ST9500423AS: Test Drive from eBay

[peterblaise]

". . . Because that SMART data doesn't exist yet . . . I prefer to write my own random data . . ."

I believe you as saying that a S.M.A.R.T. report has not been collected yet. at least not one that reports any sector reallocations.

As far as we can tell from all of your discovered and shared S.M.A.R.T. reports, no sectors have been reallocated, and the 7 sectors pending may stay that way forever, and S.M.A.R.T. does not say where they are.

I see you are using the excellent, freely-available software from Miray https://www.miray-software.com/download/free/index.html ( note: their HD Shredder is 51 MB, SpinRite is 80 KB - wow @Steve Gibson, kudos ! ).

If we prefer to 'erase' a drive because prior data should never be seen again, that is a separate issue, and random data or 0s or FF are reported to be thorough enough to prevent rediscovery of the prior data, though keeping the drive and or physically dismantling it and or physically shredding the platters it also works.

---

The opening post seems to be a comparison between SpnRite and Rapid Spar https://rapidspar.com/ with a discussion of S.M.A.R.T., and now comparing to Miray HD Shredder.

Why?

SpinRite is all about data recovery and drive maintenance.

One way or another, this drive's data appears to be recovered, as well as the drive itself appears 'maintained', though we do not yet see 'use' implementation, such as a new Windows partition and a CHKDSK /R report.

I'd produce a freely available HDDScan https://hddscan.com/ data transfer and access time response graph to confirm the consistency or quirkiness of the drive, here's a sample section of such a graph and tally from a 500GB WD drive:

The drive has no S.M.A.R.T. errors and SpinRite 6.1 LEVEL 5 has no 'events', just success end to end, and the data transfer rate meets original promised specifications, CHKDSK /R shows no errors . . . but the drive drives me crazy with it's pulsing hesitation to respond to use under Windows.

There's always more to it.

So tell us more.

---

And if you're going to do anything to that drive,

1. collect GSmartControl report,
2. do an HDDScan read-to-host report,
3. do a SPINRITE 6.1 LEVEL 5,
4. do another HDDScan read-to-host report, and compare,
5. collect another GSmartControl report, and compare.

Let us know what you find.

Thanks.

 

[ShadowMeow]

Peter,

I've already learnt a lot from doing these tests. It's almost time to move onto the next drive, with more bad sectors. I only erase drives when I've finished with them, and want to blank them out. SMART has revealed all it will reveal for this drive.

Miray HD Shredder was used to write and verify random data in a way that I can control.

I don't have this level of control using SpinRite because during a level 5 you only see what the final write contains, not what's written during the process. We're told its inverted data but cannot verify or check.

The whole point of these tests is to check that SpinRite does what it says it does.

 

[peterblaise]

". . . Peter, I've already learnt a lot from doing these tests. It's almost time to move onto the next drive, with more bad sectors. I only erase drives when I've finished with them, and want to blank them out. SMART has revealed all it will reveal for this drive. Miray HD Shredder was used to write and verify random data in a way that I can control. I don't have this level of control using SpinRite because during a level 5 you only see what the final write contains, not what's written during the process. We're told its inverted data but cannot verify or check. The whole point of these tests is to check that SpinRite does what it says it does. . . ."

Correct, we each decide what we're here for, and not everyone wants to hang in there and exhaustively test SpinRite 6.1 pre-release in development ad infinitum, preferring instead to work on drives themselves to their own satisfaction, and move on.

It's all good.

Thanks..

 

[Steve]

@fzabkar : Whoever you are, your ignorance is showing. You know enough to sound authoritative, but your facts are wrong because you are assuming that you know and understand what SpinRite is doing. You don't.

 

[Steve]

I know how Seagate's SMART attributes work. You don't.

That right. An anonymous Internet troll, who appeared and created an account here an hour ago, and who has no reputation, is telling the well-known author of a hard disc utility which has saved the data of many hundreds of thousands of mass storage drives through the years, that he doesn't know how SMART attributes function. Who is more likely to be correct?

 

[peterblaise]

". . . No, [ SpinRite ] isn't [ all about data recovery and drive maintenance ]. SpinRite breaks the cardinal rule of data recovery, and that is never to recover the hard won data back to the source drive. If you can read a sector, then recover it to a known good drive. That's what RapidSpar and all other professional tools do. Drives usually retest any pending sectors prior to reallocation. If they test OK, then they are returned to service, otherwise they are reallocated. An example where this may occur is if the drive is bumped during writing, resulting in an off-track write. As for your WD drive, could you tell us the model number and interface (SATA, USB)? . . . "

Great point, yes, there are many different 'data recovery' options,

SpinRite is aimed at the self-service market of end users,​

​

not aimed at technicians doing commercial data recovery for hire.​

Recovery-in-place is historically a cascade on down from SpinRite re-interleaving drives on the fly in times when the only backup, if any, was an endless stack of diskettes, so SpinRite had to re-interleve and rewrite data in place, meaning it had to recover data in place.

Modern methods are based on the ready availability of cloning targets, opting for recovery-by-retrieval, recovery by copying out.

Different tools, different market trajectories.

---

The HDDScan graph shows ~101 MB/s to ~114 MB/s data transfer rate, implying to me SATA3:

Western Digital Caviar Blue 500GB SATA3 CMR WD5000AAKX 16 MB cache 126 MB/s 2014 Thailand​

https://www.farnell.com/datasheets/650649.pdf

​

I have two of those drives that behave the same, implying something intrinsic in the design or susceptibility to firmware invasion or corruption ( yes, I cleaned the circuit board contacts ).

Windows is happy-ish, and only a savvy user notices odd delays.

We live in a world of technical wonders and magic.

---

Still unanswered:

The opening post seems to be a comparison between SpnRite and Rapid Spar https://rapidspar.com/ with a discussion of S.M.A.R.T.,​

and then comparing SpinRite to Miray HD Shredder.​

​

Why?​

​

The reasons to compare all of those were not explicitly enumerated, and I hesitate to assume.

The reason that I mentioned:

test GRC SpinRite 6.1 pre-release under development,​

test data recovery,​

test drive maintenance,​

. . . is because those are reasons to be here at a GRC SpinRite forum.

If we're going to test SpinRite on a drive and compare drive tools, as I suggested, perhaps a routine such as this might reveal some insight:

1. collect GSmartControl report,
2. do an HDDScan sequential-read-to-host report,
3. do a SPINRITE 6.1 LEVEL 5,
4. do another HDDScan sequential-read-to-host report, and compare the results,
5. collect another GSmartControl report, and compare the results.

Otherwise, it seems to have become a 'compare drive tools' thread, where I see different marketplaces:

recovery in place, self-service end-user,​

​

recovery by retrieval, everyone, especially data recovery servicers for hire.​

What is your understanding of the original and subsequent intentions of the opening poster of, and subsequent contributors to, this thread?

Thanks.

 

[ShadowMeow]

Peter, I know what kinds of testing I need to do. I don't need a suggested test routine. Just Saying.

I'm using Miray HD Shredder just to test a theory about the drive's function.
I'm not trying to compare Miray HD Shredder with anything.

Hope that clears things up.

 

[peterblaise]

". . . Then why do this? https://www.grc.com/sr/smart.htm In that example SpinRite took an error-free Seagate drive, misinterpreted the seek and sector counts as errors, and then produced a screenful of bogus errors. Also, SpinRite incorrectly assumed that the normalised values were linear. Actually they are logarithmic. To be fair, Seagate's SMART data are often counterintuitive, but that just means that software developers should do some prior research, especially when they know that there is no standard for SMART attribute reporting . . ."

Great point . . . from -w-a-y- back, a decade ago, ". . . Last Edit: May 07, 2013 at 10:14 (3,845.04 days ago) . . ." <-- the date of that web link.

That's dated information, we've come a -l-o-n-g- way since then, especially with the incredible hands-on experience of hundreds of testors in the GRC SpinRite newsgroups and more, 50,000+ posts to start reading and learning.

And, @Steve Gibson, are you working on an update for the SpinRite pages? ;-)

But wait, 'historical documents' can be quite informative, maybe someone can produce contemporaneous Seagate documentation from 2013, or before, resources that at the time informed us all.

Anyone?

Oh yes, fzabkar archived from 2012:

​

Seagate's Seek Error Rate, Raw Read Error Rate, and Hardware ECC Recovered SMART attributes:​

https://web.archive.org/web/2023072...on.net/~fzabkar/HDD/Seagate_SER_RRER_HEC.html​

Wow, great work.

So a Seagate drive with . . . wait a minute, can the drive read, write, re-read, and transfer data at least at the peak promised by the specification documents?

So, all that calculation is actually as useless as any S.M.A.R.T. analysis outside a drive.

Because drive manufacturers have evolved and optimized S.M.A.R.T. in proprietary ways to best serve their own internal firmware and drive behavior, and they ain't volunteering their drives to be co-opted by any external program, no way.

But great work nevertheless.

Thanks.

 

[peterblaise]

". . . Peter, I know what kinds of testing I need to do. I don't need a suggested test routine. Just Saying. I'm using Miray HD Shredder just to test a theory about the drive's function. I'm not trying to compare Miray HD Shredder with anything. Hope that clears things up . . ."

Nope, not in the slightest.

As I asked, what's the purpose and function of this thread?

I thought we all were here to test SpinRite, and using drives, at least, and perhaps other software, to do so.

It seems you are testing drives and other software, not SpinRite.

OK.

Why here?

Thanks.

 

[peterblaise]

". . . If one is claiming that SpinRite, a commercial product, is a data recovery tool (it isn't), then it is only fair to compare it against genuine, commercial data recovery tools. OTOH, if "SpinRite is aimed at the self-service market of end users, not aimed at technicians doing commercial data recovery for hire", then why not compare it against a free, genuine data recovery tool such as OpenSuperClone (a fork of HDDSuperClone)? Even professional technicians use it . . ."

Which I did:

". . . Otherwise, it seems to have become a 'compare drive tools' thread, where I see different marketplaces:​

recovery in place, self-service end-user,​

recovery by retrieval, everyone, especially data recovery servicers for hire . . ."​

If you see another piece of software out there informing SpinRite 6.1 ( or future ) development, by all means, pitch in, and start a dedicated thread with whatever you think contributes the most useful and productive engagement of the SpinRite trajectory.

Thanks.

 

[ShadowMeow]

I thought we all were here to test SpinRite, and using drives, at least, and perhaps other software, to do so.

I am. Just because the discussion went in a different direction. The plan remains the same.

 

[lcoughey]

I've been reading this back and forth dialogue and couldn't sit on the sideline any more.

1. a quick google search will show that Franc (fzabkar) is more than qualified to comment
2. no matter how many times you say it, Spinrite is NOT a data recovery program and never will be.

Let's think this through. A person has a hard drive with their Quickbooks database on it and the drive is no longer booting.

Scenario 1 - Spinrite

User or technician runs Spinrite on the drive and at 98%, the drive clicks, spins down and no longer functions. How much data has been recovered?

Scenario 2 - free, open source, ddrescue or hddsuperclone

User or technician runs one of these two multipass hard drive clone programs and after getting 98% cloned, the drive clicks, spins down and no longer functions. How much data has been recovered?

Sadly, I've encountered way too many data recovery cases at my lab after scenario 1 where I have to inform the customer that there are now rings etched into the platters and recovery is not possible.

But, it gets better with Scenario 3 - RapidSpar or a DeepSpar Stabilizer with R-Stutio

Technician quickly reads the file table, targets the critical files for recovery and images them right away. Data is back into the hands of the customer with a log of which files have unreadable sectors with and those sectors can be re-read over and over and over again until the drive gives actual good data or the drive dies while trying. There is nothing worse than seeing a sector in the middle of a bunch of data "fixed" by Spinrite full of 0x00.

All that to say that the worse thing a person could do on a failing drive is to assume that it is stable and run various programs that read sectors over and over again without ever copying a single sector to another drive. The first rule of data recovery is to never write to the patient drive and the second rule is to always copy every sector you read to another drive because you may never have a second chance to read that sector again. Spinrite violates both of these rules.

 

[Steve]

I understand why a commercial data recovery company such as yours ( https://www.recoveryforce.com/ ) would wish to promote a negative impression of a comparatively inexpensive data recovery utility, having a long history of proven success, that can be purchased once and used for years. There's no way to count, nor any need to, the number of SpinRite's users who have publicly stated that SpinRite has "saved their ass" and "performed miracles" more times than they can count.

It's true that the world has changed dramatically during SpinRite's 35 years of life, and that it's now playing catch-up due to a long period of neglect. But that's now over and SpinRite will be improving and evolving quickly now that my commitment to update v6.0 has been met.

 

[lcoughey]

I understand why a commercial data recovery company such as yours ( https://www.recoveryforce.com/ ) would wish to promote a negative impression of a comparatively inexpensive data recovery utility, having a long history of proven success, that can be purchased once and used for years. There's no way to count, nor any need to, the number of SpinRite's users who have publicly stated that SpinRite has "saved their ass" and "performed miracles" more times than they can count.

It's true that the world has changed dramatically during SpinRite's 35 years of life, and that it's now playing catch-up due to a long period of neglect. But that's now over and SpinRite will be improving and evolving quickly now that my commitment to update v6.0 has been met.

Thank-you for the promotion. Spin it how you want, but until Spinrite actually follows the first 2 rules of data recovery, it is not and never will be a data recovery program.

 

[Steve]

Thank-you for the promotion. Spin it how you want, but until Spinrite actually follows the first 2 rules of data recovery, it is not and never will be a data recovery program.

You're quite welcome for the promotion. There's definitely a place for the services you offer, and GRC's technical support often refers SpinRite's users to services like yours when that's what they need. As for “it is not and never will be a data recovery program” — that's unsupportable from 35 years of success doing exactly that, and it also presumes that you know what the future holds.

 

[Steve]

I happen to know fzabkar from various other places, such as hddguru and hddoracle forums and know fzabkar know tons on hard drives, is not afraid to say "I don't know" or "you were correct" if you manage to show him he's wrong about something.

fzabkar happens to be someone on who many, professional data recovery engineers included, rely to trouble shoot board problems (HDD or SSD PCBs etc.) and firmware issues. One of the most striking things about fzabkar, I think, is that he shares knowledge and his time without asking anything in return.

It sounds like he'll make a great addition to this place, should he choose to remain.

 

[lcoughey]

SpinRite overwrote the bad sectors. If DynaStat is unable to ever recover any data from a sector, I believe it gets overwritten with all 0's.

That is the opposite of data recovery.

 

[lcoughey]

I understand why a commercial data recovery company such as yours ( https://www.recoveryforce.com/ ) would wish to promote a negative impression of a comparatively inexpensive data recovery utility, having a long history of proven success, that can be purchased once and used for years. There's no way to count, nor any need to, the number of SpinRite's users who have publicly stated that SpinRite has "saved their ass" and "performed miracles" more times than they can count.

It's true that the world has changed dramatically during SpinRite's 35 years of life, and that it's now playing catch-up due to a long period of neglect. But that's now over and SpinRite will be improving and evolving quickly now that my commitment to update v6.0 has been met.

I can't let this slide. You are implying that my motives are not pure for exposing that Spinrite is not a data recovery program while it could just as easily be said that you are hiding the fact that Spinrite is not a data recovery program because if people were to find out that they'd get better results and it would be 1000 times safer to use a free pgram like ddrescue or hddsuperclone, you would lose a lot of sales.

Please address my scenario. Which of the 2 situations actually recovered data after the drive dies at 98%? I'm also curious as to how replacing a sector's contents with 0x00 falls under the category of data recovery.

 

[Steve]

The only thing I know of your motives is from your behavior here. You come into my company's public forums and attack my commercial product, claiming — and ignoring many decades of evidence to the contrary — that SpinRite is not a data recovery program. The most generous reading of what you've written is that you're operating under a different definition of "data recovery" than I and the rest of the world. That's certainly your right, but it doesn't seem unreasonable, then, to wonder why you're here and what you're hoping to accomplish.

One thing I'm certain of is that I don't owe you an explanation of anything, least of all a detailed explanation and defense of the operation of my company's proprietary technology. You're welcome to love what you and your data recovery company does. It's unclear to me why you feel the need to attack mine.

 

[peterblaise]

". . . SpinRite overwrote the bad sectors. If DynaStat is unable to ever recover any data from a sector, I believe it gets overwritten with all 0's . . ."

". . . That is the opposite of data recovery . . ."

---

Excellent read of the situation, yes, of course, ". . . If [ SpinRite ] is unable to recover any data . . .", that definitely means no data recovery.

---

I have a pending [ suggestion ] in @Steve Gibson's development environment to have DynaStat not write anything if it can't read everything, at least make that a command line option, such as SPINRITE DYNASTAT NOWRITEONERROR.

Perhaps others also have insightful suggestions to enhance SpinRite 6.1 features and benefits, or suggestions for SpinRite 7+.

Thanks.