ClickHouse/programs/local/LocalServer.cpp

642 lines
22 KiB
C++
Raw Normal View History

#include "LocalServer.h"
#include <Poco/Util/XMLConfiguration.h>
#include <Poco/Util/HelpFormatter.h>
#include <Poco/Util/OptionCallback.h>
#include <Poco/String.h>
#include <Poco/Logger.h>
#include <Poco/NullChannel.h>
#include <Databases/DatabaseMemory.h>
#include <Storages/System/attachSystemTables.h>
#include <Interpreters/ProcessList.h>
#include <Interpreters/executeQuery.h>
#include <Interpreters/loadMetadata.h>
2020-02-03 12:54:36 +00:00
#include <Interpreters/DatabaseCatalog.h>
#include <Common/Exception.h>
#include <Common/Macros.h>
#include <Common/Config/ConfigProcessor.h>
#include <Common/escapeForFileName.h>
#include <Common/ClickHouseRevision.h>
2019-01-28 11:18:00 +00:00
#include <Common/ThreadStatus.h>
2018-07-18 09:48:45 +00:00
#include <Common/config_version.h>
#include <Common/quoteString.h>
#include <IO/ReadBufferFromFile.h>
#include <IO/ReadBufferFromString.h>
#include <IO/WriteBufferFromFileDescriptor.h>
#include <IO/UseSSL.h>
#include <IO/ReadHelpers.h>
#include <Parsers/parseQuery.h>
#include <Parsers/IAST.h>
#include <common/ErrorHandlers.h>
2018-06-05 20:09:51 +00:00
#include <Common/StatusFile.h>
2017-04-21 17:47:27 +00:00
#include <Functions/registerFunctions.h>
#include <AggregateFunctions/registerAggregateFunctions.h>
#include <TableFunctions/registerTableFunctions.h>
#include <Storages/registerStorages.h>
#include <Dictionaries/registerDictionaries.h>
2019-11-27 09:39:44 +00:00
#include <Disks/registerDisks.h>
2020-10-29 03:39:43 +00:00
#include <Formats/registerFormats.h>
#include <boost/program_options/options_description.hpp>
#include <boost/program_options.hpp>
#include <common/argsToConfig.h>
2019-08-23 15:47:27 +00:00
#include <Common/TerminalSize.h>
#include <Common/randomSeed.h>
#include <filesystem>
namespace DB
{
namespace ErrorCodes
{
2020-06-24 22:07:01 +00:00
extern const int BAD_ARGUMENTS;
extern const int SYNTAX_ERROR;
extern const int CANNOT_LOAD_CONFIG;
extern const int FILE_ALREADY_EXISTS;
}
LocalServer::LocalServer() = default;
LocalServer::~LocalServer()
{
2020-10-22 07:37:03 +00:00
if (global_context)
global_context->shutdown(); /// required for properly exception handling
}
void LocalServer::initialize(Poco::Util::Application & self)
{
Poco::Util::Application::initialize(self);
/// Load config files if exists
if (config().has("config-file") || Poco::File("config.xml").exists())
{
const auto config_path = config().getString("config-file", "config.xml");
ConfigProcessor config_processor(config_path, false, true);
config_processor.setConfigPath(Poco::Path(config_path).makeParent().toString());
auto loaded_config = config_processor.loadConfig();
config_processor.savePreprocessedConfig(loaded_config, loaded_config.configuration->getString("path", "."));
config().add(loaded_config.configuration.duplicate(), PRIO_DEFAULT, false);
}
2020-06-28 23:40:43 +00:00
if (config().has("logger.console") || config().has("logger.level") || config().has("logger.log"))
{
2020-06-28 23:40:43 +00:00
// force enable logging
config().setString("logger", "logger");
2019-06-20 07:17:21 +00:00
// sensitive data rules are not used here
2020-06-28 23:40:43 +00:00
buildLoggers(config(), logger(), "clickhouse-local");
}
else
{
// Turn off server logging to stderr
if (!config().has("verbose"))
{
Poco::Logger::root().setLevel("none");
Poco::Logger::root().setChannel(Poco::AutoPtr<Poco::NullChannel>(new Poco::NullChannel()));
}
}
}
2020-10-22 07:37:03 +00:00
void LocalServer::applyCmdSettings(Context & context)
{
2020-10-22 07:37:03 +00:00
context.applySettingsChanges(cmd_settings.changes());
}
/// If path is specified and not empty, will try to setup server environment and load existing metadata
void LocalServer::tryInitPath()
{
std::string path;
if (config().has("path"))
{
// User-supplied path.
path = config().getString("path");
Poco::trimInPlace(path);
if (path.empty())
{
throw Exception(ErrorCodes::BAD_ARGUMENTS,
2020-08-08 01:21:04 +00:00
"Cannot work with empty storage path that is explicitly specified"
2020-06-24 22:07:01 +00:00
" by the --path option. Please check the program options and"
" correct the --path.");
}
}
else
{
2020-10-26 07:59:15 +00:00
// The path is not provided explicitly - use a unique path in the system temporary directory
// (or in the current dir if temporary don't exist)
Poco::Logger * log = &logger();
std::filesystem::path parent_folder;
std::filesystem::path default_path;
try
{
// try to guess a tmp folder name, and check if it's a directory (throw exception otherwise)
parent_folder = std::filesystem::temp_directory_path();
}
catch (const std::filesystem::filesystem_error& e)
{
// tmp folder don't exists? misconfiguration? chroot?
LOG_DEBUG(log, "Can not get temporary folder: {}", e.what());
parent_folder = std::filesystem::current_path();
std::filesystem::is_directory(parent_folder); // that will throw an exception if it's not a directory
LOG_DEBUG(log, "Will create working directory inside current directory: {}", parent_folder.string());
}
2020-10-26 07:59:15 +00:00
/// we can have another clickhouse-local running simultaneously, even with the same PID (for ex. - several dockers mounting the same folder)
/// or it can be some leftovers from other clickhouse-local runs
/// as we can't accurately distinguish those situations we don't touch any existent folders
/// we just try to pick some free name for our working folder
default_path = parent_folder / fmt::format("clickhouse-local-{}-{}-{}", getpid(), time(nullptr), randomSeed());
if (exists(default_path))
2020-10-27 11:04:03 +00:00
throw Exception(ErrorCodes::FILE_ALREADY_EXISTS, "Unsuccessful attempt to create working directory: {} exist!", default_path.string());
create_directory(default_path);
temporary_directory_to_delete = default_path;
path = default_path.string();
LOG_DEBUG(log, "Working directory created: {}", path);
}
if (path.back() != '/')
path += '/';
2020-10-22 07:37:03 +00:00
global_context->setPath(path);
2020-10-22 07:37:03 +00:00
global_context->setTemporaryStorage(path + "tmp");
global_context->setFlagsPath(path + "flags");
2020-10-22 07:37:03 +00:00
global_context->setUserFilesPath(""); // user's files are everywhere
}
2020-05-28 20:10:45 +00:00
static void attachSystemTables(const Context & context)
2020-03-18 00:57:00 +00:00
{
DatabasePtr system_database = DatabaseCatalog::instance().tryGetDatabase(DatabaseCatalog::SYSTEM_DATABASE);
if (!system_database)
{
/// TODO: add attachTableDelayed into DatabaseMemory to speedup loading
2020-05-28 20:10:45 +00:00
system_database = std::make_shared<DatabaseMemory>(DatabaseCatalog::SYSTEM_DATABASE, context);
2020-03-18 00:57:00 +00:00
DatabaseCatalog::instance().attachDatabase(DatabaseCatalog::SYSTEM_DATABASE, system_database);
}
attachSystemTablesLocal(*system_database);
}
2017-12-02 02:47:12 +00:00
int LocalServer::main(const std::vector<std::string> & /*args*/)
try
{
2020-05-30 21:57:37 +00:00
Poco::Logger * log = &logger();
2019-01-28 11:18:00 +00:00
ThreadStatus thread_status;
UseSSL use_ssl;
if (!config().has("query") && !config().has("table-structure") && !config().has("queries-file")) /// Nothing to process
{
if (config().hasOption("verbose"))
std::cerr << "There are no queries to process." << '\n';
return Application::EXIT_OK;
}
if (config().has("query") && config().has("queries-file"))
{
throw Exception("Specify either `query` or `queries-file` option", ErrorCodes::BAD_ARGUMENTS);
}
2020-04-17 20:56:38 +00:00
shared_context = Context::createShared();
2020-10-22 07:37:03 +00:00
global_context = std::make_unique<Context>(Context::createGlobal(shared_context.get()));
global_context->makeGlobalContext();
global_context->setApplicationType(Context::ApplicationType::LOCAL);
tryInitPath();
std::optional<StatusFile> status;
/// Skip temp path installation
/// We will terminate process on error
static KillingErrorHandler error_handler;
Poco::ErrorHandler::set(&error_handler);
2019-01-22 19:56:53 +00:00
/// Don't initialize DateLUT
2017-04-21 17:47:27 +00:00
registerFunctions();
registerAggregateFunctions();
registerTableFunctions();
registerStorages();
registerDictionaries();
2019-11-27 09:39:44 +00:00
registerDisks();
2020-10-29 03:39:43 +00:00
registerFormats();
2017-04-21 17:47:27 +00:00
/// Maybe useless
if (config().has("macros"))
2020-10-22 07:37:03 +00:00
global_context->setMacros(std::make_unique<Macros>(config(), "macros", log));
/// Skip networking
/// Sets external authenticators config (LDAP).
2020-10-22 07:37:03 +00:00
global_context->setExternalAuthenticatorsConfig(config());
setupUsers();
/// Limit on total number of concurrently executing queries.
2020-03-28 03:02:26 +00:00
/// There is no need for concurrent queries, override max_concurrent_queries.
2020-10-22 07:37:03 +00:00
global_context->getProcessList().setMaxSize(0);
/// Size of cache for uncompressed blocks. Zero means disabled.
size_t uncompressed_cache_size = config().getUInt64("uncompressed_cache_size", 0);
if (uncompressed_cache_size)
2020-10-22 07:37:03 +00:00
global_context->setUncompressedCache(uncompressed_cache_size);
/// Size of cache for marks (index of MergeTree family of tables). It is necessary.
/// Specify default value for mark_cache_size explicitly!
size_t mark_cache_size = config().getUInt64("mark_cache_size", 5368709120);
if (mark_cache_size)
2020-10-22 07:37:03 +00:00
global_context->setMarkCache(mark_cache_size);
/// Load global settings from default_profile and system_profile.
2020-10-22 07:37:03 +00:00
global_context->setDefaultProfiles(config());
/** Init dummy default DB
2020-01-11 09:50:41 +00:00
* NOTE: We force using isolated default database to avoid conflicts with default database from server environment
* Otherwise, metadata of temporary File(format, EXPLICIT_PATH) tables will pollute metadata/ directory;
* if such tables will not be dropped, clickhouse-server will not be able to load them due to security reasons.
*/
std::string default_database = config().getString("default_database", "_local");
2020-10-22 07:37:03 +00:00
DatabaseCatalog::instance().attachDatabase(default_database, std::make_shared<DatabaseMemory>(default_database, *global_context));
global_context->setCurrentDatabase(default_database);
applyCmdOptions(*global_context);
if (config().has("path"))
{
String path = global_context->getPath();
/// Lock path directory before read
status.emplace(path + "status", StatusFile::write_full_info);
2020-08-13 19:41:06 +00:00
LOG_DEBUG(log, "Loading metadata from {}", path);
Poco::File(path + "data/").createDirectories();
Poco::File(path + "metadata/").createDirectories();
2020-10-22 07:37:03 +00:00
loadMetadataSystem(*global_context);
attachSystemTables(*global_context);
loadMetadata(*global_context);
DatabaseCatalog::instance().loadDatabases();
2020-05-23 22:24:01 +00:00
LOG_DEBUG(log, "Loaded metadata.");
}
else if (!config().has("no-system-tables"))
{
2020-10-22 07:37:03 +00:00
attachSystemTables(*global_context);
}
processQueries();
2020-10-22 07:37:03 +00:00
global_context->shutdown();
global_context.reset();
status.reset();
cleanup();
return Application::EXIT_OK;
}
catch (const Exception & e)
{
try
{
cleanup();
}
catch (...)
{
tryLogCurrentException(__PRETTY_FUNCTION__);
}
std::cerr << getCurrentExceptionMessage(config().hasOption("stacktrace")) << '\n';
/// If exception code isn't zero, we should return non-zero return code anyway.
return e.code() ? e.code() : -1;
}
std::string LocalServer::getInitialCreateTableQuery()
{
if (!config().has("table-structure"))
return {};
auto table_name = backQuoteIfNeed(config().getString("table-name", "table"));
auto table_structure = config().getString("table-structure");
auto data_format = backQuoteIfNeed(config().getString("table-data-format", "TSV"));
String table_file;
if (!config().has("table-file") || config().getString("table-file") == "-") /// Use Unix tools stdin naming convention
table_file = "stdin";
else /// Use regular file
table_file = quoteString(config().getString("table-file"));
return
"CREATE TABLE " + table_name +
" (" + table_structure + ") " +
"ENGINE = "
"File(" + data_format + ", " + table_file + ")"
"; ";
}
void LocalServer::processQueries()
{
String initial_create_query = getInitialCreateTableQuery();
String queries_str = initial_create_query;
if (config().has("query"))
queries_str += config().getRawString("query");
else
{
String queries_from_file;
ReadBufferFromFile in(config().getString("queries-file"));
readStringUntilEOF(queries_from_file, in);
queries_str += queries_from_file;
}
2020-10-22 07:37:03 +00:00
const auto & settings = global_context->getSettingsRef();
std::vector<String> queries;
auto parse_res = splitMultipartQuery(queries_str, queries, settings.max_query_size, settings.max_parser_depth);
if (!parse_res.second)
throw Exception("Cannot parse and execute the following part of query: " + String(parse_res.first), ErrorCodes::SYNTAX_ERROR);
2020-10-22 07:37:03 +00:00
/// we can't mutate global global_context (can lead to races, as it was already passed to some background threads)
/// so we can't reuse it safely as a query context and need a copy here
2020-10-22 07:37:03 +00:00
auto context = Context(*global_context);
2020-10-22 07:37:03 +00:00
context.makeSessionContext();
context.makeQueryContext();
2020-10-21 18:36:01 +00:00
2020-10-22 07:37:03 +00:00
context.setUser("default", "", Poco::Net::SocketAddress{});
context.setCurrentQueryId("");
applyCmdSettings(context);
/// Use the same query_id (and thread group) for all queries
2020-10-22 07:37:03 +00:00
CurrentThread::QueryScope query_scope_holder(context);
bool echo_queries = config().hasOption("echo") || config().hasOption("verbose");
std::exception_ptr exception;
for (const auto & query : queries)
{
ReadBufferFromString read_buf(query);
WriteBufferFromFileDescriptor write_buf(STDOUT_FILENO);
if (echo_queries)
{
writeString(query, write_buf);
writeChar('\n', write_buf);
write_buf.next();
}
try
{
2020-10-22 07:37:03 +00:00
executeQuery(read_buf, write_buf, /* allow_into_outfile = */ true, context, {});
}
catch (...)
{
if (!config().hasOption("ignore-error"))
throw;
if (!exception)
exception = std::current_exception();
std::cerr << getCurrentExceptionMessage(config().hasOption("stacktrace")) << '\n';
}
}
if (exception)
std::rethrow_exception(exception);
}
static const char * minimal_default_user_xml =
"<yandex>"
" <profiles>"
" <default></default>"
" </profiles>"
" <users>"
" <default>"
" <password></password>"
" <networks>"
" <ip>::/0</ip>"
" </networks>"
" <profile>default</profile>"
" <quota>default</quota>"
" </default>"
" </users>"
" <quotas>"
" <default></default>"
" </quotas>"
"</yandex>";
static ConfigurationPtr getConfigurationFromXMLString(const char * xml_data)
{
2020-11-10 18:22:26 +00:00
std::stringstream ss{std::string{xml_data}}; // STYLE_CHECK_ALLOW_STD_STRING_STREAM
Poco::XML::InputSource input_source{ss};
return {new Poco::Util::XMLConfiguration{&input_source}};
}
void LocalServer::setupUsers()
{
ConfigurationPtr users_config;
if (config().has("users_config") || config().has("config-file") || Poco::File("config.xml").exists())
{
const auto users_config_path = config().getString("users_config", config().getString("config-file", "config.xml"));
ConfigProcessor config_processor(users_config_path);
const auto loaded_config = config_processor.loadConfig();
config_processor.savePreprocessedConfig(loaded_config, config().getString("path", DBMS_DEFAULT_PATH));
users_config = loaded_config.configuration;
}
else
{
users_config = getConfigurationFromXMLString(minimal_default_user_xml);
}
if (users_config)
2020-10-22 07:37:03 +00:00
global_context->setUsersConfig(users_config);
else
throw Exception("Can't load config for users", ErrorCodes::CANNOT_LOAD_CONFIG);
}
void LocalServer::cleanup()
{
2020-10-26 16:09:43 +00:00
// Delete the temporary directory if needed.
if (temporary_directory_to_delete)
{
const auto dir = *temporary_directory_to_delete;
temporary_directory_to_delete.reset();
LOG_DEBUG(&logger(), "Removing temporary directory: {}", dir.string());
remove_all(dir);
}
}
static void showClientVersion()
{
std::cout << DBMS_NAME << " client version " << VERSION_STRING << VERSION_OFFICIAL << "." << '\n';
}
2020-03-18 00:57:00 +00:00
static std::string getHelpHeader()
{
return
"usage: clickhouse-local [initial table definition] [--query <query>]\n"
"clickhouse-local allows to execute SQL queries on your data files via single command line call."
" To do so, initially you need to define your data source and its format."
" After you can execute your SQL queries in usual manner.\n"
"There are two ways to define initial table keeping your data."
" Either just in first query like this:\n"
" CREATE TABLE <table> (<structure>) ENGINE = File(<input-format>, <file>);\n"
"Either through corresponding command line parameters --table --structure --input-format and --file.";
}
2020-03-18 00:57:00 +00:00
static std::string getHelpFooter()
{
return
"Example printing memory used by each Unix user:\n"
"ps aux | tail -n +2 | awk '{ printf(\"%s\\t%s\\n\", $1, $4) }' | "
"clickhouse-local -S \"user String, mem Float64\" -q"
" \"SELECT user, round(sum(mem), 2) as mem_total FROM table GROUP BY user ORDER"
" BY mem_total DESC FORMAT PrettyCompact\"";
}
void LocalServer::init(int argc, char ** argv)
{
namespace po = boost::program_options;
/// Don't parse options with Poco library, we prefer neat boost::program_options
stopOptionsProcessing();
2019-08-23 15:47:27 +00:00
po::options_description description = createOptionsDescription("Main options", getTerminalWidth());
description.add_options()
("help", "produce help message")
("config-file,c", po::value<std::string>(), "config-file path")
("query,q", po::value<std::string>(), "query")
("queries-file, qf", po::value<std::string>(), "file path with queries to execute")
("database,d", po::value<std::string>(), "database")
("table,N", po::value<std::string>(), "name of the initial table")
/// If structure argument is omitted then initial query is not generated
("structure,S", po::value<std::string>(), "structure of the initial table (list of column and type names)")
("file,f", po::value<std::string>(), "path to file with data of the initial table (stdin if not specified)")
("input-format", po::value<std::string>(), "input format of the initial table data")
("format,f", po::value<std::string>(), "default output format (clickhouse-client compatibility)")
("output-format", po::value<std::string>(), "default output format")
("stacktrace", "print stack traces of exceptions")
("echo", "print query before execution")
("verbose", "print query and other debugging info")
2020-06-28 23:40:43 +00:00
("logger.console", po::value<bool>()->implicit_value(true), "Log to console")
("logger.log", po::value<std::string>(), "Log file name")
("logger.level", po::value<std::string>(), "Log level")
("ignore-error", "do not stop processing if a query failed")
("no-system-tables", "do not attach system tables (better startup time)")
("version,V", "print version information and exit")
;
cmd_settings.addProgramOptions(description);
/// Parse main commandline options.
po::parsed_options parsed = po::command_line_parser(argc, argv).options(description).run();
po::variables_map options;
po::store(parsed, options);
po::notify(options);
if (options.count("version") || options.count("V"))
{
showClientVersion();
exit(0);
}
if (options.empty() || options.count("help"))
{
std::cout << getHelpHeader() << "\n";
std::cout << description << "\n";
std::cout << getHelpFooter() << "\n";
exit(0);
}
/// Save received data into the internal config.
if (options.count("config-file"))
config().setString("config-file", options["config-file"].as<std::string>());
if (options.count("query"))
config().setString("query", options["query"].as<std::string>());
if (options.count("queries-file"))
config().setString("queries-file", options["queries-file"].as<std::string>());
if (options.count("database"))
config().setString("default_database", options["database"].as<std::string>());
if (options.count("table"))
config().setString("table-name", options["table"].as<std::string>());
if (options.count("file"))
config().setString("table-file", options["file"].as<std::string>());
if (options.count("structure"))
config().setString("table-structure", options["structure"].as<std::string>());
if (options.count("input-format"))
config().setString("table-data-format", options["input-format"].as<std::string>());
if (options.count("format"))
config().setString("format", options["format"].as<std::string>());
if (options.count("output-format"))
config().setString("output-format", options["output-format"].as<std::string>());
if (options.count("stacktrace"))
config().setBool("stacktrace", true);
if (options.count("echo"))
config().setBool("echo", true);
if (options.count("verbose"))
config().setBool("verbose", true);
2020-06-28 23:40:43 +00:00
if (options.count("logger.console"))
config().setBool("logger.console", options["logger.console"].as<bool>());
if (options.count("logger.log"))
config().setString("logger.log", options["logger.log"].as<std::string>());
if (options.count("logger.level"))
config().setString("logger.level", options["logger.level"].as<std::string>());
if (options.count("ignore-error"))
config().setBool("ignore-error", true);
if (options.count("no-system-tables"))
config().setBool("no-system-tables", true);
std::vector<std::string> arguments;
for (int arg_num = 1; arg_num < argc; ++arg_num)
arguments.emplace_back(argv[arg_num]);
argsToConfig(arguments, config(), 100);
}
2020-10-22 07:37:03 +00:00
void LocalServer::applyCmdOptions(Context & context)
2018-05-11 14:35:32 +00:00
{
2020-10-22 07:37:03 +00:00
context.setDefaultFormat(config().getString("output-format", config().getString("format", "TSV")));
applyCmdSettings(context);
2018-05-11 14:35:32 +00:00
}
}
2019-12-15 06:34:43 +00:00
#pragma GCC diagnostic ignored "-Wunused-function"
#pragma GCC diagnostic ignored "-Wmissing-declarations"
2017-12-02 02:47:12 +00:00
int mainEntryClickHouseLocal(int argc, char ** argv)
{
DB::LocalServer app;
try
{
app.init(argc, argv);
return app.run();
}
catch (...)
{
std::cerr << DB::getCurrentExceptionMessage(true) << '\n';
2017-12-02 02:47:12 +00:00
auto code = DB::getCurrentExceptionCode();
return code ? code : 1;
}
}