[Help] URGENT BUG - TOS7 Not Recognising TRAID on D5-300C After Reboot [SOLVED]

[digi999]

System

F4-423 connected to D5-300C (TRAID Pool on F4-423 and Separate TRAID Pool on D5-300C)

Updated to latest TOS7 , Device a corrupted RAID pool on the D5-300C , I let rebuild , accessed folders absolutely fine , no issues but then rebooted and the TRAID pool showed as corrupted again. Already had T/Master admin access my system and say the TRAID pool is unrecoverable (it isn't) this is a TOS7 initialization issue.

mdadm reported:

State : clean
Active Devices : 5
Working Devices : 5
Failed Devices : 0
Spare Devices : 0

All member event counters matched:

Events : 5769
Events : 5769
Events : 5769
Events : 5769
Events : 5769

vg1 was active:

pvs
/dev/md2 vg1

BTRFS was detected:

blkid /dev/vg1/lv2
TYPE="btrfs"

I mounted the filesystem read-only:

mkdir -p /tmp/pool2test
mount -o ro /dev/vg1/lv2 /tmp/pool2test

The filesystem mounted successfully:

51T total
37T used
15T free

BTRFS subvolumes were visible:

Plex-TV
Plex-Movies
Plex-Docs
4K-Plex
Backups

After reboot this showed as corrupted again

[TMzethar]

Hi,
Could you please clarify whether “RAID pool” specifically refers to a storage pool?
Based on the symptoms described, we cannot yet determine the exact cause. After restarting the NAS, please generate a system report and send it to our technical support email address so we can investigate the issue (please include the URL of this forum and your forum username).
Location: TOS Desktop > Support and Help > Report Issues

[digi999]

correct storage pool (the TRAID) 2 is on the D5-300C and Storage Pool 1 is on the F4-423. The F4-423 storage pool is fine , the pool on the D5-300C shows as corrupted , I let it repair , it works fine , then after a reboot it shows as corrupted again. I've already sent a report and a Terramaster engineer has accessed my system remotely and told me the raid isn't recoverable. This isn't correct as Ive twice now recovered the raid it just doesn't stay recovered after a reboot.

[Phil]

correct storage pool (the TRAID) 2 is on the D5-300C and Storage Pool 1 is on the F4-423. The F4-423 storage pool is fine , the pool on the D5-300C shows as corrupted , I let it repair , it works fine , then after a reboot it shows as corrupted again. I've already sent a report and a Terramaster engineer has accessed my system remotely and told me the raid isn't recoverable. This isn't correct as Ive twice now recovered the raid it just doesn't stay recovered after a reboot.

We have received your email, but the log file attached to your email cannot locate and resolve your issue. Please generate a system report through the TOS Desktop > Support and Help > Report Issues interface and then send it to us.

[TMzethar]

[digi999]

I ran:

/etc/tos/scripts/mntdata (as requested via email)

after rebooting into the "corrupted" state.

Before running mntdata:

md2 absent
vg1 absent
Volume3 absent

The script reported:

Detected 5 USB devices, waiting for stable...

No new USB devices for 2 consecutive checks

USB devices stable, continuing...

It then executed:

mdadm -A /dev/md2 \
/dev/sde4 \
/dev/sdf4 \
/dev/sdg4 \
/dev/sdh4 \
/dev/sdi4 --force

which returned:

mdadm: /dev/md2 has been started with 5 drives.

Afterwards:

vg1 appeared

lv2 appeared

Volume3 mounted

Plex-TV

Plex-Movies

Plex-Docs

Backups

4K-Plex

became available.

This suggests the data itself is intact.

Could there be a timing issue during boot where mntdata is executed before the D5-300C USB devices have fully stabilised?
Does TOS call mntdata too early during boot for D5-300C attached arrays?

Would delaying or rerunning mntdata after USB stabilisation be a supported workaround?

[digi999]

can confirm issue has been fixed in latest TOS thankyou

[EriChan]

Thanks for your update feedback.