NLKNguyen / alpine-mpich

MPI Cluster Automation Solution using Docker, based on Alpine Linux with MPICH (see IEEE paper)

Home Page:https://github.com/NLKNguyen/alpine-mpich

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

service logs of worker indicate Could not resolve hostname mpi-master or Connection refused

13ean opened this issue · comments

commented

I have created service
image

when inspect logs of worker, get infomation like this
image

Out of curiosity, what cloud provider are you using?

I had a real problem connecting to the containers on Google Cloud because the docker encryption on the subnet isn't supported by the NAT. I believe this is the case on AWS too:
moby/moby#37115

This isn't an issue on Digital Ocean though.

My solution was to remove the --opt encrypted line from swarm.sh

Another possible issue is that the firewall hasn't got all the necessary ports open.