This is definitely worded a bit awkwardly. Let me take a stab at some corrections:
Use SSH to connect to the manager0 and consul0 instance, and run the following on each:
$ ifconfig
From the output, copy the eth0 IP address from inet addr. These will be used in Step 5.
On the consul0 instance, run the following command to bring up a consul service:
$ docker run -d -p 8500:8500 --name=consul progrium/consul -serv
If this makes more sense, I can do a PR to help clean up this tutorial. It looks like it could use another pass or two as well to iron it out a bit more.
I think the etc0 might be a reference left over from when the guide may have been using etcd as a backend instead of consul. The instruction to copy that command out to a file does seem like unnecessary extra work.
Ha the forum’s search feature automatically found this thread as I started a post about being totally confused by the Swarm install guide.
Yes please, make those corrections. And maybe all the commands should use --restart always.
Step 5 item 1 example is missing “:8500” at the end.
Step 6 says SSL is being used, but I don’t think that’s true, port “:4000” is plain http and certs haven’t been configured. Maybe docker-machine can do that?
These docs refer to port 2375 which is not SSL or even active in recent docker. I guess for the instructions to work you have to run this on each node. docker run -d --restart always -p 2375:2375 --volume=/var/run/docker.sock:/var/run/docker.sock --name=docker-http sequenceiq/socat