Buying two used Seagate Barracuda ST4000DM004 4TB SATA drives tomorrow - plan of attack

[Dibrom]

I'm buying two used drives as per subject tomorrow of completely unknown history and health. Risky, but the price is right for the risk.

I have the latest versions of R-Studio Technician and of course I also have SpinRite 6.1R4. Before attacking them with SpinRite to see how bad they are, I thought I'd give them a going over with R-Studio and see if I can ascertain what the seller might have been using them for – or at least how well they were wiped before being put up for sale.

I've never had the pleasure of needing to use R-Studio before to recover lost software, so I thought this would represent a good opportunity to learn some things if in future I do find myself needing to recover lost software for reals.

1. Has anyone in here any experience with using R-Studio they wouldn't mind sharing with me for the described purpose I'm about to embark on?
2. Aside from SpinRite (with which I'm reasonably familiar), are there any other softwares or processes others would recommend being used on 'new' drives of completely unknown history, health and reliability, before putting them into service in a serious way with real software storage?

Obviously being 4TB each, they're not insignificantly small to not matter and I do have the time to spare making really sure they are 100% reliable before putting them into serious service. Any advice is greatly appreciated.

 

[peterblaise]

For new-to-us drives the fastest and most
thorough surface integrity confirmation might
be:

SPINRITE NORAMTEST DYNASTAT 0 LEVEL 5

Let us know what yo do and what are the results.

Thanks.

 

[PHolder]

There seems little reason to disable the RAMTEST. Yes, in theory it's a waste of time to check it every time if you think it's good, but on the other hand, it can go bad at ANY time and you really shouldn't be using a PC that isn't reliable to check the reliability of your disks.

 

[Dibrom]

Let us know what yo do and what are the results.

I will do that. I pick them up today and will temp install on a spare computer first just to mitigate any poential malware/virus issues and also to enable install of R-Studio Technician without disrupting the main computer. I'm first going to try and analyse them as-is to see if anything can be recovered just more as a learning/practice exercise in using R-Studio rather than any suspicion of anything nefarious. I rather suspect that given they're a matched pair of drives, they were one of… RAID setup in desktop; or CCTV drives; or part of a crypto mining server setup.

Only the first of those three options may result in recoverable useful data. CCTV or Chia Farm crypto mine would only reveal raw split data files which would be meaningless. Given they're not specific CCTV constant use multiwrite type drives constructions, it's unlikely they're from a professional purchased CCTV system. Perhaps a DIY home made one, but not likely a professional retail one. The Chia Farm possibility would obviously be very DIY of course. I think the RAID use is possibly the most likely, but I should find out fairly quickly after they're in the hand.

 

[peterblaise]

The SpinRite 6.1 8961 Release 4 System RAM Test at least causes
ghosts of unpartitioned drives during enumeration. If anyone finds
other misbehaviors after RAM test, let us know. @Steve has a list of
resolutions queued up for Release 5 eventually. Simply restarting
SpinRite 6.1 with the NORAMTEST command line option bypasses
the RAM test and produces no ghost drive. When I'm doing
development beta testing, I'm looking for SpinRite misbehavior.
When I'm doing data recovery and drive maintenance, I don't want
SpinRite to misbehave in any way, so I copiously employ command
line options to 'discipline' SpinRite as much a known and possible.
Hence suggesting to others to also eliminate any task that does not
directly support their goals, here, data recovery or drive
maintenance.

 

[PHolder]

The SpinRite 6.1 8961 Release 4 System RAM Test at least causes
ghosts of unpartitioned drives during enumeration.

Interesting. If it's repeatable then that clearly sounds like a bug where the RAM test fills memory that needed to remain clear. (To be quite honest, I haven't run SpinRite since late beta which was before the RAM test was added. I don't have the setup to leave SpinRite configured to scan HDDs and it got tiresome constantly tearing apart and re-configuring for every beta release.)

 

[Dibrom]

First drive DOM: Aug 2021, spent all yesterday being scanned by R-Studio Technician. It passed the sector read test no problem at all. Not even a single re-read required. The deleted files on the drive I could see indicated it was used as just a standard Windows PC system disk. There is a small Microsoft system partition and a much larger storage partition that contained nothing of any use or interest. I haven't formatted or otherwise re-partitioned it for my own use yet. Given the R-Studio analysis result, it's now undergoing an L3 SpinRite scan which will take all of today to complete. After that, I will delete the existing partitions and GPT reformat to maximum storage capacity, which again I expect will take most of a day to complete.

The second drive is identical with a DOM: Nov 2021

Update: Still working on first drive DOM Aug 2021. Going slower than anticipated. Currently at ~93% complete and been going for 24hrs 13min. It has recently slowed right down, not to DynaStat territory speed, but much slower than normal. I've now got some red ECC corrected blocks and one red command timeout block.

I've also got an error message that reads: The drive has refused to rewrite original data. Attempting rewrite to identify unwritable sectors(s). 7,319,126,016 (93.6663%). There are no RBU coloured sectors in the Graphic Status Display yet. Everything looks perfectly normal so far otherwise.

I ran SpinRite on this disk as it came to me. I haven't deleted any partitions, deleted any files, tried to format it or anything else. Should I have wiped it first before running a SpinRite L3?

Update 2: Drive one finally finished. 1 x command timeout – Longest drive reset recovery 2 seconds
Zero RBU sectors
ECC corrected = 77/94

Apart from that everything else looks completely normal

 

[Dibrom]

The SpinRite 6.1 8961 Release 4

How do you tell what exact release version your SpinRite is? Where is the 8961 bit displayed? Mine just says 6.1 Release 4. There is no more specific version information I can find.

 

[PHolder]

How do you tell what exact release version your SpinRite is?

Check the file properties of the .EXE file? (Under Windows.)