面对一个问题，我们在重启后有5个节点的集群dse尝试启动但没有成功，system.log中的最后一条记录如下... 尝试堆和48和64，节点有128GB。其中三个开始但这两个不能，日志中只有那个记录没有错误。
INFO [main] 2017-05-16 21:16:27,507 CassandraDaemon.java:487 - JVM参数：[ - DDseserver_process，-XX：+ AlwaysPreTouch，-Dcassandra.disable_auth_caches_remote_configuration = false，-Dcassandra.force_default_indexing_page_size = false，-Dcassandra.join_ring = true，-Dcassandra.load_ring_state = true，-Dcassandra.write_survey = false，-XX：CMSInitiatingOccupancyFraction = 75，-XX：CMSWaitDuration = 10000，-ea，-XX：G1RSetUpdatingPauseTimePercent = 5，-XX： + HeapDumpOnOutOfMemoryError，-Xms16G，-Djava.net.preferIPv4Stack = true，-XX：MaxGCPauseMillis = 500，-Xmx16G，-XX：MaxTenuringThreshold = 1，-Xss256k，-XX：+ PerfDisableSharedMem，-XX：+ ResizeTLAB，-XX： StringTableSize = 1000003，-XX：SurvivorRatio = 8，-XX：ThreadPriorityPolicy = 42，-XX：+ UseThreadPriorities，-XX：+ UseTLAB，-XX：+ UseG1GC，-Dcom.sun.management.jmxremote.authenticate = false， - Dcassandra.jmx.local.port = 7199，-XX：CompileCommandFile = / etc / dse / cassandra / hotspot_compiler，-javaagent：/usr/share/dse/cassandra/lib/jamm-0.3.0.jar,-Djava.library 。路径= / USR /共享/ DS E / hadoop2客户端/ lib目录/原生：在/ usr /共享/ DSE /卡桑德拉/ lib目录/ SIGAR斌：在/ usr /共享/ DSE / hadoop2客户端/ lib目录/原生：在/ usr /共享/ DSE /卡桑德拉/ lib目录/ sigar-bin：， - Dguice_include_stack_traces = OFF，-Ddse.system_memory_in_mb = 128658，-Dcassandra.config.loader = com.datastax.bdp.config.DseConfigurationLoader，-Dguice_include_stack_traces = OFF，-Ddse.system_memory_in_mb = 128658，-Dcassandra。 config.loader = com.datastax.bdp.config.DseConfigurationLoader，-Dlogback.configurationFile = logback.xml，-Dcassandra.logdir = / var / log / cassandra，-Dcassandra.storagedir = / usr / share / dse / data， - Dcassandra-pidfile = / var / run / dse / dse.pid，-Dgraph-enabled = true，-XX：HeapDumpPath = / var / lib / cassandra / java_1494958565.hprof，-XX：ErrorFile = / var / lib / cassandra / hs_err_1494958565.log，-Dguice_include_stack_traces = OFF，-Ddse.system_memory_in_mb = 128658，-Dcassandra.config.loader = com.datastax.bdp.config