ClickHouse/src/Interpreters/PartLog.cpp

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

275 lines
9.4 KiB
C++
Raw Normal View History

#include <Columns/ColumnsNumber.h>
#include <DataTypes/DataTypeArray.h>
#include <DataTypes/DataTypesNumber.h>
#include <DataTypes/DataTypeDateTime.h>
#include <DataTypes/DataTypeDateTime64.h>
#include <DataTypes/DataTypeDate.h>
#include <DataTypes/DataTypeString.h>
#include <DataTypes/DataTypeEnum.h>
#include <DataTypes/DataTypeUUID.h>
2019-10-10 16:30:30 +00:00
#include <Storages/MergeTree/IMergeTreeDataPart.h>
#include <Storages/MergeTree/MergeTreeData.h>
#include <Interpreters/PartLog.h>
#include <Interpreters/Context.h>
#include <Interpreters/ProfileEventsExt.h>
#include <Common/ProfileEvents.h>
#include <DataTypes/DataTypeMap.h>
#include <Common/CurrentThread.h>
namespace DB
{
namespace ErrorCodes
{
extern const int NOT_IMPLEMENTED;
}
PartLogElement::MergeReasonType PartLogElement::getMergeReasonType(MergeType merge_type)
{
switch (merge_type)
{
case MergeType::Regular:
return REGULAR_MERGE;
case MergeType::TTLDelete:
return TTL_DELETE_MERGE;
case MergeType::TTLRecompress:
return TTL_RECOMPRESS_MERGE;
}
throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Unknown MergeType {}", static_cast<UInt64>(merge_type));
}
PartLogElement::PartMergeAlgorithm PartLogElement::getMergeAlgorithm(MergeAlgorithm merge_algorithm_)
{
switch (merge_algorithm_)
{
case MergeAlgorithm::Undecided:
return UNDECIDED;
case MergeAlgorithm::Horizontal:
return HORIZONTAL;
case MergeAlgorithm::Vertical:
return VERTICAL;
}
throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Unknown MergeAlgorithm {}", static_cast<UInt64>(merge_algorithm_));
}
NamesAndTypesList PartLogElement::getNamesAndTypes()
{
auto event_type_datatype = std::make_shared<DataTypeEnum8>(
2018-09-09 02:23:24 +00:00
DataTypeEnum8::Values
{
2019-08-01 16:37:50 +00:00
{"NewPart", static_cast<Int8>(NEW_PART)},
{"MergeParts", static_cast<Int8>(MERGE_PARTS)},
{"DownloadPart", static_cast<Int8>(DOWNLOAD_PART)},
{"RemovePart", static_cast<Int8>(REMOVE_PART)},
{"MutatePart", static_cast<Int8>(MUTATE_PART)},
2019-09-03 11:32:25 +00:00
{"MovePart", static_cast<Int8>(MOVE_PART)},
}
);
2021-02-03 08:29:34 +00:00
auto merge_reason_datatype = std::make_shared<DataTypeEnum8>(
DataTypeEnum8::Values
{
{"NotAMerge", static_cast<Int8>(NOT_A_MERGE)},
{"RegularMerge", static_cast<Int8>(REGULAR_MERGE)},
{"TTLDeleteMerge", static_cast<Int8>(TTL_DELETE_MERGE)},
{"TTLRecompressMerge", static_cast<Int8>(TTL_RECOMPRESS_MERGE)},
}
);
auto merge_algorithm_datatype = std::make_shared<DataTypeEnum8>(
DataTypeEnum8::Values
{
{"Undecided", static_cast<Int8>(UNDECIDED)},
{"Horizontal", static_cast<Int8>(HORIZONTAL)},
{"Vertical", static_cast<Int8>(VERTICAL)},
}
);
ColumnsWithTypeAndName columns_with_type_and_name;
2021-02-03 08:29:34 +00:00
return {
{"query_id", std::make_shared<DataTypeString>()},
{"event_type", std::move(event_type_datatype)},
{"merge_reason", std::move(merge_reason_datatype)},
{"merge_algorithm", std::move(merge_algorithm_datatype)},
{"event_date", std::make_shared<DataTypeDate>()},
{"event_time", std::make_shared<DataTypeDateTime>()},
{"event_time_microseconds", std::make_shared<DataTypeDateTime64>(6)},
{"duration_ms", std::make_shared<DataTypeUInt64>()},
{"database", std::make_shared<DataTypeString>()},
{"table", std::make_shared<DataTypeString>()},
{"table_uuid", std::make_shared<DataTypeUUID>()},
{"part_name", std::make_shared<DataTypeString>()},
{"partition_id", std::make_shared<DataTypeString>()},
{"part_type", std::make_shared<DataTypeString>()},
2022-03-10 15:44:19 +00:00
{"disk_name", std::make_shared<DataTypeString>()},
{"path_on_disk", std::make_shared<DataTypeString>()},
{"rows", std::make_shared<DataTypeUInt64>()},
{"size_in_bytes", std::make_shared<DataTypeUInt64>()}, // On disk
/// Merge-specific info
{"merged_from", std::make_shared<DataTypeArray>(std::make_shared<DataTypeString>())},
{"bytes_uncompressed", std::make_shared<DataTypeUInt64>()}, // Result bytes
{"read_rows", std::make_shared<DataTypeUInt64>()},
{"read_bytes", std::make_shared<DataTypeUInt64>()},
{"peak_memory_usage", std::make_shared<DataTypeUInt64>()},
/// Is there an error during the execution or commit
{"error", std::make_shared<DataTypeUInt16>()},
{"exception", std::make_shared<DataTypeString>()},
{"ProfileEvents", std::make_shared<DataTypeMap>(std::make_shared<DataTypeString>(), std::make_shared<DataTypeUInt64>())},
};
}
NamesAndAliases PartLogElement::getNamesAndAliases()
{
return
{
{"ProfileEvents.Names", {std::make_shared<DataTypeArray>(std::make_shared<DataTypeString>())}, "mapKeys(ProfileEvents)"},
{"ProfileEvents.Values", {std::make_shared<DataTypeArray>(std::make_shared<DataTypeUInt64>())}, "mapValues(ProfileEvents)"},
2023-04-17 12:45:04 +00:00
{"name", {std::make_shared<DataTypeString>()}, "part_name"},
2018-05-07 02:01:11 +00:00
};
}
void PartLogElement::appendToBlock(MutableColumns & columns) const
{
size_t i = 0;
columns[i++]->insert(query_id);
columns[i++]->insert(event_type);
columns[i++]->insert(merge_reason);
columns[i++]->insert(merge_algorithm);
Extended range of DateTime64 to years 1925 - 2238 The Year 1925 is a starting point because most of the timezones switched to saner (mostly 15-minutes based) offsets somewhere during 1924 or before. And that significantly simplifies implementation. 2238 is to simplify arithmetics for sanitizing LUT index access; there are less than 0x1ffff days from 1925. * Extended DateLUTImpl internal LUT to 0x1ffff items, some of which represent negative (pre-1970) time values. As a collateral benefit, Date now correctly supports dates up to 2149 (instead of 2106). * Added a new strong typedef ExtendedDayNum, which represents dates pre-1970 and post 2149. * Functions that used to return DayNum now return ExtendedDayNum. * Refactored DateLUTImpl to untie DayNum from the dual role of being a value and an index (due to negative time). Index is now a different type LUTIndex with explicit conversion functions from DatNum, time_t, and ExtendedDayNum. * Updated DateLUTImpl to properly support values close to epoch start (1970-01-01 00:00), including negative ones. * Reduced resolution of DateLUTImpl::Values::time_at_offset_change to multiple of 15-minutes to allow storing 64-bits of time_t in DateLUTImpl::Value while keeping same size. * Minor performance updates to DateLUTImpl when building month LUT by skipping non-start-of-month days. * Fixed extractTimeZoneFromFunctionArguments to work correctly with DateTime64. * New unit-tests and stateless integration tests for both DateTime and DateTime64.
2020-04-17 13:26:44 +00:00
columns[i++]->insert(DateLUT::instance().toDayNum(event_time).toUnderType());
columns[i++]->insert(event_time);
columns[i++]->insert(event_time_microseconds);
columns[i++]->insert(duration_ms);
columns[i++]->insert(database_name);
columns[i++]->insert(table_name);
columns[i++]->insert(table_uuid);
columns[i++]->insert(part_name);
2018-09-13 12:48:46 +00:00
columns[i++]->insert(partition_id);
columns[i++]->insert(part_type.toString());
2022-03-10 15:44:19 +00:00
columns[i++]->insert(disk_name);
2019-09-03 11:32:25 +00:00
columns[i++]->insert(path_on_disk);
columns[i++]->insert(rows);
columns[i++]->insert(bytes_compressed_on_disk);
Array source_part_names_array;
source_part_names_array.reserve(source_part_names.size());
for (const auto & name : source_part_names)
source_part_names_array.push_back(name);
columns[i++]->insert(source_part_names_array);
columns[i++]->insert(bytes_uncompressed);
columns[i++]->insert(rows_read);
columns[i++]->insert(bytes_read_uncompressed);
2020-03-19 11:31:21 +00:00
columns[i++]->insert(peak_memory_usage);
columns[i++]->insert(error);
columns[i++]->insert(exception);
if (profile_counters)
{
auto * column = columns[i++].get();
ProfileEvents::dumpToMapColumn(*profile_counters, column, true);
}
else
{
columns[i++]->insertDefault();
}
}
bool PartLog::addNewParts(
2023-01-23 12:45:28 +00:00
ContextPtr current_context, const PartLog::PartLogEntries & parts, const ExecutionStatus & execution_status)
{
if (parts.empty())
return true;
2019-03-21 19:22:38 +00:00
std::shared_ptr<PartLog> part_log;
try
{
2023-01-23 12:45:28 +00:00
auto table_id = parts.front().part->storage.getStorageID();
part_log = current_context->getPartLog(table_id.database_name); // assume parts belong to the same table
if (!part_log)
return false;
auto query_id = CurrentThread::getQueryId();
2023-01-23 12:45:28 +00:00
for (const auto & part_log_entry : parts)
{
2023-01-23 12:45:28 +00:00
const auto & part = part_log_entry.part;
PartLogElement elem;
if (!query_id.empty())
elem.query_id.insert(0, query_id.data(), query_id.size());
2023-02-19 22:15:09 +00:00
elem.event_type = PartLogElement::NEW_PART;
// construct event_time and event_time_microseconds using the same time point
// so that the two times will always be equal up to a precision of a second.
const auto time_now = std::chrono::system_clock::now();
elem.event_time = timeInSeconds(time_now);
elem.event_time_microseconds = timeInMicroseconds(time_now);
2023-01-23 12:45:28 +00:00
elem.duration_ms = part_log_entry.elapsed_ns / 1000000;
elem.database_name = table_id.database_name;
elem.table_name = table_id.table_name;
elem.table_uuid = table_id.uuid;
2018-09-13 12:48:46 +00:00
elem.partition_id = part->info.partition_id;
2019-01-31 17:30:56 +00:00
elem.part_name = part->name;
elem.disk_name = part->getDataPartStorage().getDiskName();
elem.path_on_disk = part->getDataPartStorage().getFullPath();
elem.part_type = part->getType();
2020-03-23 13:32:02 +00:00
elem.bytes_compressed_on_disk = part->getBytesOnDisk();
elem.rows = part->rows_count;
elem.error = static_cast<UInt16>(execution_status.code);
elem.exception = execution_status.message;
2023-01-23 12:45:28 +00:00
elem.profile_counters = part_log_entry.profile_counters;
2023-07-28 07:23:34 +00:00
part_log->add(std::move(elem));
}
}
catch (...)
{
2020-05-30 21:57:37 +00:00
tryLogCurrentException(part_log ? part_log->log : &Poco::Logger::get("PartLog"), __PRETTY_FUNCTION__);
return false;
}
return true;
}
2023-01-23 12:45:28 +00:00
bool PartLog::addNewPart(ContextPtr context, const PartLog::PartLogEntry & part, const ExecutionStatus & execution_status)
{
return addNewParts(context, {part}, execution_status);
}
PartLog::PartLogEntries PartLog::createPartLogEntries(const MutableDataPartsVector & parts, UInt64 elapsed_ns, ProfileCountersSnapshotPtr profile_counters)
2023-01-23 12:45:28 +00:00
{
PartLogEntries part_log_entries;
2023-01-23 12:45:28 +00:00
part_log_entries.reserve(parts.size());
for (const auto & part : parts)
part_log_entries.emplace_back(part, elapsed_ns, profile_counters);
2023-01-23 12:45:28 +00:00
return part_log_entries;
2023-01-23 12:45:28 +00:00
}
}