Chris <bsd-li...@bsdforge.com> writes: > Warner Losh <i...@bsdimp.com> writes: > > Maybe this is a custom kernel without the label code. > It's an upgrade. It's 15 from ~11 mos ago. Yes a custom kernel. But > I'm using the same kernconf as before. Was label removed from generic > in the last 11 mos.?
Please share: - The output of the `what` command on your old (working) kernel - The output of the `what` command on your new (failing) kernel - The kernel configuration file you used to build the new kernel - The contents of `/var/run/dmesg.boot` after booting the old kernel > > What does '?' at mountroot> prompt say? > nadda, nothing. That's hard to believe. Are there any error messages relating to nvme prior to the mountroot promopt? (if necessary, use Scroll Lock followed by Page Up / Page Down to scroll back on the console). > > Does one of those work if you add 'ufs:/dev/' to the > > front of each of the geoms listed? Most likely ada0p2 or nda0p2. > Indeed. It's nda0p2. But ufs:/dev/nda0p2 returns the same error 19. > I had to move kernel to kernel.new && kernel.old to kernel to get > here to write this response. You could have just selected kernel.old from the loader menu (press 'k' to cycle between available kernels) or prompt (`boot kernel.old`)... I would however advise you to copy kernel.old to kernel.works, as it will be permanently deleted the next time you install a new kernel (unless you use `make reinstallkernel` instead of `make installkernel`). DES -- Dag-Erling Smørgrav - d...@freebsd.org
