Here's a screenshot of what happens at boot. I think the crux of the matter is the clock skew and the failed fsck.
On my RPi, I have tried:
- Setting the clock with date and saving with
Code: Select all
/lib/rc/sbin/swclock --save - Remounting with
Code: Select all
mount -o remount,rw / - Rebooting without fsck via
Code: Select all
shutdown -f - Running fsck manually
- Setting swclock to the sysinit runlevel (currently it is on the boot runlevel, and it may or may not be run before fsck)
- Combinations of the above
I have also tried taking the card out, plugging it into another machine, and running fsck from there. That gives me an entirely different error:
Code: Select all
zenbookhost chymera # fsck /dev/sdb1
fsck from util-linux 2.27.1
fsck.fat 3.0.28 (2015-05-16)
0x41: Dirty bit is set. Fs was not properly unmounted and some data may be corrupt.
1) Remove dirty bit
2) No action
? 1
Perform changes ? (y/n) y
/dev/sdb1: 18 files, 28971/142266 clusters
zenbookhost chymera # fsck /dev/sdb2
fsck from util-linux 2.27.1
e2fsck 1.42.13 (17-May-2015)
Angstrom: recovering journal
Superblock needs_recovery flag is clear, but journal has data.
Run journal anyway<y>? yes
fsck.ext3: unable to set superblock flags on Angstrom
Angstrom: ********** WARNING: Filesystem still has errors **********

