Teehehehe.

Anyone here willing to help me do a little troubleshooting?

Situation: boot process migrated from HDD to SSD, HDD now physically removed from system. Root file system migrated from plain ext3 to ext4 on RAID 1 MD (currently with only a single device out of two, but that's unrelated). GRUB command list points the Linux kernel at "root=/dev/md0 ro" with a search --set=root preceding naming the UUID of the root ext4 file system.

Updated fstab to name /dev/md0 for the root fs. On boot, the root file system check fails with "/dev/md0 mounted" and asks for root password for maintenance; if I hit Ctrl+D to abort that, then the boot completes successfully. Why is e2fsck complaining that md0 is mounted? (Well, it is mounted, but no more than any other root fs on boot...)

@MichaelKjörling - I can't help now but might later tonight. Setting up xmas tree now w/ kids, should be around 8-9PM EST ton. though.

@slm That's in like 5-6 hours, correct? That would make it around 3 in the morning for me :)

ah

yeah, sorry

i'll try researching it anyway

let me know if you figure it out

@slm It looks like it's fsck -C -R -A -a being run from somewhere (almost certainly Debian's checkroot init script) that's failing. What I don't understand is why this has cropped up with the move from root fs on a plain partition to root fs on /dev/md0. Because that's about the only thing I can think of that's changed that should be relevant to fsck.

I did find something curious though. At the point where e2fsck reports failure, / is actually mounted rw, even though the kernel reports ro in its command line.

@slm I figured it out. There was a stray entry for the same file system in a different place, from when I was migrating data. When I commented out that one, the boot proceeded without a hitch. With thanks to @Hennes.