fix issue with rke2-install

[alknopfler]

What this PR does / why we need it:

This PR fix an issue #342 adding non-temporal bootstrap.success file to avoid systemd failure after reboot the system once provisioned.

Which issue(s) this PR fixes (optional, in fixes #<issue number>(, fixes #<issue_number>, ...) format, will close the issue(s) when PR gets merged):
Fixes #342

Special notes for your reviewer:
Some lint changes because go lint discovered during the PR creation

Checklist:

• squashed commits into logical changes
• includes documentation
• adds unit tests
• adds or updates e2e tests

[alknopfler]

Tests building local image:

• after provisioning a host:

mgmt-cluster:/home/metal3 # clusterctl describe cluster xr11-3-cluster2
NAME                                                     READY  SEVERITY  REASON  SINCE  MESSAGE
Cluster/xr11-3-cluster2                                  True                     11s
├─ClusterInfrastructure - Metal3Cluster/xr11-3-cluster2  True                     15m
└─ControlPlane - RKE2ControlPlane/xr11-3-cluster2        True                     11s
  └─Machine/xr11-3-cluster2-ldn7l                        True                     11s

the systemd:

localhost:~ # systemctl status rke2-install
○ rke2-install.service - rke2-install
     Loaded: loaded (/etc/systemd/system/rke2-install.service; enabled; vendor preset: enabled)
     Active: inactive (dead) since Fri 2024-06-07 08:20:02 EDT; 38s ago
    Process: 3780 ExecStart=/etc/rke2-install.sh (code=exited, status=0/SUCCESS)
   Main PID: 3780 (code=exited, status=0/SUCCESS)

Jun 07 08:19:13 localhost.localdomain rke2-install.sh[3789]: [INFO]  downloading tarball at https://github.com/rancher/rke2/releases/download/v1.28.7+rke2r1/rke2.lin>
Jun 07 08:19:14 localhost.localdomain rke2-install.sh[3789]: [INFO]  verifying tarball
Jun 07 08:19:14 localhost.localdomain rke2-install.sh[3789]: [INFO]  unpacking tarball file to /opt/rke2
Jun 07 08:19:16 localhost.localdomain rke2-install.sh[3789]: [INFO]  updating tarball contents to reflect install path
Jun 07 08:19:16 localhost.localdomain rke2-install.sh[3789]: [INFO]  moving systemd units to /etc/systemd/system
Jun 07 08:19:16 localhost.localdomain rke2-install.sh[3789]: [INFO]  install complete; you may want to run:  export PATH=$PATH:/opt/rke2/bin
Jun 07 08:19:18 localhost.localdomain /semanage[4086]: Successful:  resrc=fcontext op=add tglob="/usr/lib/systemd/system/rke2-server.service" ftype=any tcontext=syst>
Jun 07 08:19:18 localhost.localdomain rke2-install.sh[4096]: Created symlink /etc/systemd/system/multi-user.target.wants/rke2-server.service → /etc/systemd/system/rk>
Jun 07 08:20:02 localhost.localdomain systemd[1]: rke2-install.service: Deactivated successfully.
Jun 07 08:20:02 localhost.localdomain systemd[1]: Finished rke2-install.

and both files created:

localhost:~ # cat /etc/cluster-api/bootstrap-success.complete
success
localhost:~ # cat /run/cluster-api/bootstrap-success.complete
success

After rebooting the system:

localhost:~ # systemctl status rke2-install
○ rke2-install.service - rke2-install
     Loaded: loaded (/etc/systemd/system/rke2-install.service; enabled; vendor preset: enabled)
     Active: inactive (dead)
  Condition: start condition failed at Fri 2024-06-07 08:26:50 EDT; 33s ago

Jun 07 08:26:50 localhost.localdomain systemd[1]: Condition check resulted in rke2-install being skipped.

and only the /etc/ file is present (fixing the issue with ephimeral /run in sle micro):

localhost:~ # cat /run/
NetworkManager/  chrony-helper/   dbus/            ignition.env     issue.d/         mdadm/           netns/           setroubleshoot/  udev/
agetty.reload    cockpit/         dmeventd-client  initctl          k3s/             media/           nodeagent/       systemd/         user/
blkid/           containerd/      dmeventd-server  initramfs/       lock/            motd.d/          podman/          tmpfiles.d/      utmp
calico/          credentials/     faillock/        irqbalance/      log/             mount/           salt/            tpm2-tss/        xtables.lock
chrony/          cryptsetup/      fsck/            issue            lvm/             netconfig/       salt-tmp/        tuned/
localhost:~ # cat /etc/cluster-api/bootstrap-success.complete
success

[alexander-demicev]

looks good, can you fix linter warnings?
The problem with missing /run/cluster-api/bootstrap-success.complete after reboot still remains I guess since /run/ is ephemeral, but this issue can be addressed separately.

[furkatgofurov7]

/lgtm