Enable agent restart without impacting running services

Cargo.toml

@@ -5,11 +5,14 @@ authors = ["Sönke Liebau <soenke.liebau@stackable.de>"]
 edition = "2018"
 
 [dependencies]
-# We are currently referencing the Krustlet directly from the repository, because some features that we are using have
-# (exponential backoff most prominently) have not yet been included in a release
-# We will look to move this to officially released versions as soon as possible
-kubelet = { git="https://github.com/deislabs/krustlet.git", rev="ac218b38ba564de806568e49d9e38aaef9f41537", default-features = true, features= ["derive", "cli"] }
-oci-distribution = { git="https://github.com/deislabs/krustlet.git", rev="ac218b38ba564de806568e49d9e38aaef9f41537"}
+# We are currently referencing the Krustlet directly from a Stackable fork of the official repository.
+# There are two reasons for this, the fork is needed to remove the node draining behavior of the Krustlet (see
+# https://github.com/deislabs/krustlet/issues/523)
+# Our fork / the referenced commit is quite a bit behind the official repo, as there were some breaking changes in the
+# 0.6 release that we need to adapt the Krustlet to.
+# We will look to move to the latest version as soon as possible, but may need to continue releasing from a fork.
+kubelet = { git="https://github.com/stackabletech/krustlet.git", rev="bb8bb42c9400a565df4be04f357e61934fb277c6", default-features = true, features= ["derive", "cli"] }
+oci-distribution = { git="https://github.com/stackabletech/krustlet.git", rev="bb8bb42c9400a565df4be04f357e61934fb277c6"}
 k8s-openapi = { version = "0.9", default-features = false, features = ["v1_18"] }
 kube = { version= "0.42", default-features = false, features = ["native-tls"] }
 kube-derive = "0.43"

src/provider/states/starting.rs

@@ -20,6 +20,28 @@ impl State<PodState> for Starting {
     async fn next(self: Box<Self>, pod_state: &mut PodState, _: &Pod) -> Transition<PodState> {
         if let Some(systemd_units) = &pod_state.service_units {
             for unit in systemd_units {
+                match pod_state.systemd_manager.is_running(&unit.get_name()) {
+                    Ok(true) => {
+                        debug!(
+                            "Unit [{}] for service [{}] already running, nothing to do..",
+                            &unit.get_name(),
+                            &pod_state.service_name
+                        );
+                        // Skip rest of loop as the service is already running
+                        continue;
+                    }
+                    Err(dbus_error) => {
+                        debug!(
+                            "Error retrieving activestate of unit [{}] for service [{}]: [{}]",
+                            &unit.get_name(),
+                            &pod_state.service_name,
+                            dbus_error
+                        );
+                        return Transition::Complete(Err(dbus_error));
+                    }
+                    _ => { // nothing to do, just keep going
+                    }
+                }
                 info!("Starting systemd unit [{}]", unit);
                 if let Err(start_error) = pod_state.systemd_manager.start(&unit.get_name()) {
                     error!(