void LocalServer::processConfig()
attachInformationSchema(global_context, *createMemoryDatabaseIfNotExists(global_context, DatabaseCatalog::INFORMATION_SCHEMA_UPPERCASE));
}
server_display_name = config().getString("display_name", getFQDNOrHostName());
prompt_by_server_display_name = config().getRawString("prompt_by_server_display_name.default", "{display_name} :) ");
std::map<String, String> prompt_substitutions{{"display_name", server_display_name}};
for (const auto & [key, value] : prompt_substitutions)
boost::replace_all(prompt_by_server_display_name, "{" + key + "}", value);
server_display_name = config().getString("display_name", "");
prompt_by_server_display_name = config().getRawString("prompt_by_server_display_name.default", ":) ");
global_context->setQueryKindInitial();
global_context->setQueryKind(query_kind);

View File

@ -10,6 +10,7 @@
#include <Poco/Net/NetException.h>
#include <Poco/Util/HelpFormatter.h>
#include <Poco/Environment.h>
#include <Poco/Config.h>
#include <Common/scope_guard_safe.h>
#include <Common/logger_useful.h>
#include <base/phdr_cache.h>
@ -721,11 +722,6 @@ try
CurrentMetrics::set(CurrentMetrics::Revision, ClickHouseRevision::getVersionRevision());
CurrentMetrics::set(CurrentMetrics::VersionInteger, ClickHouseRevision::getVersionInteger());
Poco::ThreadPool server_pool(3, server_settings.max_connections);
std::mutex servers_lock;
std::vector<ProtocolServerAdapter> servers;
std::vector<ProtocolServerAdapter> servers_to_start_before_tables;
/** Context contains all that query execution is dependent:
* settings, available functions, data types, aggregate functions, databases, ...
*/
@ -823,6 +819,18 @@ try
total_memory_tracker.setSampleMaxAllocationSize(server_settings.total_memory_profiler_sample_max_allocation_size);
}
Poco::ThreadPool server_pool(
/* minCapacity */3,
/* maxCapacity */server_settings.max_connections,
/* idleTime */60,
/* stackSize */POCO_THREAD_STACK_SIZE,
server_settings.global_profiler_real_time_period_ns,
server_settings.global_profiler_cpu_time_period_ns);
std::mutex servers_lock;
std::vector<ProtocolServerAdapter> servers;
std::vector<ProtocolServerAdapter> servers_to_start_before_tables;
/// Wait for all threads to avoid possible use-after-free (for example logging objects can be already destroyed).
SCOPE_EXIT({
Stopwatch watch;
@ -1372,8 +1380,8 @@ try
global_context->setQueryCache(query_cache_max_size_in_bytes, query_cache_max_entries, query_cache_query_cache_max_entry_size_in_bytes, query_cache_max_entry_size_in_rows);
#if USE_EMBEDDED_COMPILER
size_t compiled_expression_cache_max_size_in_bytes = config().getUInt64("compiled_expression_cache_size", DEFAULT_COMPILED_EXPRESSION_CACHE_MAX_SIZE);
size_t compiled_expression_cache_max_elements = config().getUInt64("compiled_expression_cache_elements_size", DEFAULT_COMPILED_EXPRESSION_CACHE_MAX_ENTRIES);
size_t compiled_expression_cache_max_size_in_bytes = server_settings.compiled_expression_cache_size;
size_t compiled_expression_cache_max_elements = server_settings.compiled_expression_cache_elements_size;
CompiledExpressionCacheFactory::instance().init(compiled_expression_cache_max_size_in_bytes, compiled_expression_cache_max_elements);
#endif
@ -1609,6 +1617,10 @@ try
0, // We don't need any threads one all the parts will be deleted
new_server_settings.max_parts_cleaning_thread_pool_size);
global_context->setMergeWorkload(new_server_settings.merge_workload);
global_context->setMutationWorkload(new_server_settings.mutation_workload);
if (config->has("resources"))
{
global_context->getResourceManager()->updateConfiguration(*config);

View File

@ -371,7 +371,7 @@
<!-- Enables asynchronous loading of databases and tables to speedup server startup.
Queries to not yet loaded entity will be blocked until load is finished.
-->
<!-- <async_load_databases>true</async_load_databases> -->
<async_load_databases>true</async_load_databases>
<!-- On memory constrained environments you may have to set this to value larger than 1.
-->
@ -1155,6 +1155,18 @@
<flush_on_crash>false</flush_on_crash>
</metric_log>
<!-- Error log contains rows with current values of errors collected with "collect_interval_milliseconds" interval. -->
<error_log>
<database>system</database>
<table>error_log</table>
<flush_interval_milliseconds>7500</flush_interval_milliseconds>
<max_size_rows>1048576</max_size_rows>
<reserved_size_rows>8192</reserved_size_rows>
<buffer_size_rows_flush_threshold>524288</buffer_size_rows_flush_threshold>
<collect_interval_milliseconds>1000</collect_interval_milliseconds>
<flush_on_crash>false</flush_on_crash>
</error_log>
<!--
Asynchronous metric log contains values of metrics from
system.asynchronous_metrics.
@ -1396,6 +1408,14 @@
<!-- <host_name>replica</host_name> -->
</distributed_ddl>
<!-- Used to regulate how resources are utilized and shared between merges, mutations and other workloads.
Specified value is used as `workload` setting value for background merge or mutation.
-->
<!--
<merge_workload>merges_and_mutations</merge_workload>
<mutation_workload>merges_and_mutations</mutation_workload>
-->
<!-- Settings to fine-tune MergeTree tables. See documentation in source code, in MergeTreeSettings.h -->
<!--
<merge_tree>

View File

@ -726,6 +726,13 @@ metric_log:
flush_interval_milliseconds: 7500
collect_interval_milliseconds: 1000
# Error log contains rows with current values of errors collected with "collect_interval_milliseconds" interval.
error_log:
database: system
table: error_log
flush_interval_milliseconds: 7500
collect_interval_milliseconds: 1000
# Asynchronous metric log contains values of metrics from
# system.asynchronous_metrics.
asynchronous_metric_log:

View File

@ -261,7 +261,24 @@ AccessControl::AccessControl()
}
AccessControl::~AccessControl() = default;
AccessControl::~AccessControl()
{
try
{
AccessControl::shutdown();
}
catch (...)
{
tryLogCurrentException(__PRETTY_FUNCTION__);
}
}
void AccessControl::shutdown()
{
MultipleAccessStorage::shutdown();
removeAllStorages();
}
void AccessControl::setUpFromMainConfig(const Poco::Util::AbstractConfiguration & config_, const String & config_path_,

View File

@ -53,6 +53,9 @@ public:
AccessControl();
~AccessControl() override;
/// Shutdown the access control and stops all background activity.
void shutdown() override;
/// Initializes access storage (user directories).
void setUpFromMainConfig(const Poco::Util::AbstractConfiguration & config_, const String & config_path_,
const zkutil::GetZooKeeper & get_zookeeper_function_);

View File

@ -31,6 +31,7 @@ namespace DB
{
namespace ErrorCodes
{
extern const int AUTHENTICATION_FAILED;
extern const int SUPPORT_IS_DISABLED;
extern const int BAD_ARGUMENTS;
extern const int LOGICAL_ERROR;
@ -90,8 +91,10 @@ bool AuthenticationData::Util::checkPasswordBcrypt(std::string_view password [[m
{
#if USE_BCRYPT
int ret = bcrypt_checkpw(password.data(), reinterpret_cast<const char *>(password_bcrypt.data()));
/// Before 24.6 we didn't validate hashes on creation, so it could be that the stored hash is invalid
/// and it could not be decoded by the library
if (ret == -1)
throw Exception(ErrorCodes::LOGICAL_ERROR, "BCrypt library failed: bcrypt_checkpw returned {}", ret);
throw Exception(ErrorCodes::AUTHENTICATION_FAILED, "Internal failure decoding Bcrypt hash");
return (ret == 0);
#else
throw Exception(
@ -230,6 +233,17 @@ void AuthenticationData::setPasswordHashBinary(const Digest & hash)
throw Exception(ErrorCodes::BAD_ARGUMENTS,
"Password hash for the 'BCRYPT_PASSWORD' authentication type has length {} "
"but must be 59 or 60 bytes.", hash.size());
auto resized = hash;
resized.resize(64);
#if USE_BCRYPT
/// Verify that it is a valid hash
int ret = bcrypt_checkpw("", reinterpret_cast<const char *>(resized.data()));
if (ret == -1)
throw Exception(ErrorCodes::BAD_ARGUMENTS, "Could not decode the provided hash with 'bcrypt_hash'");
#endif
password_hash = hash;
password_hash.resize(64);
return;

View File

@ -4,12 +4,12 @@
namespace DB
{
CachedAccessChecking::CachedAccessChecking(const std::shared_ptr<const ContextAccess> & access_, AccessFlags access_flags_)
CachedAccessChecking::CachedAccessChecking(const std::shared_ptr<const ContextAccessWrapper> & access_, AccessFlags access_flags_)
: CachedAccessChecking(access_, AccessRightsElement{access_flags_})
{
}
CachedAccessChecking::CachedAccessChecking(const std::shared_ptr<const ContextAccess> & access_, const AccessRightsElement & element_)
CachedAccessChecking::CachedAccessChecking(const std::shared_ptr<const ContextAccessWrapper> & access_, const AccessRightsElement & element_)
: access(access_), element(element_)
{
}

View File

@ -1,6 +1,7 @@
#pragma once
#include <Access/Common/AccessRightsElement.h>
#include <Access/ContextAccess.h>
#include <memory>
@ -13,14 +14,14 @@ class ContextAccess;
class CachedAccessChecking
{
public:
CachedAccessChecking(const std::shared_ptr<const ContextAccess> & access_, AccessFlags access_flags_);
CachedAccessChecking(const std::shared_ptr<const ContextAccess> & access_, const AccessRightsElement & element_);
CachedAccessChecking(const std::shared_ptr<const ContextAccessWrapper> & access_, AccessFlags access_flags_);
CachedAccessChecking(const std::shared_ptr<const ContextAccessWrapper> & access_, const AccessRightsElement & element_);
~CachedAccessChecking();
bool checkAccess(bool throw_if_denied = true);
private:
const std::shared_ptr<const ContextAccess> access;
const std::shared_ptr<const ContextAccessWrapper> access;
const AccessRightsElement element;
bool checked = false;
bool result = false;

View File

@ -20,6 +20,7 @@
#include <boost/algorithm/string/join.hpp>
#include <boost/range/algorithm/set_algorithm.hpp>
#include <cassert>
#include <unordered_set>
namespace DB
@ -271,7 +272,7 @@ namespace
std::shared_ptr<const ContextAccess> ContextAccess::fromContext(const ContextPtr & context)
{
return context->getAccess();
return ContextAccessWrapper::fromContext(context)->getAccess();
}
@ -560,7 +561,7 @@ std::shared_ptr<const AccessRights> ContextAccess::getAccessRightsWithImplicit()
template <bool throw_if_denied, bool grant_option, typename... Args>
bool ContextAccess::checkAccessImplHelper(AccessFlags flags, const Args &... args) const
bool ContextAccess::checkAccessImplHelper(const ContextPtr & context, AccessFlags flags, const Args &... args) const
{
if (user_was_dropped)
{
@ -573,8 +574,10 @@ bool ContextAccess::checkAccessImplHelper(AccessFlags flags, const Args &... arg
if (params.full_access)
return true;
auto access_granted = []
auto access_granted = [&]
{
if constexpr (throw_if_denied)
context->addQueryPrivilegesInfo(AccessRightsElement{flags, args...}.toStringWithoutOptions(), true);
return true;
};
@ -583,7 +586,10 @@ bool ContextAccess::checkAccessImplHelper(AccessFlags flags, const Args &... arg
FmtArgs && ...fmt_args [[maybe_unused]])
{
if constexpr (throw_if_denied)
{
context->addQueryPrivilegesInfo(AccessRightsElement{flags, args...}.toStringWithoutOptions(), false);
throw Exception(error_code, std::move(fmt_string), getUserName(), std::forward<FmtArgs>(fmt_args)...);
}
return false;
};
@ -686,102 +692,102 @@ bool ContextAccess::checkAccessImplHelper(AccessFlags flags, const Args &... arg
}
template <bool throw_if_denied, bool grant_option>
bool ContextAccess::checkAccessImpl(const AccessFlags & flags) const
bool ContextAccess::checkAccessImpl(const ContextPtr & context, const AccessFlags & flags) const
{
return checkAccessImplHelper<throw_if_denied, grant_option>(flags);
return checkAccessImplHelper<throw_if_denied, grant_option>(context, flags);
}
template <bool throw_if_denied, bool grant_option, typename... Args>
bool ContextAccess::checkAccessImpl(const AccessFlags & flags, std::string_view database, const Args &... args) const
bool ContextAccess::checkAccessImpl(const ContextPtr & context, const AccessFlags & flags, std::string_view database, const Args &... args) const
{
return checkAccessImplHelper<throw_if_denied, grant_option>(flags, database.empty() ? params.current_database : database, args...);
return checkAccessImplHelper<throw_if_denied, grant_option>(context, flags, database.empty() ? params.current_database : database, args...);
}
template <bool throw_if_denied, bool grant_option>
bool ContextAccess::checkAccessImplHelper(const AccessRightsElement & element) const
bool ContextAccess::checkAccessImplHelper(const ContextPtr & context, const AccessRightsElement & element) const
{
assert(!element.grant_option || grant_option);
if (element.isGlobalWithParameter())
{
if (element.any_parameter)
return checkAccessImpl<throw_if_denied, grant_option>(element.access_flags);
return checkAccessImpl<throw_if_denied, grant_option>(context, element.access_flags);
else
return checkAccessImpl<throw_if_denied, grant_option>(element.access_flags, element.parameter);
return checkAccessImpl<throw_if_denied, grant_option>(context, element.access_flags, element.parameter);
}
else if (element.any_database)
return checkAccessImpl<throw_if_denied, grant_option>(element.access_flags);
return checkAccessImpl<throw_if_denied, grant_option>(context, element.access_flags);
else if (element.any_table)
return checkAccessImpl<throw_if_denied, grant_option>(element.access_flags, element.database);
return checkAccessImpl<throw_if_denied, grant_option>(context, element.access_flags, element.database);
else if (element.any_column)
return checkAccessImpl<throw_if_denied, grant_option>(element.access_flags, element.database, element.table);
return checkAccessImpl<throw_if_denied, grant_option>(context, element.access_flags, element.database, element.table);
else
return checkAccessImpl<throw_if_denied, grant_option>(element.access_flags, element.database, element.table, element.columns);
return checkAccessImpl<throw_if_denied, grant_option>(context, element.access_flags, element.database, element.table, element.columns);
}
template <bool throw_if_denied, bool grant_option>
bool ContextAccess::checkAccessImpl(const AccessRightsElement & element) const
bool ContextAccess::checkAccessImpl(const ContextPtr & context, const AccessRightsElement & element) const
{
if constexpr (grant_option)
{
return checkAccessImplHelper<throw_if_denied, true>(element);
return checkAccessImplHelper<throw_if_denied, true>(context, element);
}
else
{
if (element.grant_option)
return checkAccessImplHelper<throw_if_denied, true>(element);
return checkAccessImplHelper<throw_if_denied, true>(context, element);
else
return checkAccessImplHelper<throw_if_denied, false>(element);
return checkAccessImplHelper<throw_if_denied, false>(context, element);
}
}
template <bool throw_if_denied, bool grant_option>
bool ContextAccess::checkAccessImpl(const AccessRightsElements & elements) const
bool ContextAccess::checkAccessImpl(const ContextPtr & context, const AccessRightsElements & elements) const
{
for (const auto & element : elements)
if (!checkAccessImpl<throw_if_denied, grant_option>(element))
if (!checkAccessImpl<throw_if_denied, grant_option>(context, element))
return false;
return true;
}
bool ContextAccess::isGranted(const AccessFlags & flags) const { return checkAccessImpl<false, false>(flags); }
bool ContextAccess::isGranted(const AccessFlags & flags, std::string_view database) const { return checkAccessImpl<false, false>(flags, database); }
bool ContextAccess::isGranted(const AccessFlags & flags, std::string_view database, std::string_view table) const { return checkAccessImpl<false, false>(flags, database, table); }
bool ContextAccess::isGranted(const AccessFlags & flags, std::string_view database, std::string_view table, std::string_view column) const { return checkAccessImpl<false, false>(flags, database, table, column); }
bool ContextAccess::isGranted(const AccessFlags & flags, std::string_view database, std::string_view table, const std::vector<std::string_view> & columns) const { return checkAccessImpl<false, false>(flags, database, table, columns); }
bool ContextAccess::isGranted(const AccessFlags & flags, std::string_view database, std::string_view table, const Strings & columns) const { return checkAccessImpl<false, false>(flags, database, table, columns); }
bool ContextAccess::isGranted(const AccessRightsElement & element) const { return checkAccessImpl<false, false>(element); }
bool ContextAccess::isGranted(const AccessRightsElements & elements) const { return checkAccessImpl<false, false>(elements); }
bool ContextAccess::isGranted(const ContextPtr & context, const AccessFlags & flags) const { return checkAccessImpl<false, false>(context, flags); }
bool ContextAccess::isGranted(const ContextPtr & context, const AccessFlags & flags, std::string_view database) const { return checkAccessImpl<false, false>(context, flags, database); }
bool ContextAccess::isGranted(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table) const { return checkAccessImpl<false, false>(context, flags, database, table); }
bool ContextAccess::isGranted(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table, std::string_view column) const { return checkAccessImpl<false, false>(context, flags, database, table, column); }
bool ContextAccess::isGranted(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table, const std::vector<std::string_view> & columns) const { return checkAccessImpl<false, false>(context, flags, database, table, columns); }
bool ContextAccess::isGranted(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table, const Strings & columns) const { return checkAccessImpl<false, false>(context, flags, database, table, columns); }
bool ContextAccess::isGranted(const ContextPtr & context, const AccessRightsElement & element) const { return checkAccessImpl<false, false>(context, element); }
bool ContextAccess::isGranted(const ContextPtr & context, const AccessRightsElements & elements) const { return checkAccessImpl<false, false>(context, elements); }
bool ContextAccess::hasGrantOption(const AccessFlags & flags) const { return checkAccessImpl<false, true>(flags); }
bool ContextAccess::hasGrantOption(const AccessFlags & flags, std::string_view database) const { return checkAccessImpl<false, true>(flags, database); }
bool ContextAccess::hasGrantOption(const AccessFlags & flags, std::string_view database, std::string_view table) const { return checkAccessImpl<false, true>(flags, database, table); }
bool ContextAccess::hasGrantOption(const AccessFlags & flags, std::string_view database, std::string_view table, std::string_view column) const { return checkAccessImpl<false, true>(flags, database, table, column); }
bool ContextAccess::hasGrantOption(const AccessFlags & flags, std::string_view database, std::string_view table, const std::vector<std::string_view> & columns) const { return checkAccessImpl<false, true>(flags, database, table, columns); }
bool ContextAccess::hasGrantOption(const AccessFlags & flags, std::string_view database, std::string_view table, const Strings & columns) const { return checkAccessImpl<false, true>(flags, database, table, columns); }
bool ContextAccess::hasGrantOption(const AccessRightsElement & element) const { return checkAccessImpl<false, true>(element); }
bool ContextAccess::hasGrantOption(const AccessRightsElements & elements) const { return checkAccessImpl<false, true>(elements); }
bool ContextAccess::hasGrantOption(const ContextPtr & context, const AccessFlags & flags) const { return checkAccessImpl<false, true>(context, flags); }
bool ContextAccess::hasGrantOption(const ContextPtr & context, const AccessFlags & flags, std::string_view database) const { return checkAccessImpl<false, true>(context, flags, database); }
bool ContextAccess::hasGrantOption(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table) const { return checkAccessImpl<false, true>(context, flags, database, table); }
bool ContextAccess::hasGrantOption(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table, std::string_view column) const { return checkAccessImpl<false, true>(context, flags, database, table, column); }
bool ContextAccess::hasGrantOption(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table, const std::vector<std::string_view> & columns) const { return checkAccessImpl<false, true>(context, flags, database, table, columns); }
bool ContextAccess::hasGrantOption(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table, const Strings & columns) const { return checkAccessImpl<false, true>(context, flags, database, table, columns); }
bool ContextAccess::hasGrantOption(const ContextPtr & context, const AccessRightsElement & element) const { return checkAccessImpl<false, true>(context, element); }
bool ContextAccess::hasGrantOption(const ContextPtr & context, const AccessRightsElements & elements) const { return checkAccessImpl<false, true>(context, elements); }
void ContextAccess::checkAccess(const AccessFlags & flags) const { checkAccessImpl<true, false>(flags); }
void ContextAccess::checkAccess(const AccessFlags & flags, std::string_view database) const { checkAccessImpl<true, false>(flags, database); }
void ContextAccess::checkAccess(const AccessFlags & flags, std::string_view database, std::string_view table) const { checkAccessImpl<true, false>(flags, database, table); }
void ContextAccess::checkAccess(const AccessFlags & flags, std::string_view database, std::string_view table, std::string_view column) const { checkAccessImpl<true, false>(flags, database, table, column); }
void ContextAccess::checkAccess(const AccessFlags & flags, std::string_view database, std::string_view table, const std::vector<std::string_view> & columns) const { checkAccessImpl<true, false>(flags, database, table, columns); }
void ContextAccess::checkAccess(const AccessFlags & flags, std::string_view database, std::string_view table, const Strings & columns) const { checkAccessImpl<true, false>(flags, database, table, columns); }
void ContextAccess::checkAccess(const AccessRightsElement & element) const { checkAccessImpl<true, false>(element); }
void ContextAccess::checkAccess(const AccessRightsElements & elements) const { checkAccessImpl<true, false>(elements); }
void ContextAccess::checkAccess(const ContextPtr & context, const AccessFlags & flags) const { checkAccessImpl<true, false>(context, flags); }
void ContextAccess::checkAccess(const ContextPtr & context, const AccessFlags & flags, std::string_view database) const { checkAccessImpl<true, false>(context, flags, database); }
void ContextAccess::checkAccess(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table) const { checkAccessImpl<true, false>(context, flags, database, table); }
void ContextAccess::checkAccess(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table, std::string_view column) const { checkAccessImpl<true, false>(context, flags, database, table, column); }
void ContextAccess::checkAccess(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table, const std::vector<std::string_view> & columns) const { checkAccessImpl<true, false>(context, flags, database, table, columns); }
void ContextAccess::checkAccess(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table, const Strings & columns) const { checkAccessImpl<true, false>(context, flags, database, table, columns); }
void ContextAccess::checkAccess(const ContextPtr & context, const AccessRightsElement & element) const { checkAccessImpl<true, false>(context, element); }
void ContextAccess::checkAccess(const ContextPtr & context, const AccessRightsElements & elements) const { checkAccessImpl<true, false>(context, elements); }
void ContextAccess::checkGrantOption(const AccessFlags & flags) const { checkAccessImpl<true, true>(flags); }
void ContextAccess::checkGrantOption(const AccessFlags & flags, std::string_view database) const { checkAccessImpl<true, true>(flags, database); }
void ContextAccess::checkGrantOption(const AccessFlags & flags, std::string_view database, std::string_view table) const { checkAccessImpl<true, true>(flags, database, table); }
void ContextAccess::checkGrantOption(const AccessFlags & flags, std::string_view database, std::string_view table, std::string_view column) const { checkAccessImpl<true, true>(flags, database, table, column); }
void ContextAccess::checkGrantOption(const AccessFlags & flags, std::string_view database, std::string_view table, const std::vector<std::string_view> & columns) const { checkAccessImpl<true, true>(flags, database, table, columns); }
void ContextAccess::checkGrantOption(const AccessFlags & flags, std::string_view database, std::string_view table, const Strings & columns) const { checkAccessImpl<true, true>(flags, database, table, columns); }
void ContextAccess::checkGrantOption(const AccessRightsElement & element) const { checkAccessImpl<true, true>(element); }
void ContextAccess::checkGrantOption(const AccessRightsElements & elements) const { checkAccessImpl<true, true>(elements); }
void ContextAccess::checkGrantOption(const ContextPtr & context, const AccessFlags & flags) const { checkAccessImpl<true, true>(context, flags); }
void ContextAccess::checkGrantOption(const ContextPtr & context, const AccessFlags & flags, std::string_view database) const { checkAccessImpl<true, true>(context, flags, database); }
void ContextAccess::checkGrantOption(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table) const { checkAccessImpl<true, true>(context, flags, database, table); }
void ContextAccess::checkGrantOption(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table, std::string_view column) const { checkAccessImpl<true, true>(context, flags, database, table, column); }
void ContextAccess::checkGrantOption(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table, const std::vector<std::string_view> & columns) const { checkAccessImpl<true, true>(context, flags, database, table, columns); }
void ContextAccess::checkGrantOption(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table, const Strings & columns) const { checkAccessImpl<true, true>(context, flags, database, table, columns); }
void ContextAccess::checkGrantOption(const ContextPtr & context, const AccessRightsElement & element) const { checkAccessImpl<true, true>(context, element); }
void ContextAccess::checkGrantOption(const ContextPtr & context, const AccessRightsElements & elements) const { checkAccessImpl<true, true>(context, elements); }
template <bool throw_if_denied, typename Container, typename GetNameFunction>
bool ContextAccess::checkAdminOptionImplHelper(const Container & role_ids, const GetNameFunction & get_name_function) const
bool ContextAccess::checkAdminOptionImplHelper(const ContextPtr & context, const Container & role_ids, const GetNameFunction & get_name_function) const
{
auto show_error = []<typename... FmtArgs>(int error_code [[maybe_unused]],
FormatStringHelper<FmtArgs...> fmt_string [[maybe_unused]],
@ -804,7 +810,7 @@ bool ContextAccess::checkAdminOptionImplHelper(const Container & role_ids, const
if (!std::size(role_ids))
return true;
if (isGranted(AccessType::ROLE_ADMIN))
if (isGranted(context, AccessType::ROLE_ADMIN))
return true;
auto info = getRolesInfo();
@ -840,54 +846,54 @@ bool ContextAccess::checkAdminOptionImplHelper(const Container & role_ids, const
}
template <bool throw_if_denied>
bool ContextAccess::checkAdminOptionImpl(const UUID & role_id) const
bool ContextAccess::checkAdminOptionImpl(const ContextPtr & context, const UUID & role_id) const
{
return checkAdminOptionImplHelper<throw_if_denied>(to_array(role_id), [this](const UUID & id, size_t) { return access_control->tryReadName(id); });
return checkAdminOptionImplHelper<throw_if_denied>(context, to_array(role_id), [this](const UUID & id, size_t) { return access_control->tryReadName(id); });
}
template <bool throw_if_denied>
bool ContextAccess::checkAdminOptionImpl(const UUID & role_id, const String & role_name) const
bool ContextAccess::checkAdminOptionImpl(const ContextPtr & context, const UUID & role_id, const String & role_name) const
{
return checkAdminOptionImplHelper<throw_if_denied>(to_array(role_id), [&role_name](const UUID &, size_t) { return std::optional<String>{role_name}; });
return checkAdminOptionImplHelper<throw_if_denied>(context, to_array(role_id), [&role_name](const UUID &, size_t) { return std::optional<String>{role_name}; });
}
template <bool throw_if_denied>
bool ContextAccess::checkAdminOptionImpl(const UUID & role_id, const std::unordered_map<UUID, String> & names_of_roles) const
bool ContextAccess::checkAdminOptionImpl(const ContextPtr & context, const UUID & role_id, const std::unordered_map<UUID, String> & names_of_roles) const
{
return checkAdminOptionImplHelper<throw_if_denied>(to_array(role_id), [&names_of_roles](const UUID & id, size_t) { auto it = names_of_roles.find(id); return (it != names_of_roles.end()) ? it->second : std::optional<String>{}; });
return checkAdminOptionImplHelper<throw_if_denied>(context, to_array(role_id), [&names_of_roles](const UUID & id, size_t) { auto it = names_of_roles.find(id); return (it != names_of_roles.end()) ? it->second : std::optional<String>{}; });
}
template <bool throw_if_denied>
bool ContextAccess::checkAdminOptionImpl(const std::vector<UUID> & role_ids) const
bool ContextAccess::checkAdminOptionImpl(const ContextPtr & context, const std::vector<UUID> & role_ids) const
{
return checkAdminOptionImplHelper<throw_if_denied>(role_ids, [this](const UUID & id, size_t) { return access_control->tryReadName(id); });
return checkAdminOptionImplHelper<throw_if_denied>(context, role_ids, [this](const UUID & id, size_t) { return access_control->tryReadName(id); });
}
template <bool throw_if_denied>
bool ContextAccess::checkAdminOptionImpl(const std::vector<UUID> & role_ids, const Strings & names_of_roles) const
bool ContextAccess::checkAdminOptionImpl(const ContextPtr & context, const std::vector<UUID> & role_ids, const Strings & names_of_roles) const
{
return checkAdminOptionImplHelper<throw_if_denied>(role_ids, [&names_of_roles](const UUID &, size_t i) { return std::optional<String>{names_of_roles[i]}; });
return checkAdminOptionImplHelper<throw_if_denied>(context, role_ids, [&names_of_roles](const UUID &, size_t i) { return std::optional<String>{names_of_roles[i]}; });
}
template <bool throw_if_denied>
bool ContextAccess::checkAdminOptionImpl(const std::vector<UUID> & role_ids, const std::unordered_map<UUID, String> & names_of_roles) const
bool ContextAccess::checkAdminOptionImpl(const ContextPtr & context, const std::vector<UUID> & role_ids, const std::unordered_map<UUID, String> & names_of_roles) const
{
return checkAdminOptionImplHelper<throw_if_denied>(role_ids, [&names_of_roles](const UUID & id, size_t) { auto it = names_of_roles.find(id); return (it != names_of_roles.end()) ? it->second : std::optional<String>{}; });
return checkAdminOptionImplHelper<throw_if_denied>(context, role_ids, [&names_of_roles](const UUID & id, size_t) { auto it = names_of_roles.find(id); return (it != names_of_roles.end()) ? it->second : std::optional<String>{}; });
}
bool ContextAccess::hasAdminOption(const UUID & role_id) const { return checkAdminOptionImpl<false>(role_id); }
bool ContextAccess::hasAdminOption(const UUID & role_id, const String & role_name) const { return checkAdminOptionImpl<false>(role_id, role_name); }
bool ContextAccess::hasAdminOption(const UUID & role_id, const std::unordered_map<UUID, String> & names_of_roles) const { return checkAdminOptionImpl<false>(role_id, names_of_roles); }
bool ContextAccess::hasAdminOption(const std::vector<UUID> & role_ids) const { return checkAdminOptionImpl<false>(role_ids); }
bool ContextAccess::hasAdminOption(const std::vector<UUID> & role_ids, const Strings & names_of_roles) const { return checkAdminOptionImpl<false>(role_ids, names_of_roles); }
bool ContextAccess::hasAdminOption(const std::vector<UUID> & role_ids, const std::unordered_map<UUID, String> & names_of_roles) const { return checkAdminOptionImpl<false>(role_ids, names_of_roles); }
bool ContextAccess::hasAdminOption(const ContextPtr & context, const UUID & role_id) const { return checkAdminOptionImpl<false>(context, role_id); }
bool ContextAccess::hasAdminOption(const ContextPtr & context, const UUID & role_id, const String & role_name) const { return checkAdminOptionImpl<false>(context, role_id, role_name); }
bool ContextAccess::hasAdminOption(const ContextPtr & context, const UUID & role_id, const std::unordered_map<UUID, String> & names_of_roles) const { return checkAdminOptionImpl<false>(context, role_id, names_of_roles); }
bool ContextAccess::hasAdminOption(const ContextPtr & context, const std::vector<UUID> & role_ids) const { return checkAdminOptionImpl<false>(context, role_ids); }
bool ContextAccess::hasAdminOption(const ContextPtr & context, const std::vector<UUID> & role_ids, const Strings & names_of_roles) const { return checkAdminOptionImpl<false>(context, role_ids, names_of_roles); }
bool ContextAccess::hasAdminOption(const ContextPtr & context, const std::vector<UUID> & role_ids, const std::unordered_map<UUID, String> & names_of_roles) const { return checkAdminOptionImpl<false>(context, role_ids, names_of_roles); }
void ContextAccess::checkAdminOption(const UUID & role_id) const { checkAdminOptionImpl<true>(role_id); }
void ContextAccess::checkAdminOption(const UUID & role_id, const String & role_name) const { checkAdminOptionImpl<true>(role_id, role_name); }
void ContextAccess::checkAdminOption(const UUID & role_id, const std::unordered_map<UUID, String> & names_of_roles) const { checkAdminOptionImpl<true>(role_id, names_of_roles); }
void ContextAccess::checkAdminOption(const std::vector<UUID> & role_ids) const { checkAdminOptionImpl<true>(role_ids); }
void ContextAccess::checkAdminOption(const std::vector<UUID> & role_ids, const Strings & names_of_roles) const { checkAdminOptionImpl<true>(role_ids, names_of_roles); }
void ContextAccess::checkAdminOption(const std::vector<UUID> & role_ids, const std::unordered_map<UUID, String> & names_of_roles) const { checkAdminOptionImpl<true>(role_ids, names_of_roles); }
void ContextAccess::checkAdminOption(const ContextPtr & context, const UUID & role_id) const { checkAdminOptionImpl<true>(context, role_id); }
void ContextAccess::checkAdminOption(const ContextPtr & context, const UUID & role_id, const String & role_name) const { checkAdminOptionImpl<true>(context, role_id, role_name); }
void ContextAccess::checkAdminOption(const ContextPtr & context, const UUID & role_id, const std::unordered_map<UUID, String> & names_of_roles) const { checkAdminOptionImpl<true>(context, role_id, names_of_roles); }
void ContextAccess::checkAdminOption(const ContextPtr & context, const std::vector<UUID> & role_ids) const { checkAdminOptionImpl<true>(context, role_ids); }
void ContextAccess::checkAdminOption(const ContextPtr & context, const std::vector<UUID> & role_ids, const Strings & names_of_roles) const { checkAdminOptionImpl<true>(context, role_ids, names_of_roles); }
void ContextAccess::checkAdminOption(const ContextPtr & context, const std::vector<UUID> & role_ids, const std::unordered_map<UUID, String> & names_of_roles) const { checkAdminOptionImpl<true>(context, role_ids, names_of_roles); }
void ContextAccess::checkGranteeIsAllowed(const UUID & grantee_id, const IAccessEntity & grantee) const
@ -919,4 +925,10 @@ void ContextAccess::checkGranteesAreAllowed(const std::vector<UUID> & grantee_id
}
}
std::shared_ptr<const ContextAccessWrapper> ContextAccessWrapper::fromContext(const ContextPtr & context)
{
return context->getAccess();
}
}

View File

@ -4,9 +4,12 @@
#include <Access/ContextAccessParams.h>
#include <Access/EnabledRowPolicies.h>
#include <Interpreters/ClientInfo.h>
#include <Access/QuotaUsage.h>
#include <Common/SettingsChanges.h>
#include <Core/UUID.h>
#include <base/scope_guard.h>
#include <boost/container/flat_set.hpp>
#include <memory>
#include <mutex>
#include <optional>
#include <unordered_map>
@ -71,59 +74,59 @@ public:
/// Checks if a specified access is granted, and throws an exception if not.
/// Empty database means the current database.
void checkAccess(const AccessFlags & flags) const;
void checkAccess(const AccessFlags & flags, std::string_view database) const;
void checkAccess(const AccessFlags & flags, std::string_view database, std::string_view table) const;
void checkAccess(const AccessFlags & flags, std::string_view database, std::string_view table, std::string_view column) const;
void checkAccess(const AccessFlags & flags, std::string_view database, std::string_view table, const std::vector<std::string_view> & columns) const;
void checkAccess(const AccessFlags & flags, std::string_view database, std::string_view table, const Strings & columns) const;
void checkAccess(const AccessRightsElement & element) const;
void checkAccess(const AccessRightsElements & elements) const;
void checkAccess(const ContextPtr & context, const AccessFlags & flags) const;
void checkAccess(const ContextPtr & context, const AccessFlags & flags, std::string_view database) const;
void checkAccess(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table) const;
void checkAccess(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table, std::string_view column) const;
void checkAccess(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table, const std::vector<std::string_view> & columns) const;
void checkAccess(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table, const Strings & columns) const;
void checkAccess(const ContextPtr & context, const AccessRightsElement & element) const;
void checkAccess(const ContextPtr & context, const AccessRightsElements & elements) const;
void checkGrantOption(const AccessFlags & flags) const;
void checkGrantOption(const AccessFlags & flags, std::string_view database) const;
void checkGrantOption(const AccessFlags & flags, std::string_view database, std::string_view table) const;
void checkGrantOption(const AccessFlags & flags, std::string_view database, std::string_view table, std::string_view column) const;
void checkGrantOption(const AccessFlags & flags, std::string_view database, std::string_view table, const std::vector<std::string_view> & columns) const;
void checkGrantOption(const AccessFlags & flags, std::string_view database, std::string_view table, const Strings & columns) const;
void checkGrantOption(const AccessRightsElement & element) const;
void checkGrantOption(const AccessRightsElements & elements) const;
void checkGrantOption(const ContextPtr & context, const AccessFlags & flags) const;
void checkGrantOption(const ContextPtr & context, const AccessFlags & flags, std::string_view database) const;
void checkGrantOption(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table) const;
void checkGrantOption(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table, std::string_view column) const;
void checkGrantOption(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table, const std::vector<std::string_view> & columns) const;
void checkGrantOption(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table, const Strings & columns) const;
void checkGrantOption(const ContextPtr & context, const AccessRightsElement & element) const;
void checkGrantOption(const ContextPtr & context, const AccessRightsElements & elements) const;
/// Checks if a specified access is granted, and returns false if not.
/// Empty database means the current database.
bool isGranted(const AccessFlags & flags) const;
bool isGranted(const AccessFlags & flags, std::string_view database) const;
bool isGranted(const AccessFlags & flags, std::string_view database, std::string_view table) const;
bool isGranted(const AccessFlags & flags, std::string_view database, std::string_view table, std::string_view column) const;
bool isGranted(const AccessFlags & flags, std::string_view database, std::string_view table, const std::vector<std::string_view> & columns) const;
bool isGranted(const AccessFlags & flags, std::string_view database, std::string_view table, const Strings & columns) const;
bool isGranted(const AccessRightsElement & element) const;
bool isGranted(const AccessRightsElements & elements) const;
bool isGranted(const ContextPtr & context, const AccessFlags & flags) const;
bool isGranted(const ContextPtr & context, const AccessFlags & flags, std::string_view database) const;
bool isGranted(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table) const;
bool isGranted(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table, std::string_view column) const;
bool isGranted(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table, const std::vector<std::string_view> & columns) const;
bool isGranted(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table, const Strings & columns) const;
bool isGranted(const ContextPtr & context, const AccessRightsElement & element) const;
bool isGranted(const ContextPtr & context, const AccessRightsElements & elements) const;
bool hasGrantOption(const AccessFlags & flags) const;
bool hasGrantOption(const AccessFlags & flags, std::string_view database) const;
bool hasGrantOption(const AccessFlags & flags, std::string_view database, std::string_view table) const;
bool hasGrantOption(const AccessFlags & flags, std::string_view database, std::string_view table, std::string_view column) const;
bool hasGrantOption(const AccessFlags & flags, std::string_view database, std::string_view table, const std::vector<std::string_view> & columns) const;
bool hasGrantOption(const AccessFlags & flags, std::string_view database, std::string_view table, const Strings & columns) const;
bool hasGrantOption(const AccessRightsElement & element) const;
bool hasGrantOption(const AccessRightsElements & elements) const;
bool hasGrantOption(const ContextPtr & context, const AccessFlags & flags) const;
bool hasGrantOption(const ContextPtr & context, const AccessFlags & flags, std::string_view database) const;
bool hasGrantOption(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table) const;
bool hasGrantOption(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table, std::string_view column) const;
bool hasGrantOption(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table, const std::vector<std::string_view> & columns) const;
bool hasGrantOption(const ContextPtr & context, const AccessFlags & flags, std::string_view database, std::string_view table, const Strings & columns) const;
bool hasGrantOption(const ContextPtr & context, const AccessRightsElement & element) const;
bool hasGrantOption(const ContextPtr & context, const AccessRightsElements & elements) const;
/// Checks if a specified role is granted with admin option, and throws an exception if not.
void checkAdminOption(const UUID & role_id) const;
void checkAdminOption(const UUID & role_id, const String & role_name) const;
void checkAdminOption(const UUID & role_id, const std::unordered_map<UUID, String> & names_of_roles) const;
void checkAdminOption(const std::vector<UUID> & role_ids) const;
void checkAdminOption(const std::vector<UUID> & role_ids, const Strings & names_of_roles) const;
void checkAdminOption(const std::vector<UUID> & role_ids, const std::unordered_map<UUID, String> & names_of_roles) const;
void checkAdminOption(const ContextPtr & context, const UUID & role_id) const;
void checkAdminOption(const ContextPtr & context, const UUID & role_id, const String & role_name) const;
void checkAdminOption(const ContextPtr & context, const UUID & role_id, const std::unordered_map<UUID, String> & names_of_roles) const;
void checkAdminOption(const ContextPtr & context, const std::vector<UUID> & role_ids) const;
void checkAdminOption(const ContextPtr & context, const std::vector<UUID> & role_ids, const Strings & names_of_roles) const;
void checkAdminOption(const ContextPtr & context, const std::vector<UUID> & role_ids, const std::unordered_map<UUID, String> & names_of_roles) const;
/// Checks if a specified role is granted with admin option, and returns false if not.
bool hasAdminOption(const UUID & role_id) const;
bool hasAdminOption(const UUID & role_id, const String & role_name) const;
bool hasAdminOption(const UUID & role_id, const std::unordered_map<UUID, String> & names_of_roles) const;
bool hasAdminOption(const std::vector<UUID> & role_ids) const;
bool hasAdminOption(const std::vector<UUID> & role_ids, const Strings & names_of_roles) const;
bool hasAdminOption(const std::vector<UUID> & role_ids, const std::unordered_map<UUID, String> & names_of_roles) const;
bool hasAdminOption(const ContextPtr & context, const UUID & role_id) const;
bool hasAdminOption(const ContextPtr & context, const UUID & role_id, const String & role_name) const;
bool hasAdminOption(const ContextPtr & context, const UUID & role_id, const std::unordered_map<UUID, String> & names_of_roles) const;
bool hasAdminOption(const ContextPtr & context, const std::vector<UUID> & role_ids) const;
bool hasAdminOption(const ContextPtr & context, const std::vector<UUID> & role_ids, const Strings & names_of_roles) const;
bool hasAdminOption(const ContextPtr & context, const std::vector<UUID> & role_ids, const std::unordered_map<UUID, String> & names_of_roles) const;
/// Checks if a grantee is allowed for the current user, throws an exception if not.
void checkGranteeIsAllowed(const UUID & grantee_id, const IAccessEntity & grantee) const;
@ -142,43 +145,43 @@ private:
void calculateAccessRights() const TSA_REQUIRES(mutex);
template <bool throw_if_denied, bool grant_option>
bool checkAccessImpl(const AccessFlags & flags) const;
bool checkAccessImpl(const ContextPtr & context, const AccessFlags & flags) const;
template <bool throw_if_denied, bool grant_option, typename... Args>
bool checkAccessImpl(const AccessFlags & flags, std::string_view database, const Args &... args) const;
bool checkAccessImpl(const ContextPtr & context, const AccessFlags & flags, std::string_view database, const Args &... args) const;
template <bool throw_if_denied, bool grant_option>
bool checkAccessImpl(const AccessRightsElement & element) const;
bool checkAccessImpl(const ContextPtr & context, const AccessRightsElement & element) const;
template <bool throw_if_denied, bool grant_option>
bool checkAccessImpl(const AccessRightsElements & elements) const;
bool checkAccessImpl(const ContextPtr & context, const AccessRightsElements & elements) const;
template <bool throw_if_denied, bool grant_option, typename... Args>
bool checkAccessImplHelper(AccessFlags flags, const Args &... args) const;
bool checkAccessImplHelper(const ContextPtr & context, AccessFlags flags, const Args &... args) const;
template <bool throw_if_denied, bool grant_option>
bool checkAccessImplHelper(const AccessRightsElement & element) const;
bool checkAccessImplHelper(const ContextPtr & context, const AccessRightsElement & element) const;
template <bool throw_if_denied>
bool checkAdminOptionImpl(const UUID & role_id) const;
bool checkAdminOptionImpl(const ContextPtr & context, const UUID & role_id) const;
template <bool throw_if_denied>
bool checkAdminOptionImpl(const UUID & role_id, const String & role_name) const;
bool checkAdminOptionImpl(const ContextPtr & context, const UUID & role_id, const String & role_name) const;
template <bool throw_if_denied>
bool checkAdminOptionImpl(const UUID & role_id, const std::unordered_map<UUID, String> & names_of_roles) const;
bool checkAdminOptionImpl(const ContextPtr & context, const UUID & role_id, const std::unordered_map<UUID, String> & names_of_roles) const;
template <bool throw_if_denied>
bool checkAdminOptionImpl(const std::vector<UUID> & role_ids) const;
bool checkAdminOptionImpl(const ContextPtr & context, const std::vector<UUID> & role_ids) const;
template <bool throw_if_denied>
bool checkAdminOptionImpl(const std::vector<UUID> & role_ids, const Strings & names_of_roles) const;
bool checkAdminOptionImpl(const ContextPtr & context, const std::vector<UUID> & role_ids, const Strings & names_of_roles) const;
template <bool throw_if_denied>
bool checkAdminOptionImpl(const std::vector<UUID> & role_ids, const std::unordered_map<UUID, String> & names_of_roles) const;
bool checkAdminOptionImpl(const ContextPtr & context, const std::vector<UUID> & role_ids, const std::unordered_map<UUID, String> & names_of_roles) const;
template <bool throw_if_denied, typename Container, typename GetNameFunction>
bool checkAdminOptionImplHelper(const Container & role_ids, const GetNameFunction & get_name_function) const;
bool checkAdminOptionImplHelper(const ContextPtr & context, const Container & role_ids, const GetNameFunction & get_name_function) const;
const AccessControl * access_control = nullptr;
const Params params;
@ -203,4 +206,115 @@ private:
mutable std::shared_ptr<const EnabledSettings> enabled_settings TSA_GUARDED_BY(mutex);
};
/// This wrapper was added to be able to pass the current context to the access
/// without the need to change the signature and all calls to the ContextAccess itself.
/// Right now a context is used to store privileges that are checked for a query,
/// and might be useful for something else in the future as well.
class ContextAccessWrapper : public std::enable_shared_from_this<ContextAccessWrapper>
{
public:
using ContextAccessPtr = std::shared_ptr<const ContextAccess>;
ContextAccessWrapper(const ContextAccessPtr & access_, const ContextPtr & context_): access(access_), context(context_) {}
~ContextAccessWrapper() = default;
static std::shared_ptr<const ContextAccessWrapper> fromContext(const ContextPtr & context);
const ContextAccess::Params & getParams() const { return access->getParams(); }
const ContextAccessPtr & getAccess() const { return access; }
/// Returns the current user. Throws if user is nullptr.
ALWAYS_INLINE UserPtr getUser() const { return access->getUser(); }
/// Same as above, but can return nullptr.
ALWAYS_INLINE UserPtr tryGetUser() const { return access->tryGetUser(); }
ALWAYS_INLINE String getUserName() const { return access->getUserName(); }
ALWAYS_INLINE std::optional<UUID> getUserID() const { return access->getUserID(); }
/// Returns information about current and enabled roles.
ALWAYS_INLINE std::shared_ptr<const EnabledRolesInfo> getRolesInfo() const { return access->getRolesInfo(); }
/// Returns the row policy filter for a specified table.
/// The function returns nullptr if there is no filter to apply.
ALWAYS_INLINE RowPolicyFilterPtr getRowPolicyFilter(const String & database, const String & table_name, RowPolicyFilterType filter_type) const { return access->getRowPolicyFilter(database, table_name, filter_type); }
/// Returns the quota to track resource consumption.
ALWAYS_INLINE std::shared_ptr<const EnabledQuota> getQuota() const { return access->getQuota(); }
ALWAYS_INLINE std::optional<QuotaUsage> getQuotaUsage() const { return access->getQuotaUsage(); }
/// Returns the default settings, i.e. the settings which should be applied on user's login.
ALWAYS_INLINE SettingsChanges getDefaultSettings() const { return access->getDefaultSettings(); }
ALWAYS_INLINE std::shared_ptr<const SettingsProfilesInfo> getDefaultProfileInfo() const { return access->getDefaultProfileInfo(); }
/// Returns the current access rights.
ALWAYS_INLINE std::shared_ptr<const AccessRights> getAccessRights() const { return access->getAccessRights(); }
ALWAYS_INLINE std::shared_ptr<const AccessRights> getAccessRightsWithImplicit() const { return access->getAccessRightsWithImplicit(); }
/// Checks if a specified access is granted, and throws an exception if not.
/// Empty database means the current database.
ALWAYS_INLINE void checkAccess(const AccessFlags & flags) const { access->checkAccess(context, flags); }
ALWAYS_INLINE void checkAccess(const AccessFlags & flags, std::string_view database) const { access->checkAccess(context, flags, database); }
ALWAYS_INLINE void checkAccess(const AccessFlags & flags, std::string_view database, std::string_view table) const { access->checkAccess(context, flags, database, table); }
ALWAYS_INLINE void checkAccess(const AccessFlags & flags, std::string_view database, std::string_view table, std::string_view column) const { access->checkAccess(context, flags, database, table, column); }
ALWAYS_INLINE void checkAccess(const AccessFlags & flags, std::string_view database, std::string_view table, const std::vector<std::string_view> & columns) const { access->checkAccess(context, flags, database, table, columns); }
ALWAYS_INLINE void checkAccess(const AccessFlags & flags, std::string_view database, std::string_view table, const Strings & columns) const { access->checkAccess(context, flags, database, table, columns); }
ALWAYS_INLINE void checkAccess(const AccessRightsElement & element) const { access->checkAccess(context, element); }
ALWAYS_INLINE void checkAccess(const AccessRightsElements & elements) const { access->checkAccess(context, elements); }
ALWAYS_INLINE void checkGrantOption(const AccessFlags & flags) const { access->checkGrantOption(context, flags); }
ALWAYS_INLINE void checkGrantOption(const AccessFlags & flags, std::string_view database) const { access->checkGrantOption(context, flags, database); }
ALWAYS_INLINE void checkGrantOption(const AccessFlags & flags, std::string_view database, std::string_view table) const { access->checkGrantOption(context, flags, database, table); }
ALWAYS_INLINE void checkGrantOption(const AccessFlags & flags, std::string_view database, std::string_view table, std::string_view column) const { access->checkGrantOption(context, flags, database, table, column); }
ALWAYS_INLINE void checkGrantOption(const AccessFlags & flags, std::string_view database, std::string_view table, const std::vector<std::string_view> & columns) const { access->checkGrantOption(context, flags, database, table, columns); }
ALWAYS_INLINE void checkGrantOption(const AccessFlags & flags, std::string_view database, std::string_view table, const Strings & columns) const { access->checkGrantOption(context, flags, database, table, columns); }
ALWAYS_INLINE void checkGrantOption(const AccessRightsElement & element) const { access->checkGrantOption(context, element); }
ALWAYS_INLINE void checkGrantOption(const AccessRightsElements & elements) const { access->checkGrantOption(context, elements); }
/// Checks if a specified access is granted, and returns false if not.
/// Empty database means the current database.
ALWAYS_INLINE bool isGranted(const AccessFlags & flags) const { return access->isGranted(context, flags); }
ALWAYS_INLINE bool isGranted(const AccessFlags & flags, std::string_view database) const { return access->isGranted(context, flags, database); }
ALWAYS_INLINE bool isGranted(const AccessFlags & flags, std::string_view database, std::string_view table) const { return access->isGranted(context, flags, database, table); }
ALWAYS_INLINE bool isGranted(const AccessFlags & flags, std::string_view database, std::string_view table, std::string_view column) const { return access->isGranted(context, flags, database, table, column); }
ALWAYS_INLINE bool isGranted(const AccessFlags & flags, std::string_view database, std::string_view table, const std::vector<std::string_view> & columns) const { return access->isGranted(context, flags, database, table, columns); }
ALWAYS_INLINE bool isGranted(const AccessFlags & flags, std::string_view database, std::string_view table, const Strings & columns) const { return access->isGranted(context, flags, database, table, columns); }
ALWAYS_INLINE bool isGranted(const AccessRightsElement & element) const { return access->isGranted(context, element); }
ALWAYS_INLINE bool isGranted(const AccessRightsElements & elements) const { return access->isGranted(context, elements); }
ALWAYS_INLINE bool hasGrantOption(const AccessFlags & flags) const { return access->hasGrantOption(context, flags); }
ALWAYS_INLINE bool hasGrantOption(const AccessFlags & flags, std::string_view database) const { return access->hasGrantOption(context, flags, database); }
ALWAYS_INLINE bool hasGrantOption(const AccessFlags & flags, std::string_view database, std::string_view table) const { return access->hasGrantOption(context, flags, database, table); }
ALWAYS_INLINE bool hasGrantOption(const AccessFlags & flags, std::string_view database, std::string_view table, std::string_view column) const { return access->hasGrantOption(context, flags, database, table, column); }
ALWAYS_INLINE bool hasGrantOption(const AccessFlags & flags, std::string_view database, std::string_view table, const std::vector<std::string_view> & columns) const { return access->hasGrantOption(context, flags, database, table, columns); }
ALWAYS_INLINE bool hasGrantOption(const AccessFlags & flags, std::string_view database, std::string_view table, const Strings & columns) const { return access->hasGrantOption(context, flags, database, table, columns); }
ALWAYS_INLINE bool hasGrantOption(const AccessRightsElement & element) const { return access->hasGrantOption(context, element); }
ALWAYS_INLINE bool hasGrantOption(const AccessRightsElements & elements) const { return access->hasGrantOption(context, elements); }
/// Checks if a specified role is granted with admin option, and throws an exception if not.
ALWAYS_INLINE void checkAdminOption(const UUID & role_id) const { access->checkAdminOption(context, role_id); }
ALWAYS_INLINE void checkAdminOption(const UUID & role_id, const String & role_name) const { access->checkAdminOption(context, role_id, role_name); }
ALWAYS_INLINE void checkAdminOption(const UUID & role_id, const std::unordered_map<UUID, String> & names_of_roles) const { access->checkAdminOption(context, role_id, names_of_roles); }
ALWAYS_INLINE void checkAdminOption(const std::vector<UUID> & role_ids) const { access->checkAdminOption(context, role_ids); }
ALWAYS_INLINE void checkAdminOption(const std::vector<UUID> & role_ids, const Strings & names_of_roles) const { access->checkAdminOption(context, role_ids, names_of_roles); }
ALWAYS_INLINE void checkAdminOption(const std::vector<UUID> & role_ids, const std::unordered_map<UUID, String> & names_of_roles) const { access->checkAdminOption(context, role_ids, names_of_roles); }
/// Checks if a specified role is granted with admin option, and returns false if not.
ALWAYS_INLINE bool hasAdminOption(const UUID & role_id) const { return access->hasAdminOption(context, role_id); }
ALWAYS_INLINE bool hasAdminOption(const UUID & role_id, const String & role_name) const { return access->hasAdminOption(context, role_id, role_name); }
ALWAYS_INLINE bool hasAdminOption(const UUID & role_id, const std::unordered_map<UUID, String> & names_of_roles) const { return access->hasAdminOption(context, role_id, names_of_roles); }
ALWAYS_INLINE bool hasAdminOption(const std::vector<UUID> & role_ids) const { return access->hasAdminOption(context, role_ids); }
ALWAYS_INLINE bool hasAdminOption(const std::vector<UUID> & role_ids, const Strings & names_of_roles) const { return access->hasAdminOption(context, role_ids, names_of_roles); }
ALWAYS_INLINE bool hasAdminOption(const std::vector<UUID> & role_ids, const std::unordered_map<UUID, String> & names_of_roles) const { return access->hasAdminOption(context, role_ids, names_of_roles); }
/// Checks if a grantee is allowed for the current user, throws an exception if not.
ALWAYS_INLINE void checkGranteeIsAllowed(const UUID & grantee_id, const IAccessEntity & grantee) const { access->checkGranteeIsAllowed(grantee_id, grantee); }
/// Checks if grantees are allowed for the current user, throws an exception if not.
ALWAYS_INLINE void checkGranteesAreAllowed(const std::vector<UUID> & grantee_ids) const { access->checkGranteesAreAllowed(grantee_ids); }
private:
ContextAccessPtr access;
ContextPtr context;
};
}

View File

@ -194,11 +194,9 @@ DiskAccessStorage::DiskAccessStorage(const String & storage_name_, const String
DiskAccessStorage::~DiskAccessStorage()
{
stopListsWritingThread();
try
{
writeLists();
DiskAccessStorage::shutdown();
}
catch (...)
{
@ -207,6 +205,17 @@ DiskAccessStorage::~DiskAccessStorage()
}
void DiskAccessStorage::shutdown()
{
stopListsWritingThread();
{
std::lock_guard lock{mutex};
writeLists();
}
}
String DiskAccessStorage::getStorageParamsJSON() const
{
std::lock_guard lock{mutex};

View File

@ -18,6 +18,8 @@ public:
DiskAccessStorage(const String & storage_name_, const String & directory_path_, AccessChangesNotifier & changes_notifier_, bool readonly_, bool allow_backup_);
~DiskAccessStorage() override;
void shutdown() override;
const char * getStorageType() const override { return STORAGE_TYPE; }
String getStorageParamsJSON() const override;

View File

@ -44,6 +44,11 @@ public:
explicit IAccessStorage(const String & storage_name_) : storage_name(storage_name_) {}
virtual ~IAccessStorage() = default;
/// If the AccessStorage has to do some complicated work when destroying - do it in advance.
/// For example, if the AccessStorage contains any threads for background work - ask them to complete and wait for completion.
/// By default, does nothing.
virtual void shutdown() {}
/// Returns the name of this storage.
const String & getStorageName() const { return storage_name; }
virtual const char * getStorageType() const = 0;

View File

@ -34,11 +34,23 @@ MultipleAccessStorage::MultipleAccessStorage(const String & storage_name_)
MultipleAccessStorage::~MultipleAccessStorage()
{
/// It's better to remove the storages in the reverse order because they could depend on each other somehow.
try
{
MultipleAccessStorage::shutdown();
}
catch (...)
{
tryLogCurrentException(__PRETTY_FUNCTION__);
}
}
void MultipleAccessStorage::shutdown()
{
/// It's better to shutdown the storages in the reverse order because they could depend on each other somehow.
const auto storages = getStoragesPtr();
for (const auto & storage : *storages | boost::adaptors::reversed)
{
removeStorage(storage);
storage->shutdown();
}
}
@ -72,6 +84,16 @@ void MultipleAccessStorage::removeStorage(const StoragePtr & storage_to_remove)
ids_cache.clear();
}
void MultipleAccessStorage::removeAllStorages()
{
/// It's better to remove the storages in the reverse order because they could depend on each other somehow.
const auto storages = getStoragesPtr();
for (const auto & storage : *storages | boost::adaptors::reversed)
{
removeStorage(storage);
}
}
std::vector<StoragePtr> MultipleAccessStorage::getStorages()
{
return *getStoragesPtr();

View File

@ -21,6 +21,8 @@ public:
explicit MultipleAccessStorage(const String & storage_name_ = STORAGE_TYPE);
~MultipleAccessStorage() override;
void shutdown() override;
const char * getStorageType() const override { return STORAGE_TYPE; }
bool isReadOnly() const override;
bool isReadOnly(const UUID & id) const override;
@ -32,6 +34,7 @@ public:
void setStorages(const std::vector<StoragePtr> & storages);
void addStorage(const StoragePtr & new_storage);
void removeStorage(const StoragePtr & storage_to_remove);
void removeAllStorages();
std::vector<StoragePtr> getStorages();
std::vector<ConstStoragePtr> getStorages() const;
std::shared_ptr<const std::vector<StoragePtr>> getStoragesPtr();

View File

@ -66,6 +66,18 @@ ReplicatedAccessStorage::ReplicatedAccessStorage(
}
ReplicatedAccessStorage::~ReplicatedAccessStorage()
{
try
{
ReplicatedAccessStorage::shutdown();
}
catch (...)
{
tryLogCurrentException(__PRETTY_FUNCTION__);
}
}
void ReplicatedAccessStorage::shutdown()
{
stopWatchingThread();
}

View File

@ -23,6 +23,8 @@ public:
ReplicatedAccessStorage(const String & storage_name, const String & zookeeper_path, zkutil::GetZooKeeper get_zookeeper, AccessChangesNotifier & changes_notifier_, bool allow_backup);
~ReplicatedAccessStorage() override;
void shutdown() override;
const char * getStorageType() const override { return STORAGE_TYPE; }
void startPeriodicReloading() override { startWatchingThread(); }

View File

@ -15,22 +15,8 @@ namespace ErrorCodes
bool operator==(const SettingsProfilesInfo & lhs, const SettingsProfilesInfo & rhs)
{
if (lhs.settings != rhs.settings)
return false;
if (lhs.constraints != rhs.constraints)
return false;
if (lhs.profiles != rhs.profiles)
return false;
if (lhs.profiles_with_implicit != rhs.profiles_with_implicit)
return false;
if (lhs.names_of_profiles != rhs.names_of_profiles)
return false;
return true;
return std::tie(lhs.settings, lhs.constraints, lhs.profiles, lhs.profiles_with_implicit, lhs.names_of_profiles)
== std::tie(rhs.settings, rhs.constraints, rhs.profiles, rhs.profiles_with_implicit, rhs.names_of_profiles);
}
std::shared_ptr<const SettingsConstraintsAndProfileIDs>
@ -66,18 +52,20 @@ Strings SettingsProfilesInfo::getProfileNames() const
{
Strings result;
result.reserve(profiles.size());
for (const auto & profile_id : profiles)
for (const UUID & profile_uuid : profiles)
{
const auto p = names_of_profiles.find(profile_id);
if (p != names_of_profiles.end())
result.push_back(p->second);
const auto names_it = names_of_profiles.find(profile_uuid);
if (names_it != names_of_profiles.end())
{
result.push_back(names_it->second);
}
else
{
if (const auto name = access_control.tryReadName(profile_id))
if (const auto name = access_control.tryReadName(profile_uuid))
// We could've updated cache here, but it is a very rare case, so don't bother.
result.push_back(*name);
else
throw Exception(ErrorCodes::LOGICAL_ERROR, "Unable to get profile name for {}", toString(profile_id));
throw Exception(ErrorCodes::LOGICAL_ERROR, "Unable to get profile name for {}", toString(profile_uuid));
}
}

View File

@ -29,7 +29,11 @@ struct SettingsProfilesInfo
/// Names of all the profiles in `profiles`.
std::unordered_map<UUID, String> names_of_profiles;
explicit SettingsProfilesInfo(const AccessControl & access_control_) : constraints(access_control_), access_control(access_control_) {}
explicit SettingsProfilesInfo(const AccessControl & access_control_)
: constraints(access_control_), access_control(access_control_)
{
}
std::shared_ptr<const SettingsConstraintsAndProfileIDs> getConstraintsAndProfileIDs(
const std::shared_ptr<const SettingsConstraintsAndProfileIDs> & previous = nullptr) const;

View File

@ -228,6 +228,11 @@ public:
return prefix_size + nested_func->sizeOfData();
}
size_t alignOfData() const override
{
return std::max(alignof(Data), nested_func->alignOfData());
}
void create(AggregateDataPtr __restrict place) const override
{
new (place) Data;

View File

@ -985,18 +985,18 @@ std::string QueryAnalyzer::rewriteAggregateFunctionNameIfNeeded(
{
result_aggregate_function_name = settings.count_distinct_implementation;
}
else if (aggregate_function_name_lowercase == "countdistinctif" || aggregate_function_name_lowercase == "countifdistinct")
else if (aggregate_function_name_lowercase == "countifdistinct" ||
(settings.rewrite_count_distinct_if_with_count_distinct_implementation && aggregate_function_name_lowercase == "countdistinctif"))
{
result_aggregate_function_name = settings.count_distinct_implementation;
result_aggregate_function_name += "If";
}
/// Replace aggregateFunctionIfDistinct into aggregateFunctionDistinctIf to make execution more optimal
if (result_aggregate_function_name.ends_with("ifdistinct"))
else if (aggregate_function_name_lowercase.ends_with("ifdistinct"))
{
/// Replace aggregateFunctionIfDistinct into aggregateFunctionDistinctIf to make execution more optimal
size_t prefix_length = result_aggregate_function_name.size() - strlen("ifdistinct");
result_aggregate_function_name = result_aggregate_function_name.substr(0, prefix_length) + "DistinctIf";
}
}
bool need_add_or_null = settings.aggregate_functions_null_for_empty && !result_aggregate_function_name.ends_with("OrNull");
if (need_add_or_null)

View File

@ -9,6 +9,8 @@
#include <Interpreters/convertFieldToType.h>
#include <Interpreters/Set.h>
#include <Common/assert_cast.h>
namespace DB
{
@ -54,8 +56,9 @@ size_t getCompoundTypeDepth(const IDataType & type)
}
template <typename Collection>
Block createBlockFromCollection(const Collection & collection, const DataTypes & block_types, bool transform_null_in)
Block createBlockFromCollection(const Collection & collection, const DataTypes& value_types, const DataTypes & block_types, bool transform_null_in)
{
assert(collection.size() == value_types.size());
size_t columns_size = block_types.size();
MutableColumns columns(columns_size);
for (size_t i = 0; i < columns_size; ++i)
@ -66,13 +69,17 @@ Block createBlockFromCollection(const Collection & collection, const DataTypes &
Row tuple_values;
for (const auto & value : collection)
for (size_t collection_index = 0; collection_index < collection.size(); ++collection_index)
{
const auto & value = collection[collection_index];
if (columns_size == 1)
{
auto field = convertFieldToTypeStrict(value, *block_types[0]);
const DataTypePtr & data_type = value_types[collection_index];
auto field = convertFieldToTypeStrict(value, *data_type, *block_types[0]);
if (!field)
{
continue;
}
bool need_insert_null = transform_null_in && block_types[0]->isNullable();
if (!field->isNull() || need_insert_null)
@ -87,6 +94,9 @@ Block createBlockFromCollection(const Collection & collection, const DataTypes &
value.getTypeName());
const auto & tuple = value.template get<const Tuple &>();
const DataTypePtr & value_type = value_types[collection_index];
const DataTypes & tuple_value_type = typeid_cast<const DataTypeTuple *>(value_type.get())->getElements();
size_t tuple_size = tuple.size();
if (tuple_size != columns_size)
@ -101,7 +111,7 @@ Block createBlockFromCollection(const Collection & collection, const DataTypes &
size_t i = 0;
for (; i < tuple_size; ++i)
{
auto converted_field = convertFieldToTypeStrict(tuple[i], *block_types[i]);
auto converted_field = convertFieldToTypeStrict(tuple[i], *tuple_value_type[i], *block_types[i]);
if (!converted_field)
break;
tuple_values[i] = std::move(*converted_field);
@ -147,20 +157,28 @@ Block getSetElementsForConstantValue(const DataTypePtr & expression_type, const
if (lhs_type_depth == rhs_type_depth)
{
/// 1 in 1; (1, 2) in (1, 2); identity(tuple(tuple(tuple(1)))) in tuple(tuple(tuple(1))); etc.
Array array{value};
result_block = createBlockFromCollection(array, set_element_types, transform_null_in);
DataTypes value_types{value_type};
result_block = createBlockFromCollection(array, value_types, set_element_types, transform_null_in);
}
else if (lhs_type_depth + 1 == rhs_type_depth)
{
/// 1 in (1, 2); (1, 2) in ((1, 2), (3, 4))
WhichDataType rhs_which_type(value_type);
if (rhs_which_type.isArray())
result_block = createBlockFromCollection(value.get<const Array &>(), set_element_types, transform_null_in);
{
const DataTypeArray * value_array_type = assert_cast<const DataTypeArray *>(value_type.get());
size_t value_array_size = value.get<const Array &>().size();
DataTypes value_types(value_array_size, value_array_type->getNestedType());
result_block = createBlockFromCollection(value.get<const Array &>(), value_types, set_element_types, transform_null_in);
}
else if (rhs_which_type.isTuple())
result_block = createBlockFromCollection(value.get<const Tuple &>(), set_element_types, transform_null_in);
{
const DataTypeTuple * value_tuple_type = assert_cast<const DataTypeTuple *>(value_type.get());
const DataTypes & value_types = value_tuple_type->getElements();
result_block = createBlockFromCollection(value.get<const Tuple &>(), value_types, set_element_types, transform_null_in);
}
else
throw Exception(ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT,
"Unsupported type at the right-side of IN. Expected Array or Tuple. Actual {}",

View File

@ -24,8 +24,6 @@
#include <Poco/Util/XMLConfiguration.h>
#include <Poco/DOM/DOMParser.h>
#include <ranges>
namespace ProfileEvents
{
@ -93,6 +91,7 @@ BackupImpl::BackupImpl(
const std::optional<BackupInfo> & base_backup_info_,
std::shared_ptr<IBackupReader> reader_,
const ContextPtr & context_,
bool is_internal_backup_,
bool use_same_s3_credentials_for_base_backup_)
: backup_info(backup_info_)
, backup_name_for_logging(backup_info.toStringForLogging())
@ -101,7 +100,7 @@ BackupImpl::BackupImpl(
, open_mode(OpenMode::READ)
, reader(std::move(reader_))
, context(context_)
, is_internal_backup(false)
, is_internal_backup(is_internal_backup_)
, version(INITIAL_BACKUP_VERSION)
, base_backup_info(base_backup_info_)
, use_same_s3_credentials_for_base_backup(use_same_s3_credentials_for_base_backup_)
@ -256,6 +255,7 @@ std::shared_ptr<const IBackup> BackupImpl::getBaseBackupUnlocked() const
params.backup_info = *base_backup_info;
params.open_mode = OpenMode::READ;
params.context = context;
params.is_internal_backup = is_internal_backup;
/// use_same_s3_credentials_for_base_backup should be inherited for base backups
params.use_same_s3_credentials_for_base_backup = use_same_s3_credentials_for_base_backup;

View File

@ -40,6 +40,7 @@ public:
const std::optional<BackupInfo> & base_backup_info_,
std::shared_ptr<IBackupReader> reader_,
const ContextPtr & context_,
bool is_internal_backup_,
bool use_same_s3_credentials_for_base_backup_);
BackupImpl(

View File

@ -153,6 +153,7 @@ void registerBackupEngineAzureBlobStorage(BackupFactory & factory)
params.base_backup_info,
reader,
params.context,
params.is_internal_backup,
/* use_same_s3_credentials_for_base_backup*/ false);
}
else

View File

@ -119,6 +119,7 @@ void registerBackupEngineS3(BackupFactory & factory)
params.base_backup_info,
reader,
params.context,
params.is_internal_backup,
params.use_same_s3_credentials_for_base_backup);
}
else

View File

@ -177,6 +177,7 @@ void registerBackupEnginesFileAndDisk(BackupFactory & factory)
params.base_backup_info,
reader,
params.context,
params.is_internal_backup,
params.use_same_s3_credentials_for_base_backup);
}
else

View File

@ -44,13 +44,12 @@
#include <Parsers/ASTIdentifier.h>
#include <Parsers/ASTColumnDeclaration.h>
#include <Parsers/ASTFunction.h>
#include <Parsers/Kusto/ParserKQLStatement.h>
#include <Parsers/PRQL/ParserPRQLQuery.h>
#include <Parsers/Kusto/ParserKQLStatement.h>
#include <Parsers/Kusto/parseKQLQuery.h>
#include <Processors/Formats/Impl/NullFormat.h>
#include <Processors/Formats/IInputFormat.h>
#include <Processors/Formats/IOutputFormat.h>
#include <Processors/QueryPlan/QueryPlan.h>
#include <Processors/QueryPlan/BuildQueryPipelineSettings.h>
#include <Processors/QueryPlan/Optimizations/QueryPlanOptimizationSettings.h>

View File

@ -255,6 +255,17 @@ void HedgedConnections::sendCancel()
if (!sent_query || cancelled)
throw Exception(ErrorCodes::LOGICAL_ERROR, "Cannot cancel. Either no query sent or already cancelled.");
/// All hedged connections should be stopped, since otherwise before the
/// HedgedConnectionsFactory will be destroyed (that will happen from
/// QueryPipeline dtor) they could still do some work.
/// And not only this does not make sense, but it also could lead to
/// use-after-free of the current_thread, since the thread from which they
/// had been created differs from the thread where the dtor of
/// QueryPipeline will be called and the initial thread could be already
/// destroyed (especially when the system is under pressure).
if (hedged_connections_factory.hasEventsInProcess())
hedged_connections_factory.stopChoosingReplicas();
cancelled = true;
for (auto & offset_status : offset_states)

View File

@ -415,6 +415,15 @@ Value saveAllArenasMetric(AsynchronousMetricValues & values,
fmt::format("jemalloc.arenas.all.{}", metric_name));
}
template<typename Value>
Value saveJemallocProf(AsynchronousMetricValues & values,
const std::string & metric_name)
{
return saveJemallocMetricImpl<Value>(values,
fmt::format("prof.{}", metric_name),
fmt::format("jemalloc.prof.{}", metric_name));
}
}
#endif
@ -607,6 +616,7 @@ void AsynchronousMetrics::update(TimePoint update_time, bool force_update)
saveJemallocMetric<size_t>(new_values, "background_thread.num_threads");
saveJemallocMetric<uint64_t>(new_values, "background_thread.num_runs");
saveJemallocMetric<uint64_t>(new_values, "background_thread.run_intervals");
saveJemallocProf<size_t>(new_values, "active");
saveAllArenasMetric<size_t>(new_values, "pactive");
[[maybe_unused]] size_t je_malloc_pdirty = saveAllArenasMetric<size_t>(new_values, "pdirty");
[[maybe_unused]] size_t je_malloc_pmuzzy = saveAllArenasMetric<size_t>(new_values, "pmuzzy");

View File

@ -1,6 +1,7 @@
#include <Common/CurrentMetrics.h>
// clang-format off
/// Available metrics. Add something here as you wish.
/// If the metric is generic (i.e. not server specific)
/// it should be also added to src/Coordination/KeeperConstant.cpp

View File

@ -202,7 +202,10 @@ uint64_t readU64(std::string_view & sp)
{
SAFE_CHECK(sp.size() >= N, "underflow");
uint64_t x = 0;
memcpy(&x, sp.data(), N);
if constexpr (std::endian::native == std::endian::little)
memcpy(&x, sp.data(), N);
else
memcpy(reinterpret_cast<char*>(&x) + sizeof(uint64_t) - N, sp.data(), N);
sp.remove_prefix(N);
return x;
}

View File

@ -602,6 +602,8 @@
M(721, DEPRECATED_FUNCTION) \
M(722, ASYNC_LOAD_WAIT_FAILED) \
M(723, PARQUET_EXCEPTION) \
M(724, TOO_MANY_TABLES) \
M(725, TOO_MANY_DATABASES) \
\
M(900, DISTRIBUTED_CACHE_ERROR) \
M(901, CANNOT_USE_DISTRIBUTED_CACHE) \

View File

@ -1,8 +1,6 @@
#pragma once
#include <cstddef>
#include <cstdint>
#include <utility>
#include <mutex>
#include <string_view>
#include <vector>
@ -35,7 +33,7 @@ namespace ErrorCodes
struct Error
{
/// Number of times Exception with this ErrorCode had been throw.
/// Number of times Exception with this ErrorCode has been thrown.
Value count = 0;
/// Time of the last error.
UInt64 error_time_ms = 0;

161
src/Common/HilbertUtils.h Normal file
View File

@ -0,0 +1,161 @@
#pragma once
#include <Core/Types.h>
#include <Common/BitHelpers.h>
#include "base/types.h"
#include <Functions/hilbertDecode2DLUT.h>
#include <base/defines.h>
#include <array>
#include <set>
namespace HilbertDetails
{
struct Segment // represents [begin; end], all bounds are included
{
UInt64 begin;
UInt64 end;
};
}
/*
Given the range of values of hilbert code - and this function will return segments of the Hilbert curve
such that each of them lies in a whole domain (aka square)
0 1
0 00xxx 11xxx
| |
| |
_______________________________
| |
| |
| |
1 01xxx___________10xxx
Imagine a square, one side of which is a x-axis, other is a y-axis.
First approximation of the Hilbert curve is on the picture - U curve.
So we divide Hilbert Code Interval on 4 parts each of which is represented by a square
and look where the given interval [start, finish] is located:
[00xxxxxx | 01xxxxxx | 10xxxxxx | 11xxxxxx ]
1: [ ]
start = 0010111 end = 10111110
2: [ ] [ ]
If it contains a whole sector (that represents a domain=square),
then we take this range. In the example above - it is a sector [01000000, 01111111]
Then we dig into the recursion and check the remaining ranges.
Note that after the first call all other ranges in the recursion will have either start or finish on the end of a range,
so the complexity of the algorithm will be O(logN), where N is the maximum of hilbert code.
*/
template <typename F>
void segmentBinaryPartition(UInt64 start, UInt64 finish, UInt8 current_bits, F && callback)
{
if (current_bits == 0)
return;
const auto next_bits = current_bits - 2;
const auto history = current_bits == 64 ? 0 : (start >> current_bits) << current_bits;
const auto chunk_mask = 0b11;
const auto start_chunk = (start >> next_bits) & chunk_mask;
const auto finish_chunk = (finish >> next_bits) & chunk_mask;
auto construct_range = [next_bits, history](UInt64 chunk)
{
return HilbertDetails::Segment{
.begin = history + (chunk << next_bits),
.end = history + ((chunk + 1) << next_bits) - 1
};
};
if (start_chunk == finish_chunk)
{
if ((finish - start + 1) == (1 << next_bits)) // it means that [begin, end] is a range
{
callback(HilbertDetails::Segment{.begin = start, .end = finish});
return;
}
segmentBinaryPartition(start, finish, next_bits, callback);
return;
}
for (auto range_chunk = start_chunk + 1; range_chunk < finish_chunk; ++range_chunk)
{
callback(construct_range(range_chunk));
}
const auto start_range = construct_range(start_chunk);
if (start == start_range.begin)
{
callback(start_range);
}
else
{
segmentBinaryPartition(start, start_range.end, next_bits, callback);
}
const auto finish_range = construct_range(finish_chunk);
if (finish == finish_range.end)
{
callback(finish_range);
}
else
{
segmentBinaryPartition(finish_range.begin, finish, next_bits, callback);
}
}
// Given 2 points representing ends of the range of Hilbert Curve that lies in a whole domain.
// The are neighbour corners of some square - and the function returns ranges of both sides of this square
inline std::array<std::pair<UInt64, UInt64>, 2> createRangeFromCorners(UInt64 x1, UInt64 y1, UInt64 x2, UInt64 y2)
{
UInt64 dist_x = x1 > x2 ? x1 - x2 : x2 - x1;
UInt64 dist_y = y1 > y2 ? y1 - y2 : y2 - y1;
UInt64 range_size = std::max(dist_x, dist_y);
bool contains_minimum_vertice = x1 % (range_size + 1) == 0;
if (contains_minimum_vertice)
{
UInt64 x_min = std::min(x1, x2);
UInt64 y_min = std::min(y1, y2);
return {
std::pair<UInt64, UInt64>{x_min, x_min + range_size},
std::pair<UInt64, UInt64>{y_min, y_min + range_size}
};
}
else
{
UInt64 x_max = std::max(x1, x2);
UInt64 y_max = std::max(y1, y2);
chassert(x_max >= range_size);
chassert(y_max >= range_size);
return {
std::pair<UInt64, UInt64>{x_max - range_size, x_max},
std::pair<UInt64, UInt64>{y_max - range_size, y_max}
};
}
}
/** Unpack an interval of Hilbert curve to hyperrectangles covered by it across N dimensions.
*/
template <typename F>
void hilbertIntervalToHyperrectangles2D(UInt64 first, UInt64 last, F && callback)
{
const auto equal_bits_count = getLeadingZeroBits(last | first);
const auto even_equal_bits_count = equal_bits_count - equal_bits_count % 2;
segmentBinaryPartition(first, last, 64 - even_equal_bits_count, [&](HilbertDetails::Segment range)
{
auto interval1 = DB::FunctionHilbertDecode2DWIthLookupTableImpl<3>::decode(range.begin);
auto interval2 = DB::FunctionHilbertDecode2DWIthLookupTableImpl<3>::decode(range.end);
std::array<std::pair<UInt64, UInt64>, 2> unpacked_range = createRangeFromCorners(
std::get<0>(interval1), std::get<1>(interval1),
std::get<0>(interval2), std::get<1>(interval2));
callback(unpacked_range);
});
}

View File

@ -48,7 +48,7 @@ public:
/// HashFunction usually hashes the entire key and the found key will be equal the provided key. In such cases, use get(). It is also
/// possible to store other, non-hashed data in the key. In that case, the found key is potentially different from the provided key.
/// Then use getWithKey() to also return the found key including it's non-hashed data.
/// Then use getWithKey() to also return the found key including its non-hashed data.
virtual MappedPtr get(const Key & key) = 0;
virtual std::optional<KeyMapped> getWithKey(const Key &) = 0;

View File

@ -3,6 +3,7 @@
#include <Common/TraceSender.h>
// clang-format off
/// Available events. Add something here as you wish.
/// If the event is generic (i.e. not server specific)
/// it should be also added to src/Coordination/KeeperConstant.cpp
@ -14,6 +15,7 @@
M(QueriesWithSubqueries, "Count queries with all subqueries") \
M(SelectQueriesWithSubqueries, "Count SELECT queries with all subqueries") \
M(InsertQueriesWithSubqueries, "Count INSERT queries with all subqueries") \
M(SelectQueriesWithPrimaryKeyUsage, "Count SELECT queries which use the primary key to evaluate the WHERE condition") \
M(AsyncInsertQuery, "Same as InsertQuery, but only for asynchronous INSERT queries.") \
M(AsyncInsertBytes, "Data size in bytes of asynchronous INSERT queries.") \
M(AsyncInsertRows, "Number of rows inserted by asynchronous INSERT queries.") \

View File

@ -11,10 +11,10 @@
#include <Poco/Util/XMLConfiguration.h>
#include <boost/noncopyable.hpp>
#include <boost/intrusive/list.hpp>
#include <chrono>
#include <deque>
#include <queue>
#include <algorithm>
#include <functional>
#include <memory>
@ -30,6 +30,8 @@ namespace ErrorCodes
}
class ISchedulerNode;
class EventQueue;
using EventId = UInt64;
inline const Poco::Util::AbstractConfiguration & emptyConfig()
{
@ -82,6 +84,115 @@ struct SchedulerNodeInfo
}
};
/*
* Node of hierarchy for scheduling requests for resource. Base class for all
* kinds of scheduling elements (queues, policies, constraints and schedulers).
*
* Root node is a scheduler, which has it's thread to dequeue requests,
* execute requests (see ResourceRequest) and process events in a thread-safe manner.
* Immediate children of the scheduler represent independent resources.
* Each resource has it's own hierarchy to achieve required scheduling policies.
* Non-leaf nodes do not hold requests, but keep scheduling state
* (e.g. consumption history, amount of in-flight requests, etc).
* Leafs of hierarchy are queues capable of holding pending requests.
*
* scheduler (SchedulerRoot)
* / \
* constraint constraint (SemaphoreConstraint)
* | |
* policy policy (PriorityPolicy)
* / \ / \
* q1 q2 q3 q4 (FifoQueue)
*
* Dequeueing request from an inner node will dequeue request from one of active leaf-queues in its subtree.
* Node is considered to be active iff:
* - it has at least one pending request in one of leaves of it's subtree;
* - and enforced constraints, if any, are satisfied
* (e.g. amount of concurrent requests is not greater than some number).
*
* All methods must be called only from scheduler thread for thread-safety.
*/
class ISchedulerNode : public boost::intrusive::list_base_hook<>, private boost::noncopyable
{
public:
explicit ISchedulerNode(EventQueue * event_queue_, const Poco::Util::AbstractConfiguration & config = emptyConfig(), const String & config_prefix = {})
: event_queue(event_queue_)
, info(config, config_prefix)
{}
virtual ~ISchedulerNode() = default;
/// Checks if two nodes configuration is equal
virtual bool equals(ISchedulerNode * other)
{
return info.equals(other->info);
}
/// Attach new child
virtual void attachChild(const std::shared_ptr<ISchedulerNode> & child) = 0;
/// Detach and destroy child
virtual void removeChild(ISchedulerNode * child) = 0;
/// Get attached child by name
virtual ISchedulerNode * getChild(const String & child_name) = 0;
/// Activation of child due to the first pending request
/// Should be called on leaf node (i.e. queue) to propagate activation signal through chain to the root
virtual void activateChild(ISchedulerNode * child) = 0;
/// Returns true iff node is active
virtual bool isActive() = 0;
/// Returns number of active children
virtual size_t activeChildren() = 0;
/// Returns the first request to be executed as the first component of resulting pair.
/// The second pair component is `true` iff node is still active after dequeueing.
virtual std::pair<ResourceRequest *, bool> dequeueRequest() = 0;
/// Returns full path string using names of every parent
String getPath()
{
String result;
ISchedulerNode * ptr = this;
while (ptr->parent)
{
result = "/" + ptr->basename + result;
ptr = ptr->parent;
}
return result.empty() ? "/" : result;
}
/// Attach to a parent (used by attachChild)
virtual void setParent(ISchedulerNode * parent_)
{
parent = parent_;
}
protected:
/// Notify parents about the first pending request or constraint becoming satisfied.
/// Postponed to be handled in scheduler thread, so it is intended to be called from outside.
void scheduleActivation();
public:
EventQueue * const event_queue;
String basename;
SchedulerNodeInfo info;
ISchedulerNode * parent = nullptr;
EventId activation_event_id = 0; // Valid for `ISchedulerNode` placed in EventQueue::activations
/// Introspection
std::atomic<UInt64> dequeued_requests{0};
std::atomic<UInt64> canceled_requests{0};
std::atomic<ResourceCost> dequeued_cost{0};
std::atomic<ResourceCost> canceled_cost{0};
std::atomic<UInt64> busy_periods{0};
};
using SchedulerNodePtr = std::shared_ptr<ISchedulerNode>;
/*
* Simple waitable thread-safe FIFO task queue.
* Intended to hold postponed events for later handling (usually by scheduler thread).
@ -89,57 +200,70 @@ struct SchedulerNodeInfo
class EventQueue
{
public:
using Event = std::function<void()>;
using Task = std::function<void()>;
static constexpr EventId not_postponed = 0;
using TimePoint = std::chrono::system_clock::time_point;
using Duration = std::chrono::system_clock::duration;
static constexpr UInt64 not_postponed = 0;
struct Event
{
const EventId event_id;
Task task;
Event(EventId event_id_, Task && task_)
: event_id(event_id_)
, task(std::move(task_))
{}
};
struct Postponed
{
TimePoint key;
UInt64 id; // for canceling
std::unique_ptr<Event> event;
EventId event_id; // for canceling
std::unique_ptr<Task> task;
Postponed(TimePoint key_, UInt64 id_, Event && event_)
Postponed(TimePoint key_, EventId event_id_, Task && task_)
: key(key_)
, id(id_)
, event(std::make_unique<Event>(std::move(event_)))
, event_id(event_id_)
, task(std::make_unique<Task>(std::move(task_)))
{}
bool operator<(const Postponed & rhs) const
{
return std::tie(key, id) > std::tie(rhs.key, rhs.id); // reversed for min-heap
return std::tie(key, event_id) > std::tie(rhs.key, rhs.event_id); // reversed for min-heap
}
};
/// Add an `event` to be processed after `until` time point.
/// Returns a unique id for canceling.
[[nodiscard]] UInt64 postpone(TimePoint until, Event && event)
/// Returns a unique event id for canceling.
[[nodiscard]] EventId postpone(TimePoint until, Task && task)
{
std::unique_lock lock{mutex};
if (postponed.empty() || until < postponed.front().key)
pending.notify_one();
auto id = ++last_id;
postponed.emplace_back(until, id, std::move(event));
auto event_id = ++last_event_id;
postponed.emplace_back(until, event_id, std::move(task));
std::push_heap(postponed.begin(), postponed.end());
return id;
return event_id;
}
/// Cancel a postponed event using its unique id.
/// NOTE: Only postponed events can be canceled.
/// NOTE: If you need to cancel enqueued event, consider doing your actions inside another enqueued
/// NOTE: event instead. This ensures that all previous events are processed.
bool cancelPostponed(UInt64 postponed_id)
bool cancelPostponed(EventId postponed_event_id)
{
if (postponed_id == not_postponed)
if (postponed_event_id == not_postponed)
return false;
std::unique_lock lock{mutex};
for (auto i = postponed.begin(), e = postponed.end(); i != e; ++i)
{
if (i->id == postponed_id)
if (i->event_id == postponed_event_id)
{
postponed.erase(i);
// It is O(n), but we do not expect either big heaps or frequent cancels. So it is fine.
// It is O(n), but we do not expect neither big heaps nor frequent cancels. So it is fine.
std::make_heap(postponed.begin(), postponed.end());
return true;
}
@ -148,11 +272,23 @@ public:
}
/// Add an `event` for immediate processing
void enqueue(Event && event)
void enqueue(Task && task)
{
std::unique_lock lock{mutex};
bool was_empty = queue.empty();
queue.emplace_back(event);
bool was_empty = events.empty() && activations.empty();
auto event_id = ++last_event_id;
events.emplace_back(event_id, std::move(task));
if (was_empty)
pending.notify_one();
}
/// Add an activation `event` for immediate processing. Activations use a separate queue for performance reasons.
void enqueueActivation(ISchedulerNode * node)
{
std::unique_lock lock{mutex};
bool was_empty = events.empty() && activations.empty();
node->activation_event_id = ++last_event_id;
activations.push_back(*node);
if (was_empty)
pending.notify_one();
}
@ -163,7 +299,7 @@ public:
bool forceProcess()
{
std::unique_lock lock{mutex};
if (!queue.empty())
if (!events.empty() || !activations.empty())
{
processQueue(std::move(lock));
return true;
@ -181,7 +317,7 @@ public:
bool tryProcess()
{
std::unique_lock lock{mutex};
if (!queue.empty())
if (!events.empty() || !activations.empty())
{
processQueue(std::move(lock));
return true;
@ -205,7 +341,7 @@ public:
std::unique_lock lock{mutex};
while (true)
{
if (!queue.empty())
if (!events.empty() || !activations.empty())
{
processQueue(std::move(lock));
return;
@ -269,141 +405,69 @@ private:
void processQueue(std::unique_lock<std::mutex> && lock)
{
Event event = std::move(queue.front());
queue.pop_front();
if (events.empty())
{
processActivation(std::move(lock));
return;
}
if (activations.empty())
{
processEvent(std::move(lock));
return;
}
if (activations.front().activation_event_id < events.front().event_id)
processActivation(std::move(lock));
else
processEvent(std::move(lock));
}
void processActivation(std::unique_lock<std::mutex> && lock)
{
ISchedulerNode * node = &activations.front();
activations.pop_front();
node->activation_event_id = 0;
lock.unlock(); // do not hold queue mutex while processing events
event();
node->parent->activateChild(node);
}
void processEvent(std::unique_lock<std::mutex> && lock)
{
Task task = std::move(events.front().task);
events.pop_front();
lock.unlock(); // do not hold queue mutex while processing events
task();
}
void processPostponed(std::unique_lock<std::mutex> && lock)
{
Event event = std::move(*postponed.front().event);
Task task = std::move(*postponed.front().task);
std::pop_heap(postponed.begin(), postponed.end());
postponed.pop_back();
lock.unlock(); // do not hold queue mutex while processing events
event();
task();
}
std::mutex mutex;
std::condition_variable pending;
std::deque<Event> queue;
// `events` and `activations` logically represent one ordered queue. To preserve the common order we use `EventId`
// Activations are stored in a separate queue for performance reasons (mostly to avoid any allocations)
std::deque<Event> events;
boost::intrusive::list<ISchedulerNode> activations;
std::vector<Postponed> postponed;
UInt64 last_id = 0;
EventId last_event_id = 0;
std::atomic<TimePoint> manual_time{TimePoint()}; // for tests only
};
/*
* Node of hierarchy for scheduling requests for resource. Base class for all
* kinds of scheduling elements (queues, policies, constraints and schedulers).
*
* Root node is a scheduler, which has it's thread to dequeue requests,
* execute requests (see ResourceRequest) and process events in a thread-safe manner.
* Immediate children of the scheduler represent independent resources.
* Each resource has it's own hierarchy to achieve required scheduling policies.
* Non-leaf nodes do not hold requests, but keep scheduling state
* (e.g. consumption history, amount of in-flight requests, etc).
* Leafs of hierarchy are queues capable of holding pending requests.
*
* scheduler (SchedulerRoot)
* / \
* constraint constraint (SemaphoreConstraint)
* | |
* policy policy (PriorityPolicy)
* / \ / \
* q1 q2 q3 q4 (FifoQueue)
*
* Dequeueing request from an inner node will dequeue request from one of active leaf-queues in its subtree.
* Node is considered to be active iff:
* - it has at least one pending request in one of leaves of it's subtree;
* - and enforced constraints, if any, are satisfied
* (e.g. amount of concurrent requests is not greater than some number).
*
* All methods must be called only from scheduler thread for thread-safety.
*/
class ISchedulerNode : private boost::noncopyable
inline void ISchedulerNode::scheduleActivation()
{
public:
explicit ISchedulerNode(EventQueue * event_queue_, const Poco::Util::AbstractConfiguration & config = emptyConfig(), const String & config_prefix = {})
: event_queue(event_queue_)
, info(config, config_prefix)
{}
virtual ~ISchedulerNode() = default;
/// Checks if two nodes configuration is equal
virtual bool equals(ISchedulerNode * other)
if (likely(parent))
{
return info.equals(other->info);
// The same as `enqueue([this] { parent->activateChild(this); });` but faster
event_queue->enqueueActivation(this);
}
/// Attach new child
virtual void attachChild(const std::shared_ptr<ISchedulerNode> & child) = 0;
/// Detach and destroy child
virtual void removeChild(ISchedulerNode * child) = 0;
/// Get attached child by name
virtual ISchedulerNode * getChild(const String & child_name) = 0;
/// Activation of child due to the first pending request
/// Should be called on leaf node (i.e. queue) to propagate activation signal through chain to the root
virtual void activateChild(ISchedulerNode * child) = 0;
/// Returns true iff node is active
virtual bool isActive() = 0;
/// Returns number of active children
virtual size_t activeChildren() = 0;
/// Returns the first request to be executed as the first component of resulting pair.
/// The second pair component is `true` iff node is still active after dequeueing.
virtual std::pair<ResourceRequest *, bool> dequeueRequest() = 0;
/// Returns full path string using names of every parent
String getPath()
{
String result;
ISchedulerNode * ptr = this;
while (ptr->parent)
{
result = "/" + ptr->basename + result;
ptr = ptr->parent;
}
return result.empty() ? "/" : result;
}
/// Attach to a parent (used by attachChild)
virtual void setParent(ISchedulerNode * parent_)
{
parent = parent_;
}
protected:
/// Notify parents about the first pending request or constraint becoming satisfied.
/// Postponed to be handled in scheduler thread, so it is intended to be called from outside.
void scheduleActivation()
{
if (likely(parent))
{
event_queue->enqueue([this] { parent->activateChild(this); });
}
}
public:
EventQueue * const event_queue;
String basename;
SchedulerNodeInfo info;
ISchedulerNode * parent = nullptr;
/// Introspection
std::atomic<UInt64> dequeued_requests{0};
std::atomic<UInt64> canceled_requests{0};
std::atomic<ResourceCost> dequeued_cost{0};
std::atomic<ResourceCost> canceled_cost{0};
std::atomic<UInt64> busy_periods{0};
};
using SchedulerNodePtr = std::shared_ptr<ISchedulerNode>;
}
}

View File

@ -0,0 +1,143 @@
#include <chrono>
#include <gtest/gtest.h>
#include <Common/Scheduler/ISchedulerNode.h>
using namespace DB;
class FakeSchedulerNode : public ISchedulerNode
{
public:
explicit FakeSchedulerNode(String & log_, EventQueue * event_queue_, const Poco::Util::AbstractConfiguration & config = emptyConfig(), const String & config_prefix = {})
: ISchedulerNode(event_queue_, config, config_prefix)
, log(log_)
{}
void attachChild(const SchedulerNodePtr & child) override
{
log += " +" + child->basename;
}
void removeChild(ISchedulerNode * child) override
{
log += " -" + child->basename;
}
ISchedulerNode * getChild(const String & /* child_name */) override
{
return nullptr;
}
void activateChild(ISchedulerNode * child) override
{
log += " A" + child->basename;
}
bool isActive() override
{
return false;
}
size_t activeChildren() override
{
return 0;
}
std::pair<ResourceRequest *, bool> dequeueRequest() override
{
log += " D";
return {nullptr, false};
}
private:
String & log;
};
struct QueueTest {
String log;
EventQueue event_queue;
FakeSchedulerNode root_node;
QueueTest()
: root_node(log, &event_queue)
{}
SchedulerNodePtr makeNode(const String & name)
{
auto node = std::make_shared<FakeSchedulerNode>(log, &event_queue);
node->basename = name;
node->setParent(&root_node);
return std::static_pointer_cast<ISchedulerNode>(node);
}
void process(EventQueue::TimePoint now, const String & expected_log, size_t limit = size_t(-1))
{
event_queue.setManualTime(now);
for (;limit > 0; limit--)
{
if (!event_queue.tryProcess())
break;
}
EXPECT_EQ(log, expected_log);
log.clear();
}
void activate(const SchedulerNodePtr & node)
{
event_queue.enqueueActivation(node.get());
}
void event(const String & text)
{
event_queue.enqueue([this, text] { log += " " + text; });
}
EventId postpone(EventQueue::TimePoint until, const String & text)
{
return event_queue.postpone(until, [this, text] { log += " " + text; });
}
void cancel(EventId event_id)
{
event_queue.cancelPostponed(event_id);
}
};
TEST(SchedulerEventQueue, Smoke)
{
QueueTest t;
using namespace std::chrono_literals;
EventQueue::TimePoint start = std::chrono::system_clock::now();
t.process(start, "", 0);
// Activations
auto node1 = t.makeNode("1");
auto node2 = t.makeNode("2");
t.activate(node2);
t.activate(node1);
t.process(start + 42s, " A2 A1");
// Events
t.event("E1");
t.event("E2");
t.process(start + 100s, " E1 E2");
// Postponed events
t.postpone(start + 200s, "P200");
auto p190 = t.postpone(start + 200s, "P190");
t.postpone(start + 150s, "P150");
t.postpone(start + 175s, "P175");
t.process(start + 180s, " P150 P175");
t.event("E3");
t.cancel(p190);
t.process(start + 300s, " E3 P200");
// Ordering of events and activations
t.event("E1");
t.activate(node1);
t.event("E2");
t.activate(node2);
t.process(start + 300s, " E1 A1 E2 A2");
}

View File

@ -5,8 +5,6 @@
#include <Common/Scheduler/Nodes/FairPolicy.h>
#include <Common/Scheduler/Nodes/ThrottlerConstraint.h>
#include "Common/Scheduler/ISchedulerNode.h"
#include "Common/Scheduler/ResourceRequest.h"
using namespace DB;

View File

@ -85,9 +85,18 @@ StatusFile::StatusFile(std::string path_, FillFunction fill_)
/// Write information about current server instance to the file.
WriteBufferFromFileDescriptor out(fd, 1024);
fill(out);
/// Finalize here to avoid throwing exceptions in destructor.
out.finalize();
try
{
fill(out);
/// Finalize here to avoid throwing exceptions in destructor.
out.finalize();
}
catch (...)
{
/// Finalize in case of exception to avoid throwing exceptions in destructor
out.finalize();
throw;
}
}
catch (...)
{

View File

@ -1,5 +1,6 @@
#include <Interpreters/AsynchronousMetricLog.h>
#include <Interpreters/CrashLog.h>
#include <Interpreters/ErrorLog.h>
#include <Interpreters/MetricLog.h>
#include <Interpreters/OpenTelemetrySpanLog.h>
#include <Interpreters/PartLog.h>

View File

@ -1,9 +1,7 @@
#pragma once
#include <atomic>
#include <condition_variable>
#include <memory>
#include <thread>
#include <vector>
#include <base/types.h>
@ -32,7 +30,8 @@
M(FilesystemReadPrefetchesLogElement) \
M(AsynchronousInsertLogElement) \
M(BackupLogElement) \
M(BlobStorageLogElement)
M(BlobStorageLogElement) \
M(ErrorLogElement)
namespace Poco
{

View File

@ -609,7 +609,10 @@ void KeeperStorage::UncommittedState::commit(int64_t commit_zxid)
uncommitted_auth.pop_front();
if (uncommitted_auth.empty())
session_and_auth.erase(add_auth->session_id);
}
else if (auto * close_session = std::get_if<CloseSessionDelta>(&front_delta.operation))
{
closed_sessions.erase(close_session->session_id);
}
deltas.pop_front();
@ -682,6 +685,10 @@ void KeeperStorage::UncommittedState::rollback(int64_t rollback_zxid)
session_and_auth.erase(add_auth->session_id);
}
}
else if (auto * close_session = std::get_if<CloseSessionDelta>(&delta_it->operation))
{
closed_sessions.erase(close_session->session_id);
}
}
if (delta_it == deltas.rend())
@ -878,6 +885,10 @@ Coordination::Error KeeperStorage::commit(int64_t commit_zxid)
session_and_auth[operation.session_id].emplace_back(std::move(operation.auth_id));
return Coordination::Error::ZOK;
}
else if constexpr (std::same_as<DeltaType, KeeperStorage::CloseSessionDelta>)
{
return Coordination::Error::ZOK;
}
else
{
// shouldn't be called in any process functions
@ -2366,12 +2377,15 @@ void KeeperStorage::preprocessRequest(
ephemerals.erase(session_ephemerals);
}
new_deltas.emplace_back(transaction.zxid, CloseSessionDelta{session_id});
uncommitted_state.closed_sessions.insert(session_id);
new_digest = calculateNodesDigest(new_digest, new_deltas);
return;
}
if (check_acl && !request_processor->checkAuth(*this, session_id, false))
if ((check_acl && !request_processor->checkAuth(*this, session_id, false)) ||
uncommitted_state.closed_sessions.contains(session_id)) // Is session closed but not committed yet
{
uncommitted_state.deltas.emplace_back(new_last_zxid, Coordination::Error::ZNOAUTH);
return;

View File

@ -314,8 +314,13 @@ public:
AuthID auth_id;
};
struct CloseSessionDelta
{
int64_t session_id;
};
using Operation = std::
variant<CreateNodeDelta, RemoveNodeDelta, UpdateNodeDelta, SetACLDelta, AddAuthDelta, ErrorDelta, SubDeltaEnd, FailedMultiDelta>;
variant<CreateNodeDelta, RemoveNodeDelta, UpdateNodeDelta, SetACLDelta, AddAuthDelta, ErrorDelta, SubDeltaEnd, FailedMultiDelta, CloseSessionDelta>;
struct Delta
{
@ -351,6 +356,7 @@ public:
std::shared_ptr<Node> tryGetNodeFromStorage(StringRef path) const;
std::unordered_map<int64_t, std::list<const AuthID *>> session_and_auth;
std::unordered_set<int64_t> closed_sessions;
struct UncommittedNode
{

View File

@ -2019,6 +2019,67 @@ TEST_P(CoordinationTest, TestCreateNodeWithAuthSchemeForAclWhenAuthIsPrecommitte
EXPECT_EQ(acls[0].permissions, 31);
}
TEST_P(CoordinationTest, TestPreprocessWhenCloseSessionIsPrecommitted)
{
using namespace Coordination;
using namespace DB;
ChangelogDirTest snapshots("./snapshots");
setSnapshotDirectory("./snapshots");
ResponsesQueue queue(std::numeric_limits<size_t>::max());
SnapshotsQueue snapshots_queue{1};
int64_t session_id = 1;
size_t term = 0;
auto state_machine = std::make_shared<KeeperStateMachine>(queue, snapshots_queue, keeper_context, nullptr);
state_machine->init();
auto & storage = state_machine->getStorageUnsafe();
const auto & uncommitted_state = storage.uncommitted_state;
// Create first node for the session
String node_path_1 = "/node_1";
std::shared_ptr<ZooKeeperCreateRequest> create_req_1 = std::make_shared<ZooKeeperCreateRequest>();
create_req_1->path = node_path_1;
auto create_entry_1 = getLogEntryFromZKRequest(term, session_id, state_machine->getNextZxid(), create_req_1);
state_machine->pre_commit(1, create_entry_1->get_buf());
EXPECT_TRUE(uncommitted_state.nodes.contains(node_path_1));
state_machine->commit(1, create_entry_1->get_buf());
EXPECT_TRUE(storage.container.contains(node_path_1));
// Close session
std::shared_ptr<ZooKeeperCloseRequest> close_req = std::make_shared<ZooKeeperCloseRequest>();
auto close_entry = getLogEntryFromZKRequest(term, session_id, state_machine->getNextZxid(), close_req);
// Pre-commit close session
state_machine->pre_commit(2, close_entry->get_buf());
// Try to create second node after close session is pre-committed
String node_path_2 = "/node_2";
std::shared_ptr<ZooKeeperCreateRequest> create_req_2 = std::make_shared<ZooKeeperCreateRequest>();
create_req_2->path = node_path_2;
auto create_entry_2 = getLogEntryFromZKRequest(term, session_id, state_machine->getNextZxid(), create_req_2);
// Pre-commit creating second node
state_machine->pre_commit(3, create_entry_2->get_buf());
// Second node wasn't created
EXPECT_FALSE(uncommitted_state.nodes.contains(node_path_2));
// Rollback pre-committed closing session
state_machine->rollback(3, create_entry_2->get_buf());
state_machine->rollback(2, close_entry->get_buf());
// Pre-commit creating second node
state_machine->pre_commit(2, create_entry_2->get_buf());
// Now second node was created
EXPECT_TRUE(uncommitted_state.nodes.contains(node_path_2));
state_machine->commit(2, create_entry_2->get_buf());
EXPECT_TRUE(storage.container.contains(node_path_1));
EXPECT_TRUE(storage.container.contains(node_path_2));
}
TEST_P(CoordinationTest, TestSetACLWithAuthSchemeForAclWhenAuthIsPrecommitted)
{
using namespace Coordination;

View File

@ -14,6 +14,7 @@ class AbstractConfiguration;
namespace DB
{
// clang-format off
#define SERVER_SETTINGS(M, ALIAS) \
M(Bool, show_addresses_in_stack_traces, true, "If it is set true will show addresses in stack traces", 0) \
M(Bool, shutdown_wait_unfinished_queries, false, "If set true ClickHouse will wait for running queries finish before shutdown.", 0) \
@ -85,10 +86,12 @@ namespace DB
M(Double, index_mark_cache_size_ratio, DEFAULT_INDEX_MARK_CACHE_SIZE_RATIO, "The size of the protected queue in the secondary index mark cache relative to the cache's total size.", 0) \
M(UInt64, page_cache_chunk_size, 2 << 20, "Bytes per chunk in userspace page cache. Rounded up to a multiple of page size (typically 4 KiB) or huge page size (typically 2 MiB, only if page_cache_use_thp is enabled).", 0) \
M(UInt64, page_cache_mmap_size, 1 << 30, "Bytes per memory mapping in userspace page cache. Not important.", 0) \
M(UInt64, page_cache_size, 10ul << 30, "Amount of virtual memory to map for userspace page cache. If page_cache_use_madv_free is enabled, it's recommended to set this higher than the machine's RAM size. Use 0 to disable userspace page cache.", 0) \
M(UInt64, page_cache_size, 0, "Amount of virtual memory to map for userspace page cache. If page_cache_use_madv_free is enabled, it's recommended to set this higher than the machine's RAM size. Use 0 to disable userspace page cache.", 0) \
M(Bool, page_cache_use_madv_free, DBMS_DEFAULT_PAGE_CACHE_USE_MADV_FREE, "If true, the userspace page cache will allow the OS to automatically reclaim memory from the cache on memory pressure (using MADV_FREE).", 0) \
M(Bool, page_cache_use_transparent_huge_pages, true, "Userspace will attempt to use transparent huge pages on Linux. This is best-effort.", 0) \
M(UInt64, mmap_cache_size, DEFAULT_MMAP_CACHE_MAX_SIZE, "A cache for mmapped files.", 0) \
M(UInt64, compiled_expression_cache_size, DEFAULT_COMPILED_EXPRESSION_CACHE_MAX_SIZE, "Byte size of compiled expressions cache.", 0) \
M(UInt64, compiled_expression_cache_elements_size, DEFAULT_COMPILED_EXPRESSION_CACHE_MAX_ENTRIES, "Maximum entries in compiled expressions cache.", 0) \
\
M(Bool, disable_internal_dns_cache, false, "Disable internal DNS caching at all.", 0) \
M(UInt64, dns_cache_max_entries, 10000, "Internal DNS cache max entries.", 0) \
@ -102,6 +105,8 @@ namespace DB
M(UInt64, max_dictionary_num_to_warn, 1000lu, "If the number of dictionaries is greater than this value, the server will create a warning that will displayed to user.", 0) \
M(UInt64, max_database_num_to_warn, 1000lu, "If the number of databases is greater than this value, the server will create a warning that will displayed to user.", 0) \
M(UInt64, max_part_num_to_warn, 100000lu, "If the number of parts is greater than this value, the server will create a warning that will displayed to user.", 0) \
M(UInt64, max_table_num_to_throw, 0lu, "If number of tables is greater than this value, server will throw an exception. 0 means no limitation. View, remote tables, dictionary, system tables are not counted. Only count table in Atomic/Ordinary/Replicated/Lazy database engine.", 0) \
M(UInt64, max_database_num_to_throw, 0lu, "If number of databases is greater than this value, server will throw an exception. 0 means no limitation.", 0) \
M(UInt64, concurrent_threads_soft_limit_num, 0, "Sets how many concurrent thread can be allocated before applying CPU pressure. Zero means unlimited.", 0) \
M(UInt64, concurrent_threads_soft_limit_ratio_to_cores, 0, "Same as concurrent_threads_soft_limit_num, but with ratio to cores.", 0) \
\
@ -146,6 +151,8 @@ namespace DB
M(UInt64, global_profiler_real_time_period_ns, 0, "Period for real clock timer of global profiler (in nanoseconds). Set 0 value to turn off the real clock global profiler. Recommended value is at least 10000000 (100 times a second) for single queries or 1000000000 (once a second) for cluster-wide profiling.", 0) \
M(UInt64, global_profiler_cpu_time_period_ns, 0, "Period for CPU clock timer of global profiler (in nanoseconds). Set 0 value to turn off the CPU clock global profiler. Recommended value is at least 10000000 (100 times a second) for single queries or 1000000000 (once a second) for cluster-wide profiling.", 0) \
M(Bool, enable_azure_sdk_logging, false, "Enables logging from Azure sdk", 0) \
M(String, merge_workload, "default", "Name of workload to be used to access resources for all merges (may be overridden by a merge tree setting)", 0) \
M(String, mutation_workload, "default", "Name of workload to be used to access resources for all mutations (may be overridden by a merge tree setting)", 0) \
M(Double, gwp_asan_force_sample_probability, 0, "Probability that an allocation from specific places will be sampled by GWP Asan (i.e. PODArray allocations)", 0) \
/// If you add a setting which can be updated at runtime, please update 'changeable_settings' map in StorageSystemServerSettings.cpp

View File

@ -31,6 +31,7 @@ class IColumn;
* for tracking settings changes in different versions and for special `compatibility` setting to work correctly.
*/
// clang-format off
#define COMMON_SETTINGS(M, ALIAS) \
M(Dialect, dialect, Dialect::clickhouse, "Which dialect will be used to parse query", 0)\
M(UInt64, min_compress_block_size, 65536, "The actual size of the block to compress, if the uncompressed data less than max_compress_block_size is no less than this value and no less than the volume of data for one mark.", 0) \
@ -933,6 +934,7 @@ class IColumn;
M(Int64, prefer_warmed_unmerged_parts_seconds, 0, "Only available in ClickHouse Cloud. If a merged part is less than this many seconds old and is not pre-warmed (see cache_populated_by_fetch), but all its source parts are available and pre-warmed, SELECT queries will read from those parts instead. Only for ReplicatedMergeTree. Note that this only checks whether CacheWarmer processed the part; if the part was fetched into cache by something else, it'll still be considered cold until CacheWarmer gets to it; if it was warmed, then evicted from cache, it'll still be considered warm.", 0) \
M(Bool, iceberg_engine_ignore_schema_evolution, false, "Ignore schema evolution in Iceberg table engine and read all data using latest schema saved on table creation. Note that it can lead to incorrect result", 0) \
M(Bool, allow_deprecated_error_prone_window_functions, false, "Allow usage of deprecated error prone window functions (neighbor, runningAccumulate, runningDifferenceStartingWithFirstValue, runningDifference)", 0) \
M(Bool, allow_deprecated_snowflake_conversion_functions, false, "Enables deprecated functions snowflakeToDateTime[64] and dateTime[64]ToSnowflake.", 0) \
// End of COMMON_SETTINGS
// Please add settings related to formats into the FORMAT_FACTORY_SETTINGS, move obsolete settings to OBSOLETE_SETTINGS and obsolete format settings to OBSOLETE_FORMAT_SETTINGS.
@ -1144,7 +1146,9 @@ class IColumn;
M(UInt64, output_format_pretty_max_value_width, 10000, "Maximum width of value to display in Pretty formats. If greater - it will be cut.", 0) \
M(UInt64, output_format_pretty_max_value_width_apply_for_single_value, false, "Only cut values (see the `output_format_pretty_max_value_width` setting) when it is not a single value in a block. Otherwise output it entirely, which is useful for the `SHOW CREATE TABLE` query.", 0) \
M(UInt64Auto, output_format_pretty_color, "auto", "Use ANSI escape sequences in Pretty formats. 0 - disabled, 1 - enabled, 'auto' - enabled if a terminal.", 0) \
M(String, output_format_pretty_grid_charset, "UTF-8", "Charset for printing grid borders. Available charsets: ASCII, UTF-8 (default one).", 0) \
M(String, output_format_pretty_grid_charset, "UTF-8", "Charset for printing grid borders. Available charsets: ASCII, UTF-8 (default one).", 0) \
M(UInt64, output_format_pretty_display_footer_column_names, true, "Display column names in the footer if there are 999 or more rows.", 0) \
M(UInt64, output_format_pretty_display_footer_column_names_min_rows, 50, "Sets the minimum threshold value of rows for which to enable displaying column names in the footer. 50 (default)", 0) \
M(UInt64, output_format_parquet_row_group_size, 1000000, "Target row group size in rows.", 0) \
M(UInt64, output_format_parquet_row_group_size_bytes, 512 * 1024 * 1024, "Target row group size in bytes, before compression.", 0) \
M(Bool, output_format_parquet_string_as_string, true, "Use Parquet String type instead of Binary for String columns.", 0) \

View File

@ -75,6 +75,7 @@ namespace SettingsChangesHistory
using SettingsChanges = std::vector<SettingChange>;
}
// clang-format off
/// History of settings changes that controls some backward incompatible changes
/// across all ClickHouse versions. It maps ClickHouse version to settings changes that were done
/// in this version. This history contains both changes to existing settings and newly added settings.
@ -101,6 +102,7 @@ static const std::map<ClickHouseVersion, SettingsChangesHistory::SettingsChanges
{"input_format_parquet_max_block_size", 8192, DEFAULT_BLOCK_SIZE, "Increase block size for parquet reader."},
{"input_format_parquet_prefer_block_bytes", 0, DEFAULT_BLOCK_SIZE * 256, "Average block bytes output by parquet reader."},
{"enable_blob_storage_log", true, true, "Write information about blob storage operations to system.blob_storage_log table"},
{"allow_deprecated_snowflake_conversion_functions", true, false, "Disabled deprecated functions snowflakeToDateTime[64] and dateTime[64]ToSnowflake."},
{"allow_statistic_optimize", false, false, "Old setting which popped up here being renamed."},
{"allow_experimental_statistic", false, false, "Old setting which popped up here being renamed."},
{"allow_statistics_optimize", false, false, "The setting was renamed. The previous name is `allow_statistic_optimize`."},
@ -108,6 +110,8 @@ static const std::map<ClickHouseVersion, SettingsChangesHistory::SettingsChanges
{"enable_vertical_final", false, true, "Enable vertical final by default again after fixing bug"},
{"parallel_replicas_custom_key_range_lower", 0, 0, "Add settings to control the range filter when using parallel replicas with dynamic shards"},
{"parallel_replicas_custom_key_range_upper", 0, 0, "Add settings to control the range filter when using parallel replicas with dynamic shards. A value of 0 disables the upper limit"},
{"output_format_pretty_display_footer_column_names", 0, 1, "Add a setting to display column names in the footer if there are many rows. Threshold value is controlled by output_format_pretty_display_footer_column_names_min_rows."},
{"output_format_pretty_display_footer_column_names_min_rows", 0, 50, "Add a setting to control the threshold value for setting output_format_pretty_display_footer_column_names_min_rows. Default 50."},
{"output_format_csv_serialize_tuple_into_separate_columns", true, true, "A new way of how interpret tuples in CSV format was added."},
{"input_format_csv_deserialize_separate_columns_into_tuple", true, true, "A new way of how interpret tuples in CSV format was added."},
{"input_format_csv_try_infer_strings_from_quoted_tuples", true, true, "A new way of how interpret tuples in CSV format was added."},

View File

@ -146,7 +146,7 @@ void SerializationVariantElement::deserializeBinaryBulkWithMultipleStreams(
}
/// If we started to read a new column, reinitialize variant column in deserialization state.
if (!variant_element_state->variant || result_column->empty())
if (!variant_element_state->variant || mutable_column->empty())
{
variant_element_state->variant = mutable_column->cloneEmpty();

View File

@ -186,6 +186,7 @@ void DatabaseLazy::attachTable(ContextPtr /* context_ */, const String & table_n
throw Exception(ErrorCodes::TABLE_ALREADY_EXISTS, "Table {}.{} already exists.", backQuote(database_name), backQuote(table_name));
it->second.expiration_iterator = cache_expiration_queue.emplace(cache_expiration_queue.end(), current_time, table_name);
CurrentMetrics::add(CurrentMetrics::AttachedTable, 1);
}
@ -202,6 +203,7 @@ StoragePtr DatabaseLazy::detachTable(ContextPtr /* context */, const String & ta
if (it->second.expiration_iterator != cache_expiration_queue.end())
cache_expiration_queue.erase(it->second.expiration_iterator);
tables_cache.erase(it);
CurrentMetrics::sub(CurrentMetrics::AttachedTable, 1);
}
return res;

View File

@ -73,9 +73,10 @@ zkutil::ZooKeeperPtr DatabaseReplicated::getZooKeeper() const
return getContext()->getZooKeeper();
}
static inline String getHostID(ContextPtr global_context, const UUID & db_uuid)
static inline String getHostID(ContextPtr global_context, const UUID & db_uuid, bool secure)
{
return Cluster::Address::toString(getFQDNOrHostName(), global_context->getTCPPort()) + ':' + toString(db_uuid);
UInt16 port = secure ? global_context->getTCPPortSecure().value_or(DBMS_DEFAULT_SECURE_PORT) : global_context->getTCPPort();
return Cluster::Address::toString(getFQDNOrHostName(), port) + ':' + toString(db_uuid);
}
static inline UInt64 getMetadataHash(const String & table_name, const String & metadata)
@ -415,8 +416,10 @@ void DatabaseReplicated::tryConnectToZooKeeperAndInitDatabase(LoadingStrictnessL
return;
}
String host_id = getHostID(getContext(), db_uuid);
if (is_create_query || replica_host_id != host_id)
String host_id = getHostID(getContext(), db_uuid, cluster_auth_info.cluster_secure_connection);
String host_id_default = getHostID(getContext(), db_uuid, false);
if (is_create_query || (replica_host_id != host_id && replica_host_id != host_id_default))
{
throw Exception(
ErrorCodes::REPLICA_ALREADY_EXISTS,
@ -424,6 +427,14 @@ void DatabaseReplicated::tryConnectToZooKeeperAndInitDatabase(LoadingStrictnessL
replica_name, shard_name, zookeeper_path, replica_host_id, host_id);
}
/// Before 24.6 we always created host_id with insecure port, even if cluster_auth_info.cluster_secure_connection was true.
/// So not to break compatibility, we need to update host_id to secure one if cluster_auth_info.cluster_secure_connection is true.
if (host_id != host_id_default && replica_host_id == host_id_default)
{
current_zookeeper->set(replica_path, host_id, -1);
createEmptyLogEntry(current_zookeeper);
}
/// Check that replica_group_name in ZooKeeper matches the local one and change it if necessary.
String zk_replica_group_name;
if (!current_zookeeper->tryGet(replica_path + "/replica_group", zk_replica_group_name))
@ -550,7 +561,7 @@ void DatabaseReplicated::createReplicaNodesInZooKeeper(const zkutil::ZooKeeperPt
"already contains some data and it does not look like Replicated database path.", zookeeper_path);
/// Write host name to replica_path, it will protect from multiple replicas with the same name
auto host_id = getHostID(getContext(), db_uuid);
auto host_id = getHostID(getContext(), db_uuid, cluster_auth_info.cluster_secure_connection);
for (int attempts = 10; attempts > 0; --attempts)
{

View File

@ -260,7 +260,9 @@ StoragePtr DatabaseWithOwnTablesBase::detachTableUnlocked(const String & table_n
res = it->second;
tables.erase(it);
res->is_detached = true;
CurrentMetrics::sub(getAttachedCounterForStorage(res), 1);
if (res->isSystemStorage() == false)
CurrentMetrics::sub(getAttachedCounterForStorage(res), 1);
auto table_id = res->getStorageID();
if (table_id.hasUUID())
@ -301,7 +303,9 @@ void DatabaseWithOwnTablesBase::attachTableUnlocked(const String & table_name, c
/// It is important to reset is_detached here since in case of RENAME in
/// non-Atomic database the is_detached is set to true before RENAME.
table->is_detached = false;
CurrentMetrics::add(getAttachedCounterForStorage(table), 1);
if (table->isSystemStorage() == false && table_id.database_name != DatabaseCatalog::SYSTEM_DATABASE)
CurrentMetrics::add(getAttachedCounterForStorage(table), 1);
}
void DatabaseWithOwnTablesBase::shutdown()

View File

@ -175,8 +175,7 @@ Columns DirectDictionary<dictionary_key_type>::getColumns(
if (!mask_filled)
(*default_mask)[requested_key_index] = 1;
Field value{};
result_column->insert(value);
result_column->insertDefault();
}
else
{

View File

@ -181,6 +181,8 @@ FormatSettings getFormatSettings(const ContextPtr & context, const Settings & se
format_settings.pretty.highlight_digit_groups = settings.output_format_pretty_highlight_digit_groups;
format_settings.pretty.output_format_pretty_row_numbers = settings.output_format_pretty_row_numbers;
format_settings.pretty.output_format_pretty_single_large_number_tip_threshold = settings.output_format_pretty_single_large_number_tip_threshold;
format_settings.pretty.output_format_pretty_display_footer_column_names = settings.output_format_pretty_display_footer_column_names;
format_settings.pretty.output_format_pretty_display_footer_column_names_min_rows = settings.output_format_pretty_display_footer_column_names_min_rows;
format_settings.protobuf.input_flatten_google_wrappers = settings.input_format_protobuf_flatten_google_wrappers;
format_settings.protobuf.output_nullables_with_google_wrappers = settings.output_format_protobuf_nullables_with_google_wrappers;
format_settings.protobuf.skip_fields_with_unsupported_types_in_schema_inference = settings.input_format_protobuf_skip_fields_with_unsupported_types_in_schema_inference;

Some files were not shown because too many files have changed in this diff Show More