failed drives after secure erase/eject

[usersky]

Hi! F5-221 here, TOS 5.1.79.

I tried to secure erase two drives WDC WD140EDGZ. After secure erase started and i was informed that it would take 1402 minutes to complete I tried to stop it but there is no way to gracefully interrupt a secure erase. I tried to stop the NAS so I eject the drives but it didn't stop. At this point i just ejected the two drives manually (they are supposed to be hot swappable aren't they?). After that both drives (previously perfectly fine) were unusable. They cannot be used, included in a storage pool, erased, nor can they be initialized in a windows machine, that shows them as unalocated but cannot initialize them as MBR or GTP. They cannot be secure erased again, basically they are broken. System sees them correctly with the model/firmware version etc but cannot do anything about them.

Any sugestion would be welcome, at this point both drives are totally borked. Thanks!

[TMzethar]

Hot swapping is a hardware level feature where your hard drive is already involved in the operation of the system and various programs. Removing it will affect the normal operation of the system, programs, and even storage. If your Windows computer also supports hot swapping, you definitely wouldn't want to directly remove the hard drive that is being used by the system or programs.
Are there any other hard drives in TNAS after you unplug them?
If there are other hard drives in TNAS, you can try restarting TNAS using the hard drives inside, and then put the two disks that need to be erased into TNAS.

[usersky]

There are other drives in the NAS. I tried what you sugested but the two drives seem unusable, they cannot be erased, initialized, nor in the NAS, nor externally in other systems. I don't care at all for their content of course, they were to be erased but they seem to be defective now and this is not something to be expected by ejecting them. Is there a way to... low leve initialize them or anything? Thanks!

2023-11-13_133125.png

They look normal here:

2023-11-13_134149.png

SMART Info (nothing special, seem normal):

2023-11-13_134403.png

[Gremlin]

If you have not already tried, you might consider one or other of the free disk/partition manager software available for Windows. (I use EaseUS, but that is not a recommendation ) These can sometimes be more effective than Windows native solutions.

[usersky]

Thanks for the tip, no luck unfortunatelly. They seem defective, WD Data Lifeguard cannot perform long SMART test. All this from removing them from NAS while secure erasing. Makes me completely reevaluate if this is something I should entrust my data into.

[Gremlin]

Such a failure is not unexpected. In "secure erase" the disk is being continually written ('0', '1', or '0+1') depending on your (or the default) choice. If removed in the process of witing, the heads/disk are in motion and probably experience head crash as they would unlikely come to a dead stop.

I think the issue, as you described, is the system not responding to a 'cancel' request.

[usersky]

May I be in this situation? https://poyu.co.uk/2018/04/13/unlock-hd ... ure-erase/
Because Windows says Cyclic Redundancy Check error trying to initialize the disks. I'll dig into this.
Continuos writing while loosing power should not phisically ruin a drive, this is not a thing since early 90s AFAIK.

[Gremlin]

I pre-date the 90's by a good margin. Nothing surprises me these days.

Personally, I would not use a nas to perform these functions. Much rather trust my windows device and (fully licenced) disk manager software that allows for support in times of trouble. (I fitted a hot-swap drive caddy to make life simpler and it functions for back-ups as well.)

[usersky]

While I totally agree now, it's a bit late I contemplate 2 new WD 14TB door stoppers so now I hope the drives are locked and unlockable, will investigate shortly.

Indeed being unableto cancel the operation is deeply inconvenient, especially that you find that it will take ages just AFTER the operation starts. A sequence of unfortunate events. If power fails during 1400 minutes operations (safe erase) means a ruined drive, this really should be made exceedingly clear before the operation starts.

[usersky]

I can confirm at this point both drives became secority locked during the secure erase and now they are locked:

2023-11-13_191740.png

When I try to unlock them from within linux mint (internally connected drive) with this command:
sudo hdparm --user-master m --security-unlock WDCWDCWDCWDCWDCWDCWDCWDCWDCWDCW /dev/sda

I get an error like
SG_IO: bad/missing sense data, sb[]: 70 ....

Nothing that i found and tried helped so far.

This is frustrating beyond words, such a dangerous command as safe erase should clearly be interruptible, preceded by how long it will take BEFORE the confirmation and clearly stated that a powerd down during this will render the drive useless.