search

mazgi / docker-cuda-cv

based on: https://gitlab.com/nvidia/cuda
https://hub.docker.com/r/mazgi/cuda-cv/
BSD 3-Clause "New" or "Revised" License
0 stars 0 forks source link

Can not run chainermn #3

Open mazgi opened 6 years ago

mazgi commented 6 years ago 
$ nvidia-docker run -v $(pwd):/mnt mazgi/cuda-cv:9.0-cudnn7-devel-ubuntu16.04 mpiexec --allow-run-as-root -n 2 python3 /mnt/chainermn/examples/mnist/train_mnist.py --gpu -1
--------------------------------------------------------------------------
The value of the MCA parameter "plm_rsh_agent" was set to a path
that could not be found:

  plm_rsh_agent: ssh : rsh

Please either unset the parameter, or check that the path is correct
--------------------------------------------------------------------------
[c2d786377820:00001] [[INVALID],INVALID] FORCE-TERMINATE AT Not found:-13 - error plm_rsh_component.c(327)
[c2d786377820:00001] *** Process received signal ***
[c2d786377820:00001] Signal: Segmentation fault (11)
[c2d786377820:00001] Signal code: Address not mapped (1)
[c2d786377820:00001] Failing at address: (nil)
[c2d786377820:00001] [ 0] /lib/x86_64-linux-gnu/libpthread.so.0(+0x11390)[0x7f3950070390]
[c2d786377820:00001] *** End of error message ***
zrss commented 5 years ago

... it seems that ssh client not found