Add parameter for when podman driver is running remote

[afbjorklund]

Currently the remote flag of podman is hardcoded, to always disabled on Linux (and always enabled on Mac and Win)

We should make it a driver boolean, so that it is possible to test podman machine also on regular developer machines.

Sadly, podman requires a flag --remote or even a special binary podman-remote. Otherwise it will still run locally...

On other platforms, this flag is "always on" and the binary has been renamed/replaced. So this makes Linux "special"

---

Note: for the Docker driver, remote is always true. You can't run docker locally, you always need a dockerd running.

Hopefully the Podman behaviour will be fixed* in version 4.0, but minikube still needs to support Podman version 3 as well.

* see containers/podman#11196

Needed for: Issue #8003

[afbjorklund]

Currently this is intermixed with rootless, so might need to be fixed in cooperation with #12460 (for #8719)

        // Quickly returns an error code if service is not running
        cmd := exec.CommandContext(ctx, oci.Podman, "version", "--format", "{{.Server.Version}}")
        // Run with sudo on linux (local), otherwise podman-remote (as podman)
        if runtime.GOOS == "linux" {
                cmd = exec.CommandContext(ctx, "sudo", "-k", "-n", oci.Podman, "version", "--format", "{{.Version}}")
                cmd.Env = append(os.Environ(), "LANG=C", "LC_ALL=C") // sudo is localized
        }

So this affects both PrefixCmd and OCIBinary. Preferrably these should be a class, instead of a string ?

• Binary
• Rootless
• Remote

Binary	root local	user local	root remote	user remote
Docker	docker	docker	docker	docker
Podman	sudo -n podman	podman	podman --remote	podman --remote

---

EDIT: Also breaks with --cgroup-manager

        if ociBin == Podman && runtime.GOOS == "linux" {
                args = append(args, "--cgroup-manager", "cgroupfs")
        }

[afbjorklund]

When running remote, "rootless" is not so important to the command (it will never use sudo anyway).

But it should be possible, to query the engine if it is running as root or as user (it is in the info output)

"rootless"

"rootless: true"

docker root

 Security Options:
  apparmor
  seccomp
   Profile: default

docker rootless

 Security Options:
  seccomp
   Profile: default
  rootless

podman root

  remoteSocket:
    exists: true
    path: /run/podman/podman.sock
  rootless: false

podman rootless

  remoteSocket:
    exists: true
    path: /run/user/1000/podman/podman.sock
  rootless: true

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue or PR as fresh with /remove-lifecycle stale
• Mark this issue or PR as rotten with /lifecycle rotten
• Close this issue or PR with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

[k8s-triage-robot]

The Kubernetes project currently lacks enough active contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue or PR as fresh with /remove-lifecycle rotten
• Close this issue or PR with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle rotten