Hello,
I’m trying to deploy out some older PPC64 IBM machines (8335-GTB) but running into an issue where they constantly reboot during pxe boot. I was able to deploy them on an older MaaS 2.9 version but haven’t been able to deploy them since upgrading to 3.3 (currently 3.3.6).
In the MaaS UI logs, it just continually loops to pxe boot.
I have the Ubuntu images for the ppc64 and the machine was commissioned successfully.
Coincidentally, I’m able to boot them into rescue mode without a problem so I’m hoping I’m overlooking a small detail to get Ubuntu installed and booted. Thanks!
So I think I narrowed this specific looping error down to the Jammy OS deployment (ubuntu 22). If I change to Focal (Ubuntu 20), then I’m able to at least get to the curatin install / etc…
So something is wrong with the Jammy PXE/TFTP boot order / process for PPC64el.
If I use Focal (or any other OS), curtain doesn’t seem to install the OS to the disk and will continue to loop as well. The disks are configured so not sure what is going on
I eventually added a “ipmitool chassis bootdev disk” in my install script but since the OS isn’t installed to the disk, there isn’t anything to boot from and the deployment times out after 30 mins.
I can “manually” install the OS with by adding the Ubuntu Focal netboot kernel and initrd:
http://ports.ubuntu.com/ubuntu-ports/dists/focal/main/installer-ppc64el/current/legacy-images/netboot/ubuntu-installer/ppc64el/vmlinux
http://ports.ubuntu.com/ubuntu-ports/dists/focal/main/installer-ppc64el/current/legacy-images/netboot/ubuntu-installer/ppc64el/initrd.gz
So all subsystems / disks /etc… are fine, I’m guessing something is “wrong” in the curtain install? Any thoughts?
