No disc detected during installation

[cosnk]

I have a small HP Proliant Microserver used in our small office.
Several months ago I had installed minimal mode of Fedora 30 Server for testing and there is no data file in the server.
I'd like to move to CentOS 8.
During installation, I need CentOS to wipe out Fedora 30.
But no disc detected during installation.
Please see the attached photos.
And what should be done next ?

[tunk]

Which model? What's the hardware specs, in particular, the disk (RAID) controller?

[cosnk]

HP ProLiant MicroServer G7 N40L
AMD TurionTM II Neo N40L (1.5GHz)
8GB of RAM
3.5" SATA drive:
250GB x 1 (used for Fedora 30 installation)
2TB x 2

[cosnk]

Help me please

[mathog]

You are most likely suffering from the issues discussed in this thread (kernel missing driver needed for your hardware):

viewtopic.php?f=54&t=73634&p=310316&hil ... og#p310316

[cosnk]

I have installed CentOS 7 instead.
CentOS 7 solved this problem.

[mathog]

CentOS 7 solved this problem.

7 had some driver your system needs, 8 does not, and so the latter won't install on some systems which were fine with the previous version.

It is more than a little annoying that the installer only offers up " no disk found" when it should at least process lspci to see what it is working with and note that there is no driver for some storage controller. What happened to you happened to many of us, and many of us were baffled why a system that worked at 7 did not work at 8. CentOS Plus would fix most of these, but that product does not have an installer with a kernel which includes the missing drivers, so getting there will be a battle.

[TrevorH]

Many of the missing drivers are packaged by ELRepo and can be used as Driver Update Disks at install time. You'd have to post the output from lspci -nn for anyone to be able to identify the exact model of controller and to check what is used for it.

[mathog]

Many of the missing drivers are packaged by ELRepo and can be used as Driver Update Disks at install time. You'd have to post the output from lspci -nn for anyone to be able to identify the exact model of controller and to check what is used for it.

Exactly - the fix is a bunch of unexpected extra steps which fall into the end user's lap at a time when the computer is basically unusable. It is possible that the network driver might also be missing, but I have yet to see anybody encounter that, so rather than:

Code: Select all

no disk found

It would be much nicer if it did:

Code: Select all

no disk found
network available? yes
check El Repo for missing drivers (Y/n)?
Y
connecting to El Repo
checking for missing drivers for this hardware
Found missing driver(s): 
  WHATEVER  storage control driver
Install missing drivers (Y/n)?
Y
Installing...
Scanning for disks...
Disk found

and off to the races. Only costing the end user two "Y" presses and a few seconds. Much better than the current situation, which requires research to understand the problem, figuring out a way to get the driver onto the machine, and starting the installation over from the beginning. Admittedly it requires pieces to be added to CentOS which were not originally in Red Hat's release.

Either that or come up with a CentOS Plus installer with a CentOS Plus kernel, which would likely fix most of these "no disk found" problems.