Adding fedora28 distribution

[gdha]

• Missing RPM package to kick off ansible-playbooks

TASK [rear-test : Install the hosts file] **************************************
fatal: [server]: FAILED! => {"changed": false, "checksum": "c6e37c98e90d5a7e21a0ffd4c1d51fb7505e5683", "msg": "Aborting, target uses selinux but python bindings (libselinux-python) aren't installed!"}

[gdha]

fixed the missing packages by adding a task yaml file: install_prereq_pkgs_fedora.yml

[gdha]

TASK [rear-test : include_tasks] ***********************************************
included: /vagrant/centos7/ansible/common/roles/rear-test/tasks/update_rh_packages.yml for server

TASK [rear-test : update all packages] *****************************************
fatal: [server]: FAILED! => {"changed": false, "msg": "The Python 2 yum module is needed for this module. If you require Python 3 support use the `dnf` Ansible module instead."}

After adding some yum dependencies we go further, but we get now the following:

TASK [rear-test : update all packages] *****************************************
fatal: [client]: FAILED! => {"changed": true, "msg": "Yum command has been deprecated, use dnf instead.

all fixed

[gdha]

PXE boot recover fedora28 did not find ldlinux.c32 executable. Not to check if it was copied or even available on the client system.
this is fixed.

[gdha]

• problem 1:
  The real client was using predicitable LAN interface names while the recover was using the aliases (eth0/1). Seems same kind of issue we had in Recover VM network interfaces are biosdevnames instead of eth0/eth1 #6.
  Did a test on the client and removed the net.ifnames=0 from line:

KERNEL_CMDLINE="$KERNEL_CMDLINE net.ifnames=0"

The recover VM started to restore with this - excellent.

• problem2:
  After the restore the recover VM rebooted, but could not load the hard disk because it was missing the libcom32.c32 executable.
  Libcom32.c32 complaint vanished after adding the rule for it

[gdha]

For problem 1 have a look at /usr/lib/dracut/modules.d/40network/net-lib.sh is_persistent_ethernet_name() {

Added the required code on the client and it works fine - still need to merge it with the upstream

[gdha]

The recover VM tries to boot from HD and now it complains about a missing libutil.c32

[gdha]

Seems pstree command not found during recovery (not that important), but the recover worked fine with the latest added code.

[gdha]

provisioning failed at:

TASK [rear-test : disable RTC time in the local time zone] *********************
fatal: [server]: FAILED! => {"changed": true, "cmd": ["/usr/bin/timedatectl", "set-local-rtc", "0"], "delta": "0:00:00.024883", "end": "2018-12-05 13:52:51.524056", "msg": "non-zero return code", "rc": 1, "start": "2018-12-05 13:52:51.499173", "stderr": "Failed to set local RTC: hwclock failed: Child process exited with code 1", "stderr_lines": ["Failed to set local RTC: hwclock failed: Child process exited with code 1"], "stdout": "", "stdout_lines": []}

if you run it again then it works -- strange

[gdha]

Validation report: https://gist.github.com/d03251478980839148229ac7e164211a
Status GREEN

[gdha]

Fedora28 upgraded to fedora29 validation report: https://gist.github.com/28203b1bb8969345122f7cc3b8a5c269

Status: GREEN

[github-actions]

Stale issue message