Docker bind mounts files as owned by nobody

When bind mounting files, you may run into permission issues with creating files or folders. For example: ¹

Error response from daemon: failed to create task for container: failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: error during container init: error mounting "/var/lib/docker/231072.231072/volumes/2c5b94522b3c1bec7de4421e247b225664b5cdda47c4fe587c3c2b5d73dbb409/_data" to rootfs at "/wiki/node_modules": create mountpoint for /wiki/node_modules mount: make mountpoint "/wiki/node_modules": mkdirat /var/lib/docker/231072.231072/overlay2/238c36d31794b699ac6c41eb7d58b94a764c545ad4f949703821c8c6ed4afdf1/merged/wiki/node_modules: permission denied

This is because the files that have been mounted may be owned by a user in the container that does not exist on your host system. To fix this issue, you may need to either: ¹

• Run the container as a user that does exist on the host by adding user: <UID>:<GID> to the docker compose configuration (replacing <UID> and <GID> with a UID and GID that exists on the host).
• Disable user namespace isolation for the container by adding userns_mode: "host", as your user namespace isolation configuration may map the container user to a user on the host that does not exist.