Hama0.7.1 的坑

1. 在GraphJobRunner中的loadVertices函数中，增加打印读取的顶点，就可以避免如下问题:

owest notify.
16/03/26 20:57:27 DEBUG sync.ZooKeeperSyncClientImpl: leaveBarrier() at superstep: 3 taskid:attempt_201603262059_0001_000004_0 lowest notify other nodes.
16/03/26 20:57:27 DEBUG sync.ZooKeeperSyncClientImpl: leaveBarrier() !!! checking znodes contnains /ready node or not: at superstep:3 znode:[ready]
16/03/26 20:57:27 DEBUG sync.ZooKeeperSyncClientImpl: leaveBarrier() at superstep:3 znode size: (0) znodes:[]
16/03/26 20:57:27 DEBUG ipc.Client: IPC Client (591203588) connection to localhost/127.0.0.1:50001 from liuqiang2 sending #67
16/03/26 20:57:27 DEBUG ipc.Client: IPC Client (591203588) connection to localhost/127.0.0.1:50001 from liuqiang2 got value #67
16/03/26 20:57:27 DEBUG ipc.RPC: Call: statusUpdate 2
16/03/26 20:57:27 INFO graph.GraphJobRunner: Total time spent for broadcasting aggregation values: 118 ms
16/03/26 20:57:27 ERROR bsp.BSPTask: Error running bsp setup and bsp function.
java.lang.NullPointerException
	at org.apache.hama.util.UnsafeByteArrayInputStream.<init>(UnsafeByteArrayInputStream.java:63)
	at org.apache.hama.util.WritableUtils.unsafeDeserialize(WritableUtils.java:63)
	at org.apache.hama.graph.MapVerticesInfo.get(MapVerticesInfo.java:101)
	at org.apache.hama.graph.GraphJobRunner$ComputeRunnable.<init>(GraphJobRunner.java:368)
	at org.apache.hama.graph.GraphJobRunner.doSuperstep(GraphJobRunner.java:278)
	at org.apache.hama.graph.GraphJobRunner.bsp(GraphJobRunner.java:188)
	at org.apache.hama.bsp.BSPTask.runBSP(BSPTask.java:171)
	at org.apache.hama.bsp.BSPTask.run(BSPTask.java:144)
	at org.apache.hama.bsp.GroomServer$BSPPeerChild.main(GroomServer.java:1255)
16/03/26 20:57:27 DEBUG bsp.Counters: Adding TASK_OUTPUT_RECORDS

原因我也不知道，但是就是可以！ 怀疑是concurrencyHashMap的并发问题，真TM神奇。。。

2. zookeeper的同步问题，记得随时:

bin/zookeepers.sh  ~/zookeeper-3.4.5/bin/zkCli.sh -server 127.0.0.1:21810

删除/bsp下面的所有数据，否则会有稀奇古怪的问题。。。。