/** * Copyright (C) 2018-present MongoDB, Inc. * * This program is free software: you can redistribute it and/or modify * it under the terms of the Server Side Public License, version 1, * as published by MongoDB, Inc. * * This program is distributed in the hope that it will be useful, * but WITHOUT ANY WARRANTY; without even the implied warranty of * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the * Server Side Public License for more details. * * You should have received a copy of the Server Side Public License * along with this program. If not, see * . * * As a special exception, the copyright holders give permission to link the * code of portions of this program with the OpenSSL library under certain * conditions as described in each individual source file and distribute * linked combinations including the program with the OpenSSL library. You * must comply with the Server Side Public License in all respects for * all of the code used other than as permitted herein. If you modify file(s) * with this exception, you may extend this exception to your version of the * file(s), but you are not obligated to do so. If you do not wish to do so, * delete this exception statement from your version. If you delete this * exception statement from all source files in the program, then also delete * it in the license file. */ #pragma once #include #include "mongo/db/operation_context.h" #include "mongo/db/repl/optime.h" #include "mongo/db/service_context.h" #include "mongo/db/transactions_stats_gen.h" #include "mongo/util/concurrency/with_lock.h" namespace mongo { /** * Container for server-wide multi-document transaction statistics. */ class ServerTransactionsMetrics { MONGO_DISALLOW_COPYING(ServerTransactionsMetrics); public: ServerTransactionsMetrics() = default; static ServerTransactionsMetrics* get(ServiceContext* service); static ServerTransactionsMetrics* get(OperationContext* opCtx); unsigned long long getCurrentActive() const; void decrementCurrentActive(); void incrementCurrentActive(); unsigned long long getCurrentInactive() const; void decrementCurrentInactive(); void incrementCurrentInactive(); unsigned long long getCurrentOpen() const; void decrementCurrentOpen(); void incrementCurrentOpen(); unsigned long long getTotalStarted() const; void incrementTotalStarted(); unsigned long long getTotalAborted() const; void incrementTotalAborted(); unsigned long long getTotalCommitted() const; void incrementTotalCommitted(); unsigned long long getTotalPrepared() const; void incrementTotalPrepared(); unsigned long long getTotalPreparedThenCommitted() const; void incrementTotalPreparedThenCommitted(); unsigned long long getTotalPreparedThenAborted() const; void incrementTotalPreparedThenAborted(); unsigned long long getCurrentPrepared() const; void incrementCurrentPrepared(); void decrementCurrentPrepared(); /** * Returns the OpTime of the oldest oplog entry written across all open transactions. * Returns boost::none if there are no transaction oplog entry OpTimes stored. */ boost::optional getOldestActiveOpTime() const; /** * Add the transaction's oplog entry OpTime to oldestActiveOplogEntryOpTimes, a set of OpTimes. * Also creates a pair with this OpTime and OpTime::max() as the corresponding commit/abort * oplog entry OpTime. Finally, adds this to oldestNonMajorityCommittedOpTimes. */ void addActiveOpTime(repl::OpTime oldestOplogEntryOpTime); /** * Remove the corresponding transaction oplog entry OpTime if the transaction commits or * aborts. Also updates the pair in oldestNonMajorityCommittedOpTimes with the * oldestOplogEntryOpTime to have a valid finishOpTime instead of OpTime::max(). It's stored in * the format: < oldestOplogEntryOpTime, finishOpTime >. */ void removeActiveOpTime(repl::OpTime oldestOplogEntryOpTime, boost::optional finishOpTime); /** * Returns the number of transaction oplog entry OpTimes currently stored. */ unsigned int getTotalActiveOpTimes() const; /** * Returns the oldest oplog entry OpTime across transactions whose corresponding commit or * abort oplog entry has not been majority committed. */ boost::optional getOldestNonMajorityCommittedOpTime() const; /** * Remove the corresponding transaction oplog entry OpTime pair from * oldestNonMajorityCommittedOpTimes if the transaction is majority committed or aborted. * We determine this by checking if there are any pairs in the set whose * 'finishOpTime' <= 'committedOpTime'. */ void removeOpTimesLessThanOrEqToCommittedOpTime(repl::OpTime committedOpTime); /** * Testing function that adds an OpTime pair to oldestNonMajorityCommittedOpTimes. */ void addNonMajCommittedOpTimePair_forTest(std::pair OpTimePair); /** * Testing function that returns the oldest non-majority committed OpTime pair in the form: * < oldestOplogEntryOpTime, finishOpTime >. */ boost::optional getFinishOpTimeOfOldestNonMajCommitted_forTest() const; /** * Appends the accumulated stats to a transactions stats object. */ void updateStats(TransactionsStats* stats, OperationContext* opCtx); /** * Invalidates the in-memory state of prepared transactions during replication rollback by * clearing oldestActiveOplogEntryOpTime, oldestActiveOplogEntryOpTimes, and * oldestNonMajorityCommittedOpTimes. These variables/data structures should be properly * reconstructed during replication recovery. */ void clearOpTimes(); private: /** * Returns the first and oldest optime in the ordered set of active oplog entry optimes. * Returns boost::none if there are no transaction oplog entry optimes stored. */ boost::optional _calculateOldestActiveOpTime(WithLock) const; /** * Returns the oldest read timestamp in use by any open unprepared transaction. This will * return a null timestamp if there is no oldest open unprepared read timestamp to be * returned. */ static Timestamp _getOldestOpenUnpreparedReadTimestamp(OperationContext* opCtx); // // Member variables, excluding atomic variables, are labeled with the following code to // indicate the synchronization rules for accessing them. // // (M) Reads and writes guarded by _mutex // mutable stdx::mutex _mutex; // The number of multi-document transactions currently active. AtomicUInt64 _currentActive{0}; // The number of multi-document transactions currently inactive. AtomicUInt64 _currentInactive{0}; // The total number of open transactions. AtomicUInt64 _currentOpen{0}; // The total number of multi-document transactions started since the last server startup. AtomicUInt64 _totalStarted{0}; // The total number of multi-document transaction aborts. AtomicUInt64 _totalAborted{0}; // The total number of multi-document transaction commits. AtomicUInt64 _totalCommitted{0}; // The total number of prepared transactions since the last server startup. AtomicUInt64 _totalPrepared{0}; // The total number of prepared transaction commits. AtomicUInt64 _totalPreparedThenCommitted{0}; // The total number of prepared transaction aborts. AtomicUInt64 _totalPreparedThenAborted{0}; // The current number of transactions in the prepared state. AtomicUInt64 _currentPrepared{0}; // The optime of the oldest oplog entry for any active transaction. boost::optional _oldestActiveOplogEntryOpTime; // (M) // Maintain the oldest oplog entry OpTime across all active transactions. Currently, we only // write an oplog entry for an ongoing transaction if it is in the `prepare` state. By // maintaining an ordered set of OpTimes, the OpTime at the beginning will be the oldest. std::set _oldestActiveOplogEntryOpTimes; // (M) // Maintain the oldest oplog entry OpTime across transactions whose corresponding abort/commit // oplog entries have not been majority committed. Since this is an ordered set, the first // pair's oldestOplogEntryOpTime represents the earliest OpTime that we should pin the stable // timestamp behind. // Each pair is structured as follows: // 'oldestOplogEntryOpTime': The first oplog entry OpTime written by a transaction. // 'finishOpTime': The commit/abort oplog entry OpTime. // Once the corresponding abort/commit entry has been majority committed, remove the pair from // this set. std::set> _oldestNonMajorityCommittedOpTimes; // (M) }; } // namespace mongo