Relatively new to docker, so if there is any info needed please ask away.
Trying to set up an overlay network so I can statically assign an IP to a container but I’m finding the Engine host is losing connectivity as soon as a container begins running. I performed the following steps on the boot2docker iso (that came with the latest Docker Toolbox I should add) booted up in VMware ESXi (as in, the first commands I specify after booting the ISO up)…
docker network create --subnet 172.16.10.0/24 svr
docker run --net svr --ip 172.16.10.161 -p 80:80
Edit: to clarify, my ip subnet my docker engine host has is on the 172.16.10.x subnet, and I’m wanting to create a new container with another IP but on the same subnet.
On running the docker run command, the host loses all network connectivity and my SSH session drops. This stays like this until I perform a docker stop on the host itself against the container that was started. Networking then returns to normal. If I delete the svr network, then bring up a container using 'docker run -p 80:80 ’ the container starts, and I can access the container without any problems using the host IP. In this state docker is working without any problems. I’ve therefore come to the conclusion there is some interesting network behaviour going on.
I also installed a new Alpine Linux server, installed the docker engine and performed the same tests with the exact same behaviour.
If I ‘docker exec ifconfig’ on the running container while the overlay network is there, and I’ve created this container to use an IP from that subnet, I can see the IP is configured correctly as I specified in the docker run command.
Google searches for similar behaviour haven’t returned much in the way of help unfortunately but I must be making a newbie mistake. or missing an argument. Anyone seen this at all?