Notifications
Clear all
Functionality Issues
11
Posts
2
Users
0
Reactions
1,021
Views
Topic starter
I plugged a Mercury Elite Pro Dual U.2 enclosure (All volumes are M.2 SSDs) into a Mac mini with SoftRAID installed to move large media files to SoftRAID volumes on the mini. SoftRAID popped up a warning that two of the M.2 SSDs are predicted to fail.
The displayed status in SoftRAID for both of these SSDs is "passed test". I've have DriveDX running on the machine the Mercure Elite Pro is normally connected to and it show NO ISSUES with them. Can you explain what SoftRAID is detecting to flag these drives?
Thanks,
GB
Posted : 03/08/2024 9:40 am
Since this is USB, (SoftRAID does not pull SMART data from USB) my guess is the drives had had IO errors. Attach a SoftRAID support file and I can look.
Posted : 03/08/2024 11:22 am
Topic starter
@softraid-support This is not USB this is a Thunderbolt drive enclosure made my OWC!
Posted : 03/08/2024 11:28 am
Posted : 04/08/2024 12:08 am
Topic starter
Posted : 04/08/2024 8:50 am
I need some data for you. There is a bug in SoftRAID logging for the particular failure mode your drive has. If you can get me some data quick enough, we can fit it into the next SoftRAID update.
If you look at the SoftRAID disk tiles, each disk has a "disk identifier". the predicted to fail disk in the support file was disk15 for example. These numbers change frequently.
Please look at the disk tile for your failing NVMe drive. It may still be disk15, or it may be another number. Below is a terminal command. Replace 15 with the current number of your disk.
Open the terminal.app.
Paste this command in, and hit enter:
softraidtool disk disk15 getsmartdata
copy/paste the text and post it here. thanks!
Posted : 05/08/2024 1:07 pm
Topic starter
I've included the info for both M.2 SSDs that SoftRAID is listing failure predicted.
SMART data for disk15:
disk15, SN: 112110140130069, PCI bus 0, id 0, lun 0 (Thunderbolt)
Total Bytes: 3.73 TB (4,096,805,658,624)
This disk passed the SMART test.
SMART/Health Information
Critical Warning: 0x00
Temperature: 36.85 celsius
Available Spare: 0x64
Available Spare Threshold: 0x20
Percentage Used: 0%
Bytes Read: 4443605504000 bytes
Bytes Written: 2255619584000 bytes
Host Read Commands: 246505587
Host Write Commands: 20237509
Controller Busy Time: 0
Power Cycles: 774
Hours of Operation: 18165
Unsafe Shutdowns: 71
Media and Data Integrity Errors: 357564
Error Information Log Entries: 0
SMART data for disk10:
disk10, SN: 112207190180015, PCI bus 0, id 0, lun 0 (Thunderbolt)
Total Bytes: 3.73 TB (4,096,805,658,624)
This disk passed the SMART test.
SMART/Health Information
Critical Warning: 0x00
Temperature: 36.85 celsius
Available Spare: 0x64
Available Spare Threshold: 0x20
Percentage Used: 0%
Bytes Read: 6470385664000 bytes
Bytes Written: 5961451520000 bytes
Host Read Commands: 324669672
Host Write Commands: 52394525
Controller Busy Time: 0
Power Cycles: 432
Hours of Operation: 12846
Unsafe Shutdowns: 53
Media and Data Integrity Errors: 272403
Error Information Log Entries: 0
Posted : 05/08/2024 6:36 pm
We will try to add logging in the SoftRAID Log for the specific reason these drives are flagged as predicted to fail, but they are showing issues. Note the Media and Integrity errors. This means each has had an average of 300,000 events where the media wrote the wrong information to disk. (ECC corrects this, yes) We flag this as a state where the drive is likely to fail soon. Its statistical, not deterministic.
This post was modified 2 years ago 2 times by SoftRAID Support
Posted : 06/08/2024 12:26 pm
Topic starter
Thanks for explain what SoftRAID is flagging the drives for!
Posted : 06/08/2024 6:39 pm
Please test the current SoftRAID beta. In the SoftRAID log, it should show the details for the predicted failure;
Its at: softraid.com/sr_beta
thanks!
Posted : 06/08/2024 6:42 pm
