USB drives disappearing

thomaslr · Dec 12th 2024

I'm pretty new to OMV, less so to linux but it's been a long time that I've had drive issues so this has me stumped.

I am finding that usb attached drives occasionally go missing. They sometimes come back, even as a different device (was /dev/sdb, reappears as /dev/sdd) but don't remount back to where they are designated under /etc/fstab.

A reboot will always refind them - but that's not really a solution.

I've looked high and low onto how I can rescan the USB to reset , reattach etc... but nothing seems to be working.

macom · Dec 12th 2024

Quote from thomaslr

I've looked high and low onto how I can rescan the USB to reset , reattach etc... but nothing seems to be working.

Try sudo mount -a from CLI.

If that works you might try to add x-systemd.automount as mount option for that drive.

thomaslr · Dec 13th 2024

sudo mount -a

mount: /srv/dev-disk-by-uuid-011eedde-d7e5-4c6f-bcf8-11a987158adb: mount(2) system call failed: File exists

looks like the mount point still exists and stops mount -a from successfully remounting

if I run umount /srv/dev-....... then run mount -a it works.... which is something but still a suboptimal

looking through the last dmesg

it was operating as sdc but looks like it ran into issues & crapped out and left the mount point dangling as a result and stopping the mount -a working

360385.609294] BTRFS error (device sdc: state EA): bdev /dev/sdc errs: wr 0, rd 3817, flush 0, corrupt 0, gen 0

[360385.609408] BTRFS error (device sdc: state EA): bdev /dev/sdc errs: wr 0, rd 3818, flush 0, corrupt 0, gen 0

[360385.614866] BTRFS error (device sdc: state EA): bdev /dev/sdc errs: wr 0, rd 3819, flush 0, corrupt 0, gen 0

[360385.614996] BTRFS error (device sdc: state EA): bdev /dev/sdc errs: wr 0, rd 3820, flush 0, corrupt 0, gen 0

[360385.620541] BTRFS error (device sdc: state EA): bdev /dev/sdc errs: wr 0, rd 3821, flush 0, corrupt 0, gen 0

[360385.620616] BTRFS error (device sdc: state EA): bdev /dev/sdc errs: wr 0, rd 3822, flush 0, corrupt 0, gen 0

[360385.634639] BTRFS error (device sdc: state EA): bdev /dev/sdc errs: wr 0, rd 3823, flush 0, corrupt 0, gen 0

[360385.634764] BTRFS error (device sdc: state EA): bdev /dev/sdc errs: wr 0, rd 3824, flush 0, corrupt 0, gen 0

[360385.639537] BTRFS error (device sdc: state EA): bdev /dev/sdc errs: wr 0, rd 3825, flush 0, corrupt 0, gen 0

[360385.639672] BTRFS error (device sdc: state EA): bdev /dev/sdc errs: wr 0, rd 3826, flush 0, corrupt 0, gen 0

[360638.619696] btrfs_dev_stat_inc_and_print: 14 callbacks suppressed

[360638.619700] BTRFS error (device sdc: state EA): bdev /dev/sdc errs: wr 0, rd 3841, flush 0, corrupt 0, gen 0

[360638.619718] BTRFS error (device sdc: state EA): bdev /dev/sdc errs: wr 0, rd 3842, flush 0, corrupt 0, gen 0

[360639.032296] BTRFS error (device sdc: state EA): bdev /dev/sdc errs: wr 0, rd 3843, flush 0, corrupt 0, gen 0

[360639.032396] BTRFS error (device sdc: state EA): bdev /dev/sdc errs: wr 0, rd 3844, flush 0, corrupt 0, gen 0

[360639.032902] BTRFS error (device sdc: state EA): bdev /dev/sdc errs: wr 0, rd 3845, flush 0, corrupt 0, gen 0

[360639.032969] BTRFS error (device sdc: state EA): bdev /dev/sdc errs: wr 0, rd 3846, flush 0, corrupt 0, gen 0

[360639.033104] BTRFS error (device sdc: state EA): bdev /dev/sdc errs: wr 0, rd 3847, flush 0, corrupt 0, gen 0

[360639.033529] BTRFS error (device sdc: state EA): bdev /dev/sdc errs: wr 0, rd 3848, flush 0, corrupt 0, gen 0

[360639.033664] BTRFS error (device sdc: state EA): bdev /dev/sdc errs: wr 0, rd 3849, flush 0, corrupt 0, gen 0

[360639.033731] BTRFS error (device sdc: state EA): bdev /dev/sdc errs: wr 0, rd 3850, flush 0, corrupt 0, gen 0

[361643.347049] usb 2-2: reset SuperSpeed USB device number 43 using xhci_hcd

[361643.368349] program smartctl is using a deprecated SCSI ioctl, please convert it to SG_IO

[361719.139093] usb 2-2: reset SuperSpeed USB device number 43 using xhci_hcd

[361719.160412] program smartctl is using a deprecated SCSI ioctl, please convert it to SG_IO

[361720.904474] program smartctl is using a deprecated SCSI ioctl, please convert it to SG_IO

[361720.904530] program smartctl is using a deprecated SCSI ioctl, please convert it to SG_IO

[361721.279581] program smartctl is using a deprecated SCSI ioctl, please convert it to SG_IO

[361721.716816] program smartctl is using a deprecated SCSI ioctl, please convert it to SG_IO

[361722.187738] program smartctl is using a deprecated SCSI ioctl, please convert it to SG_IO

[361722.215225] program smartctl is using a deprecated SCSI ioctl, please convert it to SG_IO

[361722.453485] program smartctl is using a deprecated SCSI ioctl, please convert it to SG_IO

[361722.790508] program smartctl is using a deprecated SCSI ioctl, please convert it to SG_IO

[361896.428313] BTRFS warning: duplicate device /dev/sdd devid 1 generation 17447 scanned by mount (2871305)

macom · Dec 13th 2024

Quote from thomaslr

[360385.639672] BTRFS error (device sdc: state EA): bdev /dev/sdc errs: wr 0, rd 3826, flush 0, corrupt 0, gen 0
[360638.619696] btrfs_dev_stat_inc_and_print: 14 callbacks suppressed

Suggest to google for that error.

Years ago there was a recommendation on the btrfs wiki not to use btrfs on certain USB connected devices. Might be worth checking if it still exists.

Krisbee · Dec 13th 2024

thomaslr Intermittant connections on USB drives in midst IO can cause problems on any filesystem. With BTRFS, errors can force into read-only mode, or simply corrupt it.

This an example of power loss, and then regained, on a USB drive during mid file read:

Code

root@omv7:~# findmnt --real
TARGET                                                       SOURCE      FSTYPE OPTIONS
/                                                            /dev/vda1   ext4   rw,relatime,errors=remount-ro
├─/srv/Opool                                                 Opool       zfs    rw,noatime,xattr,posixacl
│ └─/srv/Opool/odata                                         Opool/odata zfs    rw,noatime,xattr,posixacl
└─/srv/dev-disk-by-uuid-d50c1327-ccb5-45e7-901b-dbdc24fc61fd /dev/sdc    btrfs  ro,relatime,space_cache=v2,subvolid=5,subvol=/
root@omv7:~# mount | grep btrfs
/dev/sdc on /srv/dev-disk-by-uuid-d50c1327-ccb5-45e7-901b-dbdc24fc61fd type btrfs (ro,relatime,space_cache=v2,subvolid=5,subvol=/)
root@omv7:~# journalctl -b -g btrfs
Dec 13 16:51:13 omv7 kernel: Btrfs loaded, crc32c=crc32c-intel, zoned=yes, fsverity=yes
Dec 13 16:52:02 omv7 kernel: BTRFS: device fsid d50c1327-ccb5-45e7-901b-dbdc24fc61fd devid 1 transid 18 /dev/sdc scanned by (udev-worker) (1017)
Dec 13 16:52:02 omv7 kernel: BTRFS info (device sdc): first mount of filesystem d50c1327-ccb5-45e7-901b-dbdc24fc61fd
Dec 13 16:52:02 omv7 kernel: BTRFS info (device sdc): using crc32c (crc32c-intel) checksum algorithm
Dec 13 16:52:02 omv7 kernel: BTRFS info (device sdc): using free space tree
Dec 13 17:07:53 omv7 kernel: BTRFS error (device sdc): bdev /dev/sdc errs: wr 1, rd 0, flush 0, corrupt 0, gen 0
Dec 13 17:07:53 omv7 kernel: BTRFS error (device sdc): bdev /dev/sdc errs: wr 2, rd 0, flush 0, corrupt 0, gen 0
Dec 13 17:07:53 omv7 kernel: BTRFS: error (device sdc) in btrfs_commit_transaction:2496: errno=-5 IO failure (Error while writing out transaction)
Dec 13 17:07:53 omv7 kernel: BTRFS info (device sdc: state E): forced readonly
Dec 13 17:07:53 omv7 kernel: BTRFS warning (device sdc: state E): Skipping commit of aborted transaction.
Dec 13 17:07:53 omv7 kernel: BTRFS: error (device sdc: state EA) in cleanup_transaction:1994: errno=-5 IO failure
root@omv7:~# btrfs dev stats /srv/dev-disk-by-uuid-d50c1327-ccb5-45e7-901b-dbdc24fc61fd
[/dev/sdc].write_io_errs    2
[/dev/sdc].read_io_errs     0
[/dev/sdc].flush_io_errs    0
[/dev/sdc].corruption_errs  0
[/dev/sdc].generation_errs  0
root@omv7:~#

Display More

While there's only 2 write errors in this case, the scrub cannot run as the filesystem is mounted read-only.

In any case, in this example the fllesystem is a "SINGLE" so only metadata could be repaired.

Your output from "dmesg" shows potential file corruption. You need to see if you can still mount it at all and what is shown by "btrfs dev stats".

USB drives disappearing

crashtest Dec 12th 2024

Participate now!

Similar Threads

cant grow/recover raid 1 array (mdadm: /dev/nvme2n1 not large enough to join array)