The following command should create a new docker machine on a shiny new Amazon EC2 instance:
docker-machine \
--storage-path /path/to/folder/docker_machines \
create \
--driver amazonec2 \
--amazonec2-access-key <my key> \
--amazonec2-secret-key <my secret> \
--amazonec2-vpc-id <my vpc> \
--amazonec2-region <my region> \
--amazonec2-zone <my AZ> \
--amazonec2-security-group <existing Sec Grp> \
--amazonec2-ami ami-da05a4a0 \
--amazonec2-ssh-keypath /path/to/private/key \
--engine-install-url=https://web.archive.org/web/20170623081500/https://get.docker.com \
awesome-new-docker-machine
I ran this command once, and encountered a legitimate problem (bad path to private key). Once I fixed that and ran the command again, I get this error:
Host already exists: "awesome-new-docker-machine"
However, I can't find this docker machine anywhere:
$ docker-machine ls
NAME ACTIVE DRIVER STATE URL SWARM DOCKER ERRORS
I even tried a docker-machine rm
and docker-machine kill
just for giggles. No difference.
I can't see a new EC2 instance on Amazon having been created from the first, erroneous run of the command.
How can I "clean up" whatever's existing (somewhere) so I can recreate the machine correctly?
So, it turns out that the first run of the command created some initial artifacts in a new folder awesome-new-docker-machine
under /path/to/folder/docker_machines
.
Deleting this folder and trying again worked perfectly.