arundasan91 commented on issue #1037: ssh: Could not resolve hostname export dmlc_role=worker; export dmlc_ps_root_port=9729; export dmlc_ps_root_uri=10.108.165.17; expor: Name or service not known URL: https://github.com/apache/incubator-mxnet/issues/1037#issuecomment-328127032 If this helps someone in the future, in my case I had to make sure that the hostnames are resolvable. I added the IP addresses to `/etc/hosts` file before running the distributed training. ``` $ cat /etc/hosts 127.0.0.1 localhost 10.40.0.252 das-1 10.40.0.46 mxnet-1 10.40.0.47 mxnet-2 10.40.0.52 mxnet-3 10.40.0.49 mxnet-4 10.40.0.48 mxnet-5 10.40.0.53 mxnet-6 10.40.0.51 mxnet-7 10.40.0.55 mxnet-8 10.40.0.54 mxnet-9 ... $ cat mxnet_hosts 10.40.0.252 10.40.0.46 10.40.0.47 10.40.0.52 10.40.0.49 10.40.0.48 10.40.0.53 10.40.0.51 10.40.0.55 10.40.0.54 ``` ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected]
With regards, Apache Git Services
