fsck EEXIST_str_hash_set, exit code -1

[m-hugo]

default output:

sudo fsck
fsck from util-linux 2.40.2
Running fsck online
bcachefs (nvme0n1p2): check_alloc_info... done
bcachefs (nvme0n1p2): check_lrus... done
bcachefs (nvme0n1p2): check_btree_backpointers... done
bcachefs (nvme0n1p2): check_backpointers_to_extents... done
bcachefs (nvme0n1p2): check_extents_to_backpointers... done
bcachefs (nvme0n1p2): check_alloc_to_lru_refs... done
bcachefs (nvme0n1p2): check_snapshot_trees... done
bcachefs (nvme0n1p2): check_snapshots... done
bcachefs (nvme0n1p2): check_subvols... done
bcachefs (nvme0n1p2): check_subvol_children... done
bcachefs (nvme0n1p2): delete_dead_snapshots... done
bcachefs (nvme0n1p2): check_root... done
bcachefs (nvme0n1p2): check_subvolume_structure... done
bcachefs (nvme0n1p2): check_directory_structure...bcachefs (nvme0n1p2): check_path(): error EEXIST_str_hash_set
bcachefs (nvme0n1p2): bch2_check_directory_structure(): error EEXIST_str_hash_set
bcachefs (nvme0n1p2): bch2_fsck_online_thread_fn(): error EEXIST_str_hash_set
thread 'main' panicked at src/bcachefs.rs:113:79:
called `Result::unwrap()` on an `Err` value: TryFromIntError(())
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace

the panic comes from trying to do u8::try_from(-1).unwrap(), you really shouldn't have any unwrap in your code and handle error reporting properly...

especially since i had to build my own version just to be able to log the error code

Running fsck online
bcachefs (nvme0n1p2): check_alloc_info... done
bcachefs (nvme0n1p2): check_lrus... done
bcachefs (nvme0n1p2): check_btree_backpointers... done
bcachefs (nvme0n1p2): check_backpointers_to_extents... done
bcachefs (nvme0n1p2): check_extents_to_backpointers... done
bcachefs (nvme0n1p2): check_alloc_to_lru_refs... done
bcachefs (nvme0n1p2): check_snapshot_trees... done
bcachefs (nvme0n1p2): check_snapshots... done
bcachefs (nvme0n1p2): check_subvols... done
bcachefs (nvme0n1p2): check_subvol_children... done
bcachefs (nvme0n1p2): delete_dead_snapshots... done
bcachefs (nvme0n1p2): check_root... done
bcachefs (nvme0n1p2): check_subvolume_structure... done
bcachefs (nvme0n1p2): check_directory_structure...bcachefs (nvme0n1p2): check_path(): error EEXIST_str_hash_set
bcachefs (nvme0n1p2): bch2_check_directory_structure(): error EEXIST_str_hash_set
bcachefs (nvme0n1p2): bch2_fsck_online_thread_fn(): error EEXIST_str_hash_set
my own logging of handle_c_command(args, symlink_cmd)'s result: -1

also you should really fix all the unclean shutdown issues "The COW filesystem for Linux that won't eat your data" is still on your website but it still eats a lot of data