[SOLVED] Unable to boot after fresh install

[linucksproxmox]

Hi, I was moving some servers around and had to power down my Proxmox 6.1 machine (HP Proliant DL380P G8). After powering it back on, it would not boot. I haven't really set it up or moved any data to it, so I wiped all disks with Gparted by creating a new GPT partition table on every disk (8 3TB drives and a 512GB SSD). With a fresh system, I booted the Proxmox 6.1 ISO and installed to the SSD using EXT4, everything looked good, and rebooted. Now I'm back to the same error I was seeing before the reinstall:
error: attempt to read or write outside of disk 'hd0'
When it goes back to the Grub menu, I press 'c' to go to the grub cli. Running 'ls' just returns "error: unknown filesystem"

I'm starting to think this is a hardware issue but I'm not sure what to look for. Any ideas?

In case it's relevant, I have an HP H220 HBA installed with the latest HP firmware in IT mode, planning on doing ZFS with the 8 disks (actually I had ZFS working nicely in RAIDZ2 before this issue).

 

[Stoiko Ivanov]

hm - things I would try in this situation:
* unplug all other disks apart from the SSD
* check the HBA-util (preboot environment, ...) for some settings
* try wiping the complete ssd (dd if=/dev/zero of=/dev/XXX bs=64M) and installing it fresh again (without the other disks present)
* try installing plain debian - and pve on top of it

I hope this helps!

 

[linucksproxmox]

Thanks for the help. I installed Debian with no issues after rebooting, but when installing the PVE packages it had some kind of failure.

I attempted to install Proxmox from the ISO again but it was showing errors about ata2, and wouldn't fully boot into the installer. I messed around with switching USB ports and rewriting the ISO to the drive, one time it didn't recognize the SSD at all. Long story short, I cleared the NVRAM, reseated all SATA power/data connections and eventually replaced the SATA cable going to the SSD and it all seems to be working great now! I think the SATA cable was the culprit since the errors were not always consistent across reboots.

Thanks again!

 

[Stoiko Ivanov]

Glad you found the issue!
please mark the thread as 'SOLVED'

Thanks!