分布式运行的错误——java/lang/NoClassDefFoundError: java/lang/Object

请教一下：

我在完成了分布式编译安装之后，当运行 bash /home/network/euler/tf_euler/scripts/dist_tf_euler.sh 时（dist_tf_euler.sh 里面的参数都已经改成本集群的数值），出现了

2019-03-22 09:58:04.826351: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
2019-03-22 09:58:04.830087: I tensorflow/core/distributed_runtime/rpc/grpc_channel.cc:222] Initialize GrpcChannelCache for job ps -> {0 -> 172.20.10.93:2222}
2019-03-22 09:58:04.830106: I tensorflow/core/distributed_runtime/rpc/grpc_channel.cc:222] Initialize GrpcChannelCache for job worker -> {0 -> 172.20.10.86:2222, 1 -> localhost:2222}
2019-03-22 09:58:04.831129: I tensorflow/core/distributed_runtime/rpc/grpc_server_lib.cc:381] Started server with target: grpc://localhost:2222
WARNING: Logging before InitGoogleLogging() is written to STDERR
I0322 09:58:04.832376 22400 remote_graph.cc:91] Initialize RemoteGraph, connect to server monitor: [172.20.11.86:2181, /home/network/zhangyongxing/zookeeper-3.4.9]

Error occurred during initialization of VM
java/lang/NoClassDefFoundError: java/lang/Object

其中，dist_tf_euler.sh 内容是
python -m tf_euler \
--data_dir hdfs://172.20.11.86:8020/ppi \
--euler_zk_addr 172.20.11.86:2181 \
--euler_zk_path /home/network/zookeeper-3.4.9 \
--max_id 56944 --feature_idx 1 --feature_dim 50 --label_idx 0 --label_dim 121 \
--model graphsage_supervised --mode train \
--ps_hosts=$PS_HOSTS \
--worker_hosts=$WORKER_HOSTS \
--job_name=worker --task_index=1

其中 PS_HOSTS 与 WORKER_HOSTS 都已指定。

谢谢！

Start with scripts/dist_tf_euler.sh and the shown `python -m tf_euler` command, then inspect the distributed runtime setup around the Java VM error. Reproduce the command with the stated PS_HOSTS, WORKER_HOSTS, HDFS path, and ZooKeeper address; done means identifying the environment or configuration cause of `java/lang/NoClassDefFoundError`.

Written by the indexing model from the issue text.