[BUG] 6.17.0, mount: /mnt/0: fsconfig() failed: File exists.

[BUG] 6.17.0, mount: /mnt/0: fsconfig() failed: File exists.

[Chris Murphy]

kernel-6.17.0-0.rc7.56.fc43.x86_64
btrfs-progs-6.16.1-1.fc43.x86_64

Downstream bug includes full dmesg attached:
https://bugzilla.redhat.com/show_bug.cgi?id=2399959

File system is already mounted, but I need access to the top-level of the file system.

$ sudo mount /dev/dm-0 /mnt/0
mount: /mnt/0: fsconfig() failed: File exists.
       dmesg(1) may have more information after failed mount system call.

dmesg

[  670.350492] BTRFS: device fsid 958beb6e-eadb-469f-9b33-a06db5a014e7 devid 1 transid 1512 /dev/mapper/luks-32d2f2ad-a545-4aba-b5c7-3fa9135d12bf (252:0) scanned by mount (1324)
[  670.352174] BTRFS info (device dm-0): first mount of filesystem 958beb6e-eadb-469f-9b33-a06db5a014e7
[  670.352302] BTRFS info (device dm-0): using crc32c (crc32c-lib) checksum algorithm
[  670.361200] sysfs: cannot create duplicate filename '/fs/btrfs/958beb6e-eadb-469f-9b33-a06db5a014e7'
[  670.361294] CPU: 3 UID: 0 PID: 1324 Comm: mount Not tainted 6.17.0-0.rc7.56.fc43.x86_64 #1 PREEMPT(full) 
[  670.361300] Hardware name:  /NUC5PPYB, BIOS PYBSWCEL.86A.0081.2022.0823.1419 08/23/2022
[  670.361303] Call Trace:
[  670.361309]  <TASK>
[  670.361314]  dump_stack_lvl+0x5d/0x80
[  670.361324]  sysfs_warn_dup.cold+0x17/0x23
[  670.361331]  sysfs_create_dir_ns+0xcc/0xe0
[  670.361341]  kobject_add_internal+0xbc/0x260
[  670.361347]  kobject_init_and_add+0x8e/0xc0
[  670.361352]  ? btrfs_init_dev_replace+0x176/0x5c0
[  670.361357]  btrfs_sysfs_add_fsid+0x5e/0x120
[  670.361363]  open_ctree+0x71c/0xb30
[  670.361369]  btrfs_get_tree_super.cold+0xb/0xac
[  670.361372]  ? security_fs_context_dup+0x52/0x100
[  670.361378]  btrfs_get_tree_subvol+0xfe/0x1f0
[  670.361383]  vfs_get_tree+0x26/0xd0
[  670.361389]  vfs_cmd_create+0x57/0xd0
[  670.361395]  __do_sys_fsconfig+0x4b6/0x650
[  670.361401]  do_syscall_64+0x7e/0x250
[  670.361406]  ? do_syscall_64+0xb6/0x250
[  670.361410]  ? do_faccessat+0x1d7/0x2d0
[  670.361414]  ? do_syscall_64+0xb6/0x250
[  670.361418]  ? __x64_sys_fsopen+0xb9/0x110
[  670.361422]  ? do_syscall_64+0xb6/0x250
[  670.361426]  ? arch_exit_to_user_mode_prepare.isra.0+0x6a/0xa0
[  670.361431]  ? irqentry_exit_to_user_mode+0x2c/0x1c0
[  670.361436]  entry_SYSCALL_64_after_hwframe+0x76/0x7e
[  670.361441] RIP: 0033:0x7f95ef1cd3ce
[  670.361470] Code: 73 01 c3 48 8b 0d 32 3a 0f 00 f7 d8 64 89 01 48 83 c8 ff c3 0f 1f 84 00 00 00 00 00 f3 0f 1e fa 49 89 ca b8 af 01 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 02 3a 0f 00 f7 d8 64 89 01 48
[  670.361473] RSP: 002b:00007ffee85650d8 EFLAGS: 00000246 ORIG_RAX: 00000000000001af
[  670.361478] RAX: ffffffffffffffda RBX: 00005621c0d2b710 RCX: 00007f95ef1cd3ce
[  670.361480] RDX: 0000000000000000 RSI: 0000000000000006 RDI: 0000000000000003
[  670.361482] RBP: 00007ffee8565220 R08: 0000000000000000 R09: 0000000000000000
[  670.361484] R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
[  670.361486] R13: 00005621c0d2d4e0 R14: 00007f95ef34ea60 R15: 00005621c0d2d578
[  670.361491]  </TASK>
[  670.361494] kobject: kobject_add_internal failed for 958beb6e-eadb-469f-9b33-a06db5a014e7 with -EEXIST, don't try to register things with the same name in the same directory.
[  670.362231] BTRFS error (device dm-0): failed to init sysfs fsid interface: -17
[  670.365833] BTRFS error (device dm-0): open_ctree failed: -17


This doesn't always happen. I'm not sure what the pattern is yet.

--
Chris Murphy

Re: [BUG] 6.17.0, mount: /mnt/0: fsconfig() failed: File exists.

[Chris Murphy]

On Sun, Sep 28, 2025, at 1:53 AM, Chris Murphy wrote:


> Downstream bug includes full dmesg attached:
> https://bugzilla.redhat.com/show_bug.cgi?id=2399959

OK so this is a case of two devices with the same fsid. I simply forgot I had plugged in the original device planning to wipe it - its replacement is a new fs using metadata_uuid, with the original fsid.

But the two identical fsid's is the source of all this confusion, both user and I think the kernel is also confused.

I've added more details to the downstream bug report, including strace. 



-- 
Chris Murphy