Description of problem: Currently, anaconda permits a /lost+fount mount point. I believe anaconda should not accept any storage configuration in which any being created mount point would mount itself over any 'lost+found' directory. I think 'lost+found' should be a reserved word or something like that. A user might still setup such configuration manually post-install, but that is another story and has nothing to do with anaconda. Examples: No /lost+found, /boot/lost+found, etc. as mount points. Version-Release number of selected component (if applicable): F19b TC4 (19.25-1) How reproducible: always. Steps to Reproduce: 0. Reach the Main Hub, set English because Portuguese is still selected by default. The correct one would be Spanish. 1. I selected 'Minimal' Base Environment 2. Enter Storage: Installation Destination and do custom partitioning (1 disk guest) using 'standard partition' partition scheme. 3. Add a swap (768mb) 4. Add a /boot (512mb) 5. Add a /lost+found (120mb) YES! it is not a joke. 6. Add a / with the rest of the disk. 7. Return to the Main Hub and perform the installation. Actual results: A fake /lost+found, with different permissions. There might be real files under the real one. Expected results: Anaconda refusal to create such storage configuration. Additional info: * LOST+FOUND on a F17 system: # ls -ld /lost+found/ drwx------. 2 root root 16384 May 22 2012 /lost+found/ # ls -ldZ /lost+found/ drwx------. root root system_u:object_r:lost_found_t:s0 /lost+found/ *A 'FAKE' LOST+FOUND on a F19b TC4 system: # ls -ld /lost+found/ drwxr-xr-x. 3 root root 1024 May 10 19:31 /lost+found/ # ls -ldZ /lost+found/ drwxr-xr-x. root root system_u:object_r:lost_found_t:s0 /lost+found/ # df -h Filesystem Size Used Avail Use% Mounted on /dev/sda2 8.2G 638M 7.1G 9% / devtmpfs 492M 0 492M 0% /dev tmpfs 498M 0 498M 0% /dev/shm tmpfs 498M 368K 498M 1% /run tmpfs 498M 0 498M 0% /sys/fs/cgroup tmpfs 498M 4.0K 498M 1% /tmp /dev/sda1 488M 61M 403M 14% /boot /dev/sda5 112M 1.6M 104M 2% /lost+found * From FSH v2.3: Section 1.10. /lost+found > [...] Each partition has its own lost+found directory. > If you find files in there, try to move them back to their original location. [...] So there could be files that only root should see on a lost+found entry after a crash, but if there is another filesystem mounted over a lost+found, those files would be not visible. On the whole section, it is described as a directory that is used/created by fsck, it is never ever called a mount point.
There are too many possible permutations of this. It doesn't cause a traceback, so I'd call it user error.