Dear Maintainer,

   * What led up to the situation?
I wanted to use virt-manager to install a debian s390x guest on an amd64 host
   * What exactly did you do (or not do) that was effective (or
     ineffective)?
I used virt-manager in the ordinary way, and selected the s390x architecture
   * What was the outcome of this action?
virt-manager got as far as attempting to boot the guest but then indicated that
the guest was not running.
   * What outcome did you expect instead?
A dialogue with the debian installer resulting in a viable s390x installation.

I have previously used a Ubuntu 26.04 amd64 host system to install a RHEL10
s390x guest with no problem.

Also when trying to do the install using a command such as
qemu-system-s390x -M s390-ccw-virtio \
-monitor stdio \
-name "guest" \
-m 4096 \
-mem-prealloc \
-kernel /home/tjcw/Downloads/dvd1/linux_vm \
-initrd /home/tjcw/Downloads/dvd1/root.bin \
-blockdev
file,node-name=c1,filename=/home/tjcw/Downloads/debian-13.4.0-s390x-DVD-1.iso
\
                  -device virtio-scsi \
                  -device scsi-cd,drive=c1 \
-netdev user,id=net0,hostfwd=tcp::2222-:22 \
    -device virtio-net-ccw,netdev=net0,id=net0_dev \
-drive
file=/home/tjcw/VMs/s390x-vm.qcow2,format=qcow2,if=none,id=drive-sata0-0-0 \
-device virtio-blk-pci,drive=drive-sata0-0-0,id=sata0-0-0
(on a debian trixie system) I could boot linux and get as far as running
the installer if I uses explicit kernel and initrd as above, but the
corresponding command attempting to use the kernel and ramdisk in the
'.iso' would not boot.
I wasn't able to do the install with 'qemu-system-s390x' because I could
not get networking to work correctly.

I am now trying to install from a RHEL10 s390x DVD. This seems to be going
OK, it certainly boots and runs code from the initrd. This suggests that
the problem is with the Debian trixie s390x ISO, and not with virt-manager.

There is a secondary problem with the 'virt-*' support, though. The first
attempt to start the DVD failed saying that network 'default' is not
active, and when I run 'virsh net-start default' I get a message that the
network can't be started because the ethernet is already active. To get
round this I click in the top right corner and turn 'wired' off
(network-manager, I think) and then 'virsh net-start default' succeeds and
I can turn 'wired' on again.

I mounted the iso on /mnt and extracted the kernel and ramdisk from
/mnt/boot .
Then I went to virt-manager and selected 'additional configuration before
boot', selected direct kernel boot, and entered the file names for the
kernel and ramdisk.

That done, the VM booted and progressed with the install.

So it looks that the problem is that the ISOs are set up in a way which
does not allow the virtual boot firmware to find the kernel and ramdisk.

Is this fixable ?

The install completed normally, but after that I needed to use 'virsh edit'
to delete the lines in the XML for kernel and initrd. Before I deleted
those, the system would boot into the installer again.

That fixed, I now have a viable emulated s390x debian system in my amd64
host.