now I run the "example/image-classification" with two machines,long time nothing respond

[n080509]

now I run the "example/image-classification" with two machines
10.136.159.116
10.136.159.117

I run the the command on the 10.136.159.116 as follow,but nothing respond,pls help me ,how can I do ?

root@116:~/mxnet/example/image-classification#../../tools/launch.py -n 2 --launcher ssh -H hosts python train_mnist.py --network lenet --kv-store dist_sync

[qiaohaijun]

Is single machine ok?

[n080509]

now i run it as follow,how can i do next

root@116:~/mxnet/example/image-classification# export PS_VERBOSE=1; ../../tools/launch.py -n 2 -H hosts --sync-dst-dir /tmp/mxnet python train_mnist.py --kv-store dist_sync
python train_mnist.py --kv-store dist_sync
2016-05-12 05:34:37,993 INFO rsync /home/slave/mxnet/example/image-classification/ -> 10.136.159.116:/tmp/mxnet
[05:34:38] src/van.cc:65: Bind to role=scheduler, id=1, ip=0.0.0.116, port=9098
2016-05-12 05:34:38,203 INFO rsync /home/slave/mxnet/example/image-classification/ -> 10.136.159.117:/tmp/mxnet
^CTraceback (most recent call last):
File "../../tools/launch.py", line 79, in
main()
File "../../tools/launch.py", line 64, in main
ssh.submit(args)
File "/home/slave/mxnet/tools/../dmlc-core/tracker/dmlc_tracker/ssh.py", line 76, in submit
pscmd=(' '.join(args.command)))
File "/home/slave/mxnet/tools/../dmlc-core/tracker/dmlc_tracker/tracker.py", line 422, in submit
pserver.join()
File "/home/slave/mxnet/tools/../dmlc-core/tracker/dmlc_tracker/tracker.py", line 371, in join
self.thread.join(100)
File "/usr/lib/python2.7/threading.py", line 960, in join
self.__block.wait(delay)
File "/usr/lib/python2.7/threading.py", line 359, in wait
_sleep(delay)
KeyboardInterrupt

[n080509]

2台机器上运行不行

[n080509]

后来试了一下单机上多WORK也不行
root@116:~/mxnet/example/image-classification# ../../tools/launch.py -n 2 -s 1 python train_mnist.py --kv-store dist_sync
python train_mnist.py --kv-store dist_sync
[06:29:44] src/van.cc:65: Bind to role=scheduler, id=1, ip=0.0.0.116, port=9102
[06:29:44] src/van.cc:65: Bind to role=server, ip=10.136.159.116, port=47526
[06:29:44[] src/van.cc:65: Bind to role=worker, ip=10.136.159.116, port=45587
06:29:44] src/van.cc:65: Bind to role=worker, ip=10.136.159.116, port=39873

^CTraceback (most recent call last):
File "../../tools/launch.py", line 79, in
main()
File "../../tools/launch.py", line 55, in main
local.submit(args)
File "/home/slave/mxnet/tools/../dmlc-core/tracker/dmlc_tracker/local.py", line 83, in submit
pscmd=(' '.join(args.command)))
File "/home/slave/mxnet/tools/../dmlc-core/tracker/dmlc_tracker/tracker.py", line 422, in submit
pserver.join()
File "/home/slave/mxnet/tools/../dmlc-core/tracker/dmlc_tracker/tracker.py", line 371, in join
self.thread.join(100)
File "/usr/lib/python2.7/threading.py", line 960, in join
self.__block.wait(delay)
File "/usr/lib/python2.7/threading.py", line 359, in wait
_sleep(delay)
KeyboardInterrupt
root@116:~/mxnet/example/image-classification#

[n080509]

单机上运行 python train_mnist.py --network lenet 是OK的

[qiaohaijun]

你配置了ssh 无密码登录了吗？
单机多卡，不是那么配置的，你可以通过qq联系我