Package: e2fsprogs Version: 1.43.4-2 Severity: critical e2fsck potentially moves blocks around in sparse files when running with -E bmap2extent, in particular when the blocks are physically adjacent on the underlying block device, but have a hole in between in the file.
This script reproduces the problem on my system (run as root in an empty directory!): #!/bin/bash -ex rm -f image mkdir -p m umount m || true dd if=/dev/zero bs=1M seek=99 count=1 of=image mkfs.ext3 -E nodiscard -b 4096 image mount -o loop image m echo -n a > m/x echo -n b | dd of=m/x bs=1 seek=8k ls -i m/x ino=$(stat -c%i m/x) sha1sum m/x umount m debugfs image -R "stat <$ino>" | cat tune2fs -O extent image e2fsck -fE bmap2extent image || true debugfs image -R "stat <$ino>" | cat mount -o loop image m sha1sum m/x umount m rm image rmdir m The e2fsck invocation turns this part of the debugfs output: BLOCKS: (0):24576, (2):24577 into this: EXTENTS: (0-1):24576-24577 With version 1.44.5-1~bpo9+1, the test above does not produce corruption on my system, but I have not investigated whether that is just coincidence or because the bug has been fixed. As this silently corrupts files, I would think it should get some fix in stretch, and be it by disabling the feature.
