To provide some back story I was working my way through the ubuntu pci passthrough tutorial (Ubuntu 17.04 -- VFIO PCIe Passthrough & Kernel Update (4.14-rc1))
when I was having issues with it showing IOMMU groups. To fix this I updated to the latest Kernel which I think was Kernel 4.18.x After this I noticed when I left a bootable USB in my USB port and rebooted it brought up an error like the below. Apologies for the bad phone photos btw.
Anyway this caused issues later on when I read that the proprietary Nvidia drivers I had previously installed could be causing issues with passthrough I decided to reinstall Ubuntu. This was not possible because of the error.
My searching online indicates this is to do with video drivers as I can determine by the fact it mentions nouveau. Eventually I managed to boot into a Ubuntu installer using the nomodeset parameter but then I still got an error installing GRUB or removing all the partitions when getting into a Ubuntu live boot.
Long story short I want to ask how this can be fixed before I start the same process now I have installed Ubuntu to my SSD instead of my HDD. Note the only way around this was to install it to my SSD instead. Before I continue trying the pci passthrough stuff again I want to know how this can be fixed or how to avoid it so I do not cause the same problem on my SSD.
I am a bit of a Linux Desktop noob by most Linux nerds standards so even basic suggestions are welcomed.
EDIT: And I should say I have done lots of research already but nothing that came up fixed it. I am thinking a full drive wipe is the way to go. How best to do this on Linux? Also I had issues booting into Dban boot and nuke via a USB bootable disk.
I think that your GPU driver/module issue is a coincidental effect, rather than the cause of your problem. Instead, I think it is more likely that something blew up on your disk.
Seriously, only in some parallel universe, is a cratered kernel module going to interfere with the partitioning of a disk.
I would completely wipe the disk, and add new partitions and a new partition table. I also wouldn’t put any data onto it that I wasn’t willing to loose and I would get the toolkit installed that allows for the monitoring of the S.M.A.R.T. data. You don’t want a pass/fail report, you want to be able to monitor the raw data and see what the error counts look like. More importantly, if they are increasing. Then, put the disk back into service and watch what happens.
Yeah I reckon that’s what I will do. Although I put my computer hardware set back to my normal Windows set up for now and will give it another go when I get the free time. From what I read I can boot into a live Ubuntu session and install/run a wipe tool. Any experience with that or any particular tools you prefer?
