MongoDB 3.2.4 发布了，MongoDB是一个介于关系数据库和非关系数据库之间的产品，是非关系数据库当中功能最丰富，最像关系数据库的。他支持的数据结构非常松散，是类似json的bjson格式，因此可以存储比较复杂的数据类型。Mongo最大的特点是他支持的查询语言非常强大，其语法有点类似于面向对象的查询语言，几乎可以实现类似关系数据库单表查询的绝大部分功能，而且还支持对数据建立索引。
改进日志：
Security
SERVER-22237 Built-in role that allows full control over data, but not security or topology
Sharding
SERVER-21758 Test behavior when ‘nearest’ config server has severe replication lag
SERVER-22184 Operations that fail against a secondary in a sharded cluster may have their error message swallowed
SERVER-22239 wait for replication after duplicate key error from insert operations
SERVER-22297 Add targeted jstests for csrs upgrade during common operations
SERVER-22299 Add a jstest that runs moveChunk directly against a mongod that is not yet sharding aware, providing an SCCC connection string for the config servers
SERVER-22524 Only interrupt mapReduce on catalog manager swap if it is outputting to a sharded collection
SERVER-22543 multi_write_target.js should not rely on the order of shard ids
SERVER-22547 add support for config server ReplSetTest options to ShardingTest
SERVER-22553 mongos_shard_failure_tolerance.js should not rely on order of shard ids
SERVER-22569 Initialization of eooElement static local variable isn’t thread safe with MSVC 2013
SERVER-22584 Make sure IncompatibleCatalogManager errors fully propagate wherever thrown
SERVER-22585 CatalogManagerLegacy needs retry logic on config server reads
SERVER-22590 applyChunkOpsDeprecated retries and throws an error on preCondition no longer matching because the original write worked
SERVER-22592 Remove duplicate check for ‘enabled’ from ShardingState::_refreshMetadata
SERVER-22627 ShardRegistry should mark hosts which failed due to OperationTimeout as faulty
SERVER-22783 CSRS catalog manager writes should retry on WriteConcernFailed error
SERVER-22789 CSRS catalog manager writes should use writeConcern majority
SERVER-22797 Calls to ShardRegistry::reload needs to be serialized
SERVER-22822 Prevent mongod step down during moveChunk in balance_repl.js and sharding_rs2.js
SERVER-22849 Shard registry should update config last visible opTime even on command errors
SERVER-22859 SCCC config server reads need to specify slaveOk
SERVER-22862 Deadlock between ReplicaSetMonitor updating the connection string for a shard and reloading the ShardRegistry
SERVER-22863 sharding read_after_optime.js test timeout should exceed ping interval
SERVER-22878 rewrite checks in csrs_upgrade_mongod_using_movechunk.js after new assert.contains implementation
SERVER-22880 add requires_persistence tag to csrs_upgrade_mongod_using_movechunk.js
Replication
SERVER-21698 Add error-checking for isMaster() return values in jstests/libs/election_timing_test.js
SERVER-21972 improve naming of ReplicationCoordinator and TopologyCoordinator unittests
SERVER-22269 ReadConcern: majority does not reflect journaled state on PRIMARY
SERVER-22276 implement “j” flag in write concern apply to secondary as well as primary
SERVER-22277 test “j” flag in write concern apply to secondary as well as primary
SERVER-22287 Merging replica sets with replication protocol version 1 may result in two primaries
SERVER-22426 priority_takeover_one_node_higher_priority.js should call ReplSetTest.awaitReplication() before stepping primary down
SERVER-22428 Log read-after-optime timeouts
SERVER-22495 Running without journaling doesn’t set all OpTimes (lastDurableOpTime)
SERVER-22521 default timeout for ReplSetTest.initiate() from 60 seconds to 120 seconds should be longer to accommodate slow hosts
SERVER-22595 Reactivate rollback4.js
SERVER-22598 ensure all default write concern options use sync unset
SERVER-22617 SnapshotThread hits invariant due to reading oplog entries out of order
SERVER-22683 enableMajorityReadConcern option cannot be disabled if specified
SERVER-22728 if journaling is disabled, update durableOpTime when appliedOpTime updates
SERVER-22731 give correct error message when running initiate on a non-replset mongod
下载地址：https://www.mongodb.org/dl/win32/x86_64-2008plus-ssl
来自:开源中国社区
