If I launch a docker container with
docker run -v /foo:/foo ... I can see the contents of /foo on the host, from within the container.
While the docker container is running, if I run
mount -t ext4 /dev/... /foo/something I will NOT see the new mount point in /foo inside the container - is there any way to make it show up? (if I launch the docker container AFTER the mount point on the host is established, it is ok).
Docker containers run in a private mount namespace, which means that mounts made on the host after the container starts do not propagate into the container. The kernel documentation on shared subtrees goes into detail about mount propagation and private vs shared vs slave mounts.
The short answer to your question is that there isn't an easy way to expose a new mount like this into a container. It's possible, probably involving the use of nsenter to run commands inside the container namespace to change the flags on the mounts, but I wouldn't go there.
In general, if you need to change the storage configuration of a container, you re-deploy the container.
