我正在批量写入MongoDB并获得OOM异常（java.lang.OutOfMemoryError：超出了GC开销限制）。我有两个问题：
FO 2016-11-15 15：19：10,437-[TS] org.mongodb.driver.cluster info（71）-WritableServerSelector从群集描述ClusterDescription {type = UNKNOWN，connectionMode = MULTIPLE，all = [ ServerDescription {地址= mongo.server1-or：30000，类型=未知，状态= CONNECTING}，ServerDescription {地址= mongo.server2：30000，类型= UNKNOWN，状态= CONNECTING}，ServerDescription {地址= mongo.server3：30000，类型=未知，状态=连接}]}。等待30000毫秒，然后超时INFO 2016-11-15 15：19：11,448-[TS] org.mongodb.driver.cluster info（71）-WritableServerSelector从群集描述ClusterDescription {type = UNKNOWN，connectionMode = MULTIPLE，all = [ServerDescription {地址= mongo.server1或：30000，类型=未知，状态= CONNECTING}，ServerDescription {地址= mongo.server2：30000，类型=未知，状态= CONNECTING}，ServerDescription {地址= mongo.server3：30000，类型=未知，状态= CONNECTING}]}。等待30000毫秒，然后超时INFO 2016-11-15 15：19：14,324-[TS] org.mongodb.driver.cluster info（76）-连接到服务器mongo.server2：30000时监视线程异常com.mongodb.MongoException：java.lang.OutOfMemoryError：超出了GC开销限制在com.mongodb.connection.InternalStreamConnection.open（InternalStreamConnection.java:125）在com.mongodb.connection.DefaultServerMonitor $ ServerMonitorRunnable.run（DefaultServerMonitor.java:128）在java.lang.Thread.run（Thread.java:745）原因：java.lang.OutOfMemoryError：超出了GC开销限制INFO 2016-11-15 15：19：14,325-[TS] com.xyz.executors.ConsumeMessageTask run（45）-预定的线程池将写入MongoDBINFO 2016-11-15 15：19：14,325-[TS] org.mongodb.driver.connection info（71）-已打开与mongo.server3：30000的连接[connectionId {localValue：28690}]错误2016-11-15 15：19：17,353-[TS] com.xyz.executors.ConsumeMessageTask run（117）-批量写入MongoDB时发生错误com.mongodb.MongoTimeoutException：等待与WritableServerSelector匹配的服务器30000毫秒后超时。群集状态的客户端视图为{type = UNKNOWN，服务器= [{address = mongo.server1-or：30000，type = UNKNOWN，state = CONNECTING}，{address = mongo.server2：30000，type = UNKNOWN，state = CONNECTING ，异常= {com.mongodb.MongoException：java.lang.OutOfMemoryError：超出了GC开销限制}，是由于{java.lang.OutOfMemoryError：超出了GC开销限制}}，{地址= mongo.server3：30000，类型= UNKNOWN ，state = CONNECTING}]在com.mongodb.connection.BaseCluster.createTimeoutException（BaseCluster.java:369）在com.mongodb.connection.BaseCluster.selectServer（BaseCluster.java:101）在com.mongodb.binding.ClusterBinding $ ClusterBindingConnectionSource。（ClusterBinding.java:75）在com.mongodb.binding.ClusterBinding $ ClusterBindingConnectionSource。（ClusterBinding.java:71）在com.mongodb.binding.ClusterBinding.getWriteConnectionSource（ClusterBinding.java:68）在com.mongodb.operation.OperationHelper.withConnection（OperationHelper.java:219）在com.mongodb.operation.MixedBulkWriteOperation.execute（MixedBulkWriteOperation.java:168）在com.mongodb.operation.MixedBulkWriteOperation.execute（MixedBulkWriteOperation.java:74）在com.mongodb.Mongo.execute（Mongo.java:781）在com.mongodb.Mongo $ 2.execute（Mongo.java:764）在com.mongodb.DBCollection.executeBulkWriteOperation（DBCollection.java:2195）在com.mongodb.DBCollection.executeBulkWriteOperation（DBCollection.java:2188）在com.mongodb.BulkWriteOperation.execute（BulkWriteOperation.java:121）在com.xyz.executors.ConsumeMessageTask.run（ConsumeMessageTask.java:112）在java.util.concurrent.Executors $ RunnableAdapter.call（Executors.java:471）在java.util.concurrent.FutureTask.run（FutureTask.java:262）在java.util.concurrent.ThreadPoolExecutor.runWorker（ThreadPoolExecutor.java:1145）在java.util.concurrent.ThreadPoolExecutor $ Worker.run（ThreadPoolExecutor.java:615）在java.lang.Thread.run（Thread.java:745）`
您的堆栈跟踪显示错误源自此行：
at com.mongodb.connection.BaseCluster.createTimeoutException(BaseCluster.java:369)
at com.mongodb.connection.BaseCluster.selectServer(BaseCluster.java:101)...
因此，客户端在尝试连接到服务器时会超时。从服务器收到的信息在上面的行中：
Client view of cluster state is {type=UNKNOWN, servers=[
{address=mongo.server1-or:30000, type=UNKNOWN, state=CONNECTING},
{address=mongo.server2:30000, type=UNKNOWN, state=CONNECTING,
exception={com.mongodb.MongoException: java.lang.OutOfMemoryError: GC overhead limit exceeded}, caused by {java.lang.OutOfMemoryError: GC overhead limit exceeded}},
{address=mongo.server3:30000, type=UNKNOWN, state=CONNECTING}]
这意味着OutOfMemoryError来自mongo.server2。最好检查该服务器的日志以查看其发生原因，但是大量的写作可能与此有关。可能是mongo.server2跟不上同步您写入主服务器的数据。