Docker host solution

docker-compose.yml

master:
  image: centos
  hostname: master
  net: my-swarm-network
  command: /bin/sh -c "while true; do echo executing master; sleep 1; done"
slave:
  image: centos
  net: my-swarm-network
  command: /bin/sh -c "while true; do echo executing slave; sleep 1; done"
  

Instructions

$ ./swarm-create.sh

$ eval "$(docker-machine env --swarm swarm-master)"

$ docker network create --driver overlay my-swarm-network

$ docker-compose up -d

$ docker exec -ti swarm-node-02/rafaelbenevides_master_1 ping master
PING master (10.0.0.3) 56(84) bytes of data.
64 bytes from master (10.0.0.3): icmp_seq=1 ttl=64 time=0.032 ms
64 bytes from master (10.0.0.3): icmp_seq=2 ttl=64 time=0.048 ms

$ docker exec -ti swarm-node-02/rafaelbenevides_master_1 ping rafaelbenevides_slave_1
PING rafaelbenevides_slave_1 (10.0.0.2) 56(84) bytes of data.
64 bytes from rafaelbenevides_slave_1 (10.0.0.2): icmp_seq=1 ttl=64 time=1.41 ms

$ docker exec -ti swarm-node-01/rafaelbenevides_slave_1 sh -c 'HOST=$(hostname); ping $HOST'
PING 6ff0763e7495 (10.0.0.2) 56(84) bytes of data.
64 bytes from 6ff0763e7495 (10.0.0.2): icmp_seq=1 ttl=64 time=0.031 ms

swarm-create.sh

cho "Creating Multi Host Keystore"
docker-machine create -d virtualbox mh-keystore
eval "$(docker-machine env mh-keystore)"
echo "Starting Consul at Keystore Machine"
docker run -d -p "8500:8500" -h "consul"  progrium/consul -server -bootstrap
echo "Creating Swarm master ..."
docker-machine create -d virtualbox --swarm --swarm-master --swarm-strategy "spread" --swarm-discovery="consul://$(docker-machine ip mh-keystore):8500" --engine-opt="cluster-store=consul://$(docker-machine ip mh-keystore):8500" --engine-opt="cluster-advertise=eth1:2376" swarm-master
echo "Creating Swarm node 01 ..."
docker-machine create -d virtualbox --swarm --swarm-discovery="consul://$(docker-machine ip mh-keystore):8500" --engine-opt="cluster-store=consul://$(docker-machine ip mh-keystore):8500" --engine-opt="cluster-advertise=eth1:2376"  swarm-node-01
echo "Creating Swarm node 02 ..."
docker-machine create -d virtualbox --swarm --swarm-discovery="consul://$(docker-machine ip mh-keystore):8500" --engine-opt="cluster-store=consul://$(docker-machine ip mh-keystore):8500" --engine-opt="cluster-advertise=eth1:2376"  swarm-node-02

That's right! The generated internal hostname doesn't follow the "docker-compose" convention.

If you force the hostname="slave" on slaves and scale it, there will be rafaelbenevides_slave_1, rafaelbenevides_slave_2 and rafaelbenevides_slave_3 for master, but all of them will call itself as "slave". Which makes sense because we forced a hostname.

To check what was the generated hostname for the slave instance, run the following:

    $ docker exec -ti swarm-node-02/rafaelbenevides_master_1 cat /etc/hosts

You will see something like this:

    10.0.0.3    master
    127.0.0.1   localhost
    ::1 localhost ip6-localhost ip6-loopback
    fe00::0 ip6-localnet
    ff00::0 ip6-mcastprefix
    ff02::1 ip6-allnodes
    ff02::2 ip6-allrouters
    10.0.0.2    rafaelbenevides_slave_1
    10.0.0.2    rafaelbenevides_slave_1.my-swarm-network

Now that you found that the generated hostname for slave is rafaelbenevides_slave_1, try to ping that hostname address from the slave instance itself:

    $ docker exec -ti swarm-node-01/rafaelbenevides_slave_1 ping rafaelbenevides_slave_1

The command above fails because /etc/hosts of the slave instance does not have the generated hostname entry the same way as in master's.

That is the issue.