Description Philipp Hahn 2014-11-18 21:17:50 CET

Created attachment 6417 [details]
partman.log

/var/log/syslog
Nov 18 19:59:20 kernel: [  129.031611] FAT-fs (sda1): utf8 is not a recommended IO charset for FAT filesystems, filesystem will be case sensitive!
Nov 18 20:00:16 init: starting pid 1553, tty '/dev/tty2': '-/bin/sh'
Nov 18 20:02:14 kernel: [  303.585068] EXT4-fs (sda1): couldn't mount as ext3 due to feature incompatibilities
Nov 18 20:02:14 kernel: [  303.585227] EXT4-fs (sda1): mounting ext2 file system using the ext4 subsystem
Nov 18 20:02:14 kernel: [  303.586180] EXT4-fs (sda1): warning: mounting unchecked fs, running e2fsck is recommended
Nov 18 20:02:14 kernel: [  303.586850] EXT4-fs (sda1): mounted filesystem without journal. Opts: (null)
Nov 18 20:03:01 main-menu[1602]: (process:5514): WARNING: could not open /lib/modules/3.16-ucs102-amd64/modules.builtin: No such file or directory
Nov 18 20:03:01 main-menu[1602]: (process:5514): /var/lib/partman/devices/=dev=sda
Nov 18 20:03:01 main-menu[1602]: (process:5514): /var/lib/partman/devices/=dev=sda
Nov 18 20:03:01 main-menu[1602]: (process:5514): mount: mounting /dev/sda1 on /target/boot/efi failed: Invalid argument

partman-efi/commit.d/format_efi has a comment about not formatting existing paritions, because it is too dangerous:
> # Formatting an EFI System Partition that already has a
> # filesystem on it is dangerous
> # (https://bugs.launchpad.net/bugs/769669).  The least bad
> # option seems to be to skip such partitions entirely.
> if [ -f $id/detected_filesystem ]; then
>         continue
> fi
As such the partition probably survived from the previous test run.