Seagate - ST3500320AS - SD15 - Is it BSY or somthing else?

[belshawm]

Hi Guys,

Is this the BSY issue or something else???

I have this drive:
Barracuda 7200.11
500G
ST3500320AS
Firmware - SD15

Symptoms of drive:
- Drive doesregistrars in Bios.
- Sometimes the drive seems to disappear from bios.
- Drive is visible in OS (Linux / window) for a period of time.
- With drive in computer (as secondary) windows take forever to boot
- In windows - windows reports that drive perimeters are incorrect and requires a format
- Partitions are visible but not mountable
- On attempt of drive clone after ~3gigs of data is copied the drive reports CRC errors ( I think thats the error).
- On further attempts at cloning the drive without restart cloning reports CRC errors from the start of the drive.


Do you guys this is the same SD15 firmware BSY problem? It seems to be that it could be as the firmware and model fit into this category. When entering the serial number into the Seagate serial checker it reports there is no problem with the firmware on this drive?

Thoughts, comments, suggestions are all welcome!!

Thanks for the help.

[drc]

Do you guys this is the same SD15 firmware BSY problem?

No; if you get access to the sectors then it is not BSY.

[Vulcan]

Is this the BSY issue or something else???

I agree with drc that it's something else.

I have experience of a fault which you might be describing, but I use specialist hardware in order to confirm the diagnosis. It's possible that perhaps MHDD would allow confirmation too (I can't use that on my systems so I never investigated further) but you'd have to make any necessary BIOS configuration changes, to get MHDD to recognise the drive, if it doesn't do so.

The quickest way I can think, for you to get a pointer about whether your drive might have the problem I've seen, would be to collect its SMART data - preferably in Linux (e.g. smartmontools), booting from a different drive, with the "suspect" drive attached via SATA and not being mounted during boot, and don't try to read anything from the "suspect" drive before running "smartctl -a /dev/--- >> tempfile.txt" (where "---" depends on your system, of course).

Attach that output file containing the SMART data to a reply to this thread, if you collect it, and we'll see what it shows.

You didn't mention how you had managed to clone part of the drive - what software & OS were you using? Also, why is there any uncertainty about what the error was, which stopped the cloning - it seemed that it might not have been CRCs, from what you said. Please explain.

The bad news for you is that, if I'm right in my suspicion, it will be probably be very difficult (perhaps impossible) for you to fully clone the drive successfully, without special skills/equipment.

[belshawm]

Hi Vulcan, drc,

I will get back to you with the details later today. Thanks for the help - much appreciated!

[belshawm]

Here is the SMART for the drive from MHDD
Let me know if there is anything else I should check.

RecoverSoft Media Tools was use for cloning and gave this error CloneDisk
The error message was:
"Failed to read sector 6360320"

I've ran the clone several times and it seems to always report the same location. Once the error occurs the drive seems to be inaccessible and the computer needs to be rebooted to bring it back.

Thanks for the help

Attachments

[thatdellguy]

Your drive is failing.

[fzabkar]

You need to clone your drive, sector by sector. Then use data recovery software on the clone.

In your case you need a cloning utility that understands how to work around bad sectors. Some freeware cloning tools are ...

dd_rescue: http://www.garloff.de/kurt/linux/ddrescue/
ddrescue: http://www.gnu.org/software/ddrescue/ddrescue.html
HDclone: http://www.miray.de/products/sat.hdclone.html

Comparison between ddrescue and dd_rescue:
http://www.forensicswiki.org/wiki/Dd_rescue

See the following thread for a discussion of commercial and freeware cloning/imaging utilities:

The Best Disk Cloning Hardware/Software:
the-best-disk-cloning-hardware-software-t10396.html

DDrescue is a multipass utility. It clones the easy sectors on the first pass and tries for the more difficult ones on subseqent passes. It can also clone your drive in reverse, effectively disabling look ahead caching.

As for your SMART report, attributes 5, 187 (Reported Uncorrectable Errors), 197, and 198 are indicating that your drive has a serious problem with bad sectors.

Furthermore, in addition to the BSY bug, your drive has other SMART bugs. For example, your drive has 2031 reallocated sectors, yet the normalised value of attribute #5 remains at 100. Attribute #5 should really have lost about 50 points (1 point per 40 reallocations).

See this article for an explanation of the attributes:
http://en.wikipedia.org/wiki/S.M.A.R.T.

[belshawm]

Hi fzabkar,

Very good info - much appreciated!

I'm working with ddrescue at the moment and it seems to be having serious issues on the drive. Just like the other cloning tool I was using it gets the first 3 gigs off the drive then reports a "errsize: 496GB" and "error: 1"
How does ddrescue know after ~1min of scanning that 496GB is the error size?

If I "cntr C" ddrescue and restart it I get this message

ddrescue: cannot open input file: No such device or address

Something is indeed fishy here.

[Doomer]

Furthermore, in addition to the BSY bug, your drive has other SMART bugs. For example, your drive has 2031 reallocated sectors, yet the normalised value of attribute #5 remains at 100. Attribute #5 should really have lost about 50 points (1 point per 40 reallocations).

You are wrong (again) but still making statements from assumptions
The reason why raw is not 0 but value is still 100 is because SMART has been cleared recently (topicstarter just didn't tell you that). And this is not supposed to be done by a person who don't understand internals of this procedure

As I told you before - not every answer can be found on wiki

[fzabkar]

Furthermore, in addition to the BSY bug, your drive has other SMART bugs. For example, your drive has 2031 reallocated sectors, yet the normalised value of attribute #5 remains at 100. Attribute #5 should really have lost about 50 points (1 point per 40 reallocations).

You are wrong (again) but still making statements from assumptions
The reason why raw is not 0 but value is still 100 is because SMART has been cleared recently (topicstarter just didn't tell you that).

As I told you before - not every answer can be found on wiki :mrgreen:

It certainly can't be found here, either. Not that I don't believe you, but let's ask the OP whether "SMART has been cleared recently", without making assumptions.

As for your Wiki snipe, I am loathe to play your self-indulgent game of "I know more than you". However, since you insist, I will endeavour to drag out my camera and respond appropriately in that particular thread. Hopefully I won't appear narcissistic.

[Doomer]

As for your Wiki snipe, I am loathe to play your self-indulgent game of "I know more than you". However, since you insist, I will endeavour to drag out my camera and respond appropriately in that particular thread. Hopefully I won't appear narcissistic.

It is not "I know more than you" (which is true BTW) but "Don't make statements or suggestions w/o knowing the right answer". Why you keep centering yourself ? Earth is not rotating around you
What you gonna do if the drive would kill media because of your suggestion to image it with a utility with unknown capabilities?

You don't know what's wrong with the drive besides read errors. You don't know why there are so many read errors. You don't know what was already tried. But still you are happy to make your suggestion to image it with something you have never tried yourself on severe damaged drives and don't even bother with a warning of how dangerous it can be

[belshawm]

Hi,

Nope I didn't think the SMART flags were ever reset. Although that hamster in my computer has a mind of his own someday.

[Doomer]

I guess I will take it back and apologize for my invalid statement
Judging but Att #9 SMART has not been cleared

[fzabkar]

Nope I didn't think the SMART flags were ever reset. Although that hamster in my computer has a mind of his own someday.

Sit him in front of your computer, teach him to punch a keyboard, and start your own data recovery business. :mrgreen:

[guru]

If only data recovery was a simple as you seem to believe.

I assume that a large percentage of data recovery companies do use "Hamsters/Gerbils" in their labs and I have seen one lab which use their engineers to power Salvation tools.

Attachments

hamster_1240313c.jpg (54.98 KiB) Viewed 16168 times

hamster_1240313c.jpg (54.98 KiB) Viewed 16168 times

hamster_1240313c.jpg (54.98 KiB) Viewed 16168 times

[fzabkar]

As for your Wiki snipe, I am loathe to play your self-indulgent game of "I know more than you". However, since you insist, I will endeavour to drag out my camera and respond appropriately in that particular thread. Hopefully I won't appear narcissistic.

It is not "I know more than you" (which is true BTW) ...

Careful. Some might see that statement as narcissistic.

... but "Don't make statements or suggestions w/o knowing the right answer". Why you keep centering yourself ? Earth is not rotating around you
What you gonna do if the drive would kill media because of your suggestion to image it with a utility with unknown capabilities?

You are targeting me. I'm defending, not "centering" myself.

You don't know what's wrong with the drive besides read errors. You don't know why there are so many read errors. You don't know what was already tried. But still you are happy to make your suggestion to image it with something you have never tried yourself on severe damaged drives and don't even bother with a warning of how dangerous it can be.

I have referred the OP to an old HDD Guru thread which specifically discusses the issue:
the-best-disk-cloning-hardware-software-t10396.html

If you have problems with my recommendations, then thrash it out with your colleagues.

[fzabkar]

I guess I will take it back and apologize for my invalid statement
Judging but Att #9 SMART has not been cleared

There were several invalid statements, specifically those about me. :)

As for your misunderstanding in regard to Seagate's SMART attributes and firmware issues, I'm finding this very hard to comprehend. You are the author of a popular SMART diagnostic tool (HDDScan), you have been in the data recovery profession for many years, and I believe you are an employee of Seagate or one of its affiliates. Furthermore, SD15 firmware is well known to be buggy, both within Seagate and without. Despite this, you have made an astounding blunder which calls your own professional credibility into question. :o

BTW, does that mean I know more than you? :P

[thatdellguy]

No disrespect fzabkar, but I would trust Doomers professional knowledge and expertise over yourself any day. Your nothing more than a search engine for end users who are too lazy to click a search button.

[guru]

ID 5 may suggest that the original defect limit for G-List entries is circa 2030, which would keep the drive at 100

If more defects are found and appended to the g-list you will see the value reduce

[Vulcan]

@belshawm: I see that things have moved on in this thread overnight (UK time). I don't want to get into the middle of this argument but since I asked for the SMART data which you've now kindly supplied, the least I can do is to reply - but it's not good news.

Here is the SMART for the drive from MHDD

Thanks for supplying that and, unfortunately, it shows what I have seen before on this model of drive. As you've already been told, that drive is very sick, and the behaviour you are seeing (where the drive will "disappear" from the OS after cloning reaches the same place on the disk each time) is what I would expect, based on some specific parts of that SMART data which you supplied.

However most OSs (due to HBA f/w and/or driver limitations) do a poor job of reporting exactly what is happening. Which leads on to...

RecoverSoft Media Tools was use for cloning and gave this error CloneDisk
The error message was:
"Failed to read sector 6360320"

I've ran the clone several times and it seems to always report the same location. Once the error occurs the drive seems to be inaccessible and the computer needs to be rebooted to bring it back.

By "rebooted" I expect you mean "power-cycled" - a Ctrl+Alt+Del should not be enough to make the drive accessible again, if this is the issue I've seen.

FYI, those weren't CRC errors that the drive was reporting, leading to that "Failed to read" error message by Media Tools (or at least not after the first message), but as you see, the error message doesn't give that level of detail - hence why the problem isn't easy to identify. Seeing the same location reported each time for the "start of the problem", is expected behaviour.

So at one level, from that SMART data you've now supplied and knowledge of what that drive is likely to be reporting via SATA, I understand why you're seeing the specific "drive disappearing from the OS and BIOS" behaviour. However as I said before, with the drive having reached this state, IMHO it would be very difficult, perhaps impossible (and high risk, due to how long it would take to even try) to attempt to clone it using software only.

My suggestion is that, if the data is important to you, you use the services of a DR professional - specifically one who is experienced with these Seagate 7200.11 drives (since this is not an uncommon behaviour with those drives) and who has hardware-assisted drive imaging equipment, in order to try to image your drive. If you say where you are in the world, you may get some recommendations of nearby board members who could assist.

If you choose not to take the drive to a pro, and decide to accept the risks of DIY (which have been explained elsewhere on this forum), then reverse imaging the disk might be able to clone some more data - or you might just kill the drive a little more. There are also other DIY approaches, but as I said, DIY carries a significant risk in this specific situation, seeing how much the drive has already deteriorated IMHO. The more DIY you attempt now, could significantly increase the cost (and decrease your chances of successful data recovery) if you later decide to take the drive to a pro - all depending on what is actually happening inside the disk.

It's your data, so it's your choice as to how important it is to you, and therefore what risks you want to take. Good luck!

(Just for full disclosure - I work in a different part of the electronics & data storage industry, so although I deal directly with engineers from several disk manufacturers, I'm not a DR pro. The reason I've been able to identify a little more about the behaviour you are seeing, is that I had to investigate the consequences of this specific behaviour as part of my job.)