Hi, I just updated to the latest snapshot and now my PC boots to emergency mode. I logged in and ran journalctl -xb and in the log there was CPA detected w^x violation followed by hex numbers I assume are memory address ranges.
This happens when I select either of the last two kernals and with read only snapshots. Basically my PC is unbootable.
I am typing this on my roommates win10 box.
I asked gemini and it says: The kernel’s CPA system caught an attempt to assign a memory region permissions that allow writing and executing simultaneously.
It gave three possible causes two relating to VMs and the one that may be the culprit is proprietary gpu drivers.
I do not use VMs.
I guess my question is - how can I boot? All the options I tried - previous kernal and read only snapshots fail. I am at a loss.
I was able to get a read only snapshot to boot. It is 20260813.
My pc sometimes hangs with an error popup saying nvidia-smi cannot connect to the drivers. So it seems nvidia is the problem. My problem is I cannot boot to a writeable snapshot.
Anyone with an idea?
Thanks I did that and it seems to work. Except the nvidia drivers fail to load. System info says my Intel 630 is the only gpu. I have two nvidia cards, an RTX3060 and a Titan RTX. They are now invisible. I guess I will have to wait for a driver update. The current ones put me straight into emergency mode.
I didn’t know sudo snapper rollback without an image number would make things writable. Thank you.
Ironically, about 1.1/2 hours ago, I had to do a “snapper rollback” on a Leap 16.1 install I did earlier. I goofed up a system configuration … so I rebooted and chose the previous Snapshot
When you’re in a Snapshot and don’t supply a number, it assumes you want to rollback to the one you’re in
Anyway, at least you’re in a running system and can take your time to determine next steps.