miq 15 RL8 migration

[tjyang]

Hi

I did a POC test to convert oparin's CentOS Steam 8 OS to run on RL8.

• Download and sping up miq15 VMWare ova
• In a test lab with subnet like 192.168.1.0/24.
• Create a all-in-one miq instance by making adding a new disk as /dev/sdb and create a LV partition to /var/lib/postgresql
• Backup the VM.
• add this line "exclude=ansible,ansible-core" at the end of /etc/dnf/dnf.conf, without this line, migration script will failed.

echo "exclude=ansible,ansible-core" >> /etc/dnf/dnf.conf

• installonly_limit=2 in /etc/dnf/dnf.conf

sed -i 's#limit=3#limit=2#' /etc/dnf/dnf.conf

• remove splash bootup GUI with 3 dots in the middle.

sed -i 's#rhgb\squiet##'  /etc/default/grub
grub2-mkconfig -o /boot/grub2/grub.cfg

• Disable and stop miq

systemctl  disable --now httpd postgresql evmserverd
# systemctl disable --now  cloud-init

• download,chmod 755 and execute migrate2rocky.sh

wget https://raw.githubusercontent.com/learn-rocky/rocky-tools/main/migrate2rocky/migrate2rocky.sh && chmod 755 ./migrate2rocky.sh  && ./migrate2rocky.sh -r

• re-enable all miq related software.

systemctl  enable --now httpd postgresql evmserverd

• dnf repolist after migration to RL8, Centos Stream RPMs and repos were nicely replaced.

repo id                          repo name
ansible-runner                   Ansible Runner for EL 8 - x86_64
appstream                        Rocky Linux 8 - AppStream
baseos                           Rocky Linux 8 - BaseOS
epel                             Extra Packages for Enterprise Linux 8 - x86_64
extras                           Rocky Linux 8 - Extras
manageiq-14-najdorf              ManageIQ 14 (Najdorf) Release- x86_64
manageiq-14-najdorf-noarch       ManageIQ 14 (Najdorf) Release- noarch

• "cp /var/log/migrate2rocky.log /root" to keep a copy of migration log.
• Run "appliance_console" , Select 17 to exit after seeing the current miq status.
• Remove vmware snapshot backup.

[Fryguy]

See also ManageIQ/manageiq-rpm_build#364 regarding the exclude=ansible,ansible-core workaround.

[Fryguy]

Wow, very cool @tjyang ! Would you mind perhaps writing something for the manageiq-documentation with your steps?

[tjyang]

Submit PR to here https://github.com/ManageIQ/manageiq-documentation, right ?

[tjyang]

• the graphical boot up screen kind of stuck at last dot. I disabled the quiet mode and get the following last boot up job.
• Any idea why manageiq-web (or manageiq-web after) is taking so long to finish ?

[Fryguy]

It's probably whatever is after manageiq-web_service. @agrare do you know how this is supposed to work and what might be next?

[agrare]

None of the manageiq services should be blocking the user login tty. If you can ssh to the machine then systemctl status and systemctl should give you some good info

# systemctl status
● manageiq.localdomain
    State: running
     Jobs: 0 queued
   Failed: 0 units
   etc...

If you just run systemctl it will show you a list of all of the services with their states in a useful way to see what is still "starting"

[tjyang]

• Using miq 15.
• The boot up stuck at cloud-init for a long time.

[agrare]

Is cloud-init part of the rocky migration process? That isn't something that we set up as far as I know. You could probably systemctl disable the cloud-init service that is running then reboot.

[tjyang]

Is cloud-init part of the rocky migration process? That isn't something that we set up as far as I know.

I saw cloud-init was running when I first imported the ova and started up the VM.

You could probably systemctl disable the cloud-init service that is running then reboot.

I tried "systemctl disable cloud-init cloud-init-local" and rebooted.
But the splash screen with 3 dots come back again.
I do VMWare snapshots before I made big changes so I can roll back to previous good change.